Around 2 years ago I picked up this oriental board game called Go (English) / Igo (Japanese) Weiqi (Chinese) / Baduk (Korean). It is a board game involving Black and White pieces on a 19 x 19 board. Recently, I found out that quite a bit of the Go community is aware of the SC proscene, and some of them are lurking around on TL. I'm starting this thread both to flush out Go players on TL as well as to introduce people to this great game!
I think Go has many qualities which should attract the SC player. Just like SC, it is easy to learn but hard to master. It appeals to those who are relentlessly competitive and seek improvement in what they do. And lastly, Koreans tend to be cho-gosu and have a similarly feared reputation in both games. Its also been said that a lot of the top Korean and Chinese players are more than casual players of SC /WC3 - apparently in a tournament within the last couple months in Korea a lot of the Chinese pros went to play SC at a PC bang in their spare time.
Here are some resources to learn more about Go:
The Interactive Way to Go - Easy tutorial to learn the rules. Recommended if you are a complete beginner who wants to get into the game.
Sensei's Library A community maintained wikipedia. A great resource for learning about specific moves, history, or just common go terms.
Godiscussions Forum This used to be one of the focal points of the community. As of long, it is supplanted by the Life in 19x19 forum (listed below) as the owner no longer had time to maintain the site, which was suffering from chronic technical issues. However, its archives are still a treasure trove of information.
Life in 19x19 Forums The replacement forum for Godiscussions forums. Most of the veteran members have transfered to here.
KGS Go Server Great free server to play online go. Very newbie friendly with good community atmosphere , with lots of stronger players willing to help out. There is already a teamliquid room there, but its rather sparsely populated right now. To join go to the Rooms -> Social -> teamliquid.
If any TLers already play go or wish to learn more about the game, feel free to PM me or ask me for a game on KGS. I'm Aphelion02 on KGS and am 6k in rank, which is by no means good, but it would be nice to play with a fellow SC fan in Go.
I asked my girlfriend to get me a Go board for christmas a couple years ago, the poor thing looked everywhere to no avail. I'm very interested in a TL Go community, I'm doing the interactive right now =)
I've been meaning to learn Go for a while now. Ever since I played a few games against a friend a few years ago. Unfortunately by the time I learned about online Go I was consumed by my Starcraft addiction. I might give it a proper chance this time around.
Stopped playing serious when I reach 1d on KGS because the local go club sucks and I can't see any future with this game in my country. But sometimes I still play for fun, handa on KGS.
On June 07 2010 08:34 Nytefish wrote: I'm somewhere between 10-15k but I mostly play on 13x13. Mostly because of time, but to me it feels very different to the 19x19 game.
Have you tried playing with a shorter clock, to fit your time constraints? In fast games, the time spent is about the same as a full length starcraft game.
As you said, the formats make for very different games, and if at all possible, I'd recommend sticking with 19x19. To me, 13x13 is kind of like starcraft without the macromanagement aspect, so there's a lot you're missing out on.
On June 07 2010 20:17 Zona wrote: I've got to start playing again.
Oh - any recommended beginner books? I need to restart from the basics.
The Learn to Play Go series 1-5 by Janice Kim are pretty good for some one learning the basics. You really only need the first three books though. Graded Go problems is also a good series for practice problems.
This thread on the L19x19 forums has a very comprehensive list of good books, sorted by skill level.
I got a Go board for christmas a few years ago, but my wife got bored of it after about two games. I know a kid in high school who will play, but he's totally antisocial and will spend like two hours on each move so it's not very much fun.
I tried playing several games on KGS; got to 16k or so after learning the absolute basics and some really basic fuseki. I was impressed with how helpful the go community is compared to b.net (this was before I discovered TL).
I think I'll get back into it this summer after my finals are over and try to get to single digit kyu. What's the TL channel?
Hey, I'm a C- Terran player and a 4 kyu Go player. Actually, at a tournament today I met an old 6 Dan Korean guy who said was disappointed in the Korean youth for playing video games (SC) and not thinking enough (Go). Obviously the guy has never ventured into the Starcraft pro scene, although Go is definitely the mentally harder game.
I don't play on KGS often, but my username is "Nate".
The 2010 US Go Congress is fast approaching by the way. It's at the University of Colorado at Colorado Springs from July 31 - August 8. Go Professionals from all around the world will give lectures, review games, etc etc. More information at www.gocongress.org If you can gather the money and fly to Colorado, I'd highly suggest everybody to attend. It's a great way to get started at Go or improve your game.
Ive joined the TL go channel on KGS now. Im all up for playing stronger or weaker players so dont be afraid to PM me for a game if you want to play. I usally play on my "Rochan" or "Robinsa" account.
Have played Go for a long time but never more than a few games between friends now and then. It's an awesome game with depth like no other. I know Jinro plays it sometimes too.
I'd love to start a TL go community! I am Wren on KGS (and wren.308 on bnet) if you want to add me. I'm happy to tutor people if you are interested. I'm high kyu, close to 1d.
I'd also second the recommendation for Janice Kim's Learn to Play Go series. The "get strong at (the beginning, the endgame, tesuji, etc)" series is also great for quick puzzles that expand your knowledge rapidly.
Also, if you want a board on the cheap, I recommend making one! It is quite fun and you can decorated to suit your taste. 3/4 inch particle board, a sharpie, and a rich wood stain produce a classy looking board for under $20! Glass stones can be acquired separately at lots of game shops, asian marts, and online stores.
Anything new here? I'd love to play some more Go than I am doing right now, so if any of you would like to play, please tell! (: I used to be about 4d-5d on KGS, don't know my current rank though. Still after not playing much for the last 4 years my style hasn't gotten any better so I guess I might be 1 oder 2 stones weaker. Since I have way too many accounts on KGS, just tell me if you would like to play any of you here or on ICQ (#130123681)! (: I also don't mind tutoring a little, just want to play some more again!
PS: As for books I really would recommend "In the Beginning" by James Davies. Best book for 30kyu - 1dan, imho. Also helping out a lot when you are even stronger. (:
Anyone here still play? I'm new at this and I don't know anyone else that plays it would be great if someone could help me out. I've just been playing games here and there on KGS and getting my ass kicked
Started end of july/beginning of august and rushed my way to about 6k strength in 3 months
Unfortunately i've been taking a break recently (havnt played in about 3 weeks?) haha but i'll probably play some more again once i re-pick th einterest
Bumping this up. It's probably dead anyways, but I'm trying to learn Go and I started this week. I was wondering if anyone knew the ideal site to play vs. AI.
Additionally, I'm learning about False Eye right now but sometimes I don't understand situations where it does look like I surrounded someone, but I don't see it.
GO AI is sorta lacking, it's hard to learn off of computers. A much better option is just a patient human teacher, KGS is the most user friendly go server for learning the game. You can go to the teamliquid room or just message me or some other tl players and we can give some teaching games, if you like.
I'm using this to learn and I'm on #21. But the previous problems didn't make sense to me:
I've already figured them out, but the 'failed' situations don't look like failures to me:
For problem 1. How is F5 a false eye and why would it be catastrophic to be in this situation.
For problem 2. I don't understand how D9 is a false eye and if he placed it there, would he be capturing stones (how if he isn't surrounding me)
This is #17 in the table of contents.
So far, some ataris don't make sense, or rather; how they 'surround me' and capture stones don't make sense. Especially when it comes to long chains or captures that involve a side/wall (sorry for poor terminology).
Also, is there much difference between Korean and Japanese rules? Everyone I know plays Korean, but I only know Japanese I believe.
In both cases the stone that is forming the false eye doesn't have a solid (straight line) connection to the rest of black's stones and is thus vulnerable to capture. The only way the stone at G5 in the first problem and E9 in the second could avoid immediate capture by white is by connecting in the space where the "eye" is located (which is why they're given the name "false eyes").
Problem 1: The black stone on G5 can be captured with white playing on F5. Problem 2: When the liberty on F9 gets filled, the E9 stone will be in atari.
A key characteristic of an eye is that the opponent cannot play in there.
Afaik there are only Chinese (area) and Japanese (territory) rules. For now it's not very important. http://senseis.xmp.net/?Scoring "Which system is used will not normally affect who wins the game, and the difference between the margin of victory for the two methods will be zero or at most one point in the vast majority of games."
On May 04 2014 07:36 Roggle wrote: In both cases the stone that is forming the false eye doesn't have a solid (straight line) connection to the rest of black's stones and is thus vulnerable to capture. The only way the stone at G5 in the first problem and E9 in the second could avoid immediate capture by white is by connecting in the space where the "eye" is located (which is why they're given the name "false eyes").
So in both cases, I only lose one stone?
In the second problem, wouldn't it take two stones to capture E9?
On May 04 2014 07:37 nimbim wrote: Problem 1: The black stone on G5 can be captured with white playing on F5. Problem 2: When the liberty on F9 gets filled, the E9 stone will be in atari.
A key characteristic of an eye is that the opponent cannot play in there.
Afaik there are only Chinese (area) and Japanese (territory) rules. For now it's not very important. http://senseis.xmp.net/?Scoring "Which system is used will not normally affect who wins the game, and the difference between the margin of victory for the two methods will be zero or at most one point in the vast majority of games."
For Problem 1. it does look like a move that could have consequential effects. But for Problem 2, it doesn't seem that bad if it ends up in atari, does it?
There are no Korean rules? I wonder why he said Korean then...
In the second problem, wouldn't it take two stones to capture E9?
Nope, you lose the entire group in both cases because the remaining stones only have one eye. In the first problem, the remaining black group only has two liberties, so in two more moves white could take the entire thing. Even if black tried to force a ko fight by recapturing F5 white would eventually be able to take the group.
In the second problem it would take two more moves to capture E9, but there's nothing black can do to stop it from happening. So eventually the corner black group would also be reduced to only one eye and die.
In the second problem, wouldn't it take two stones to capture E9?
Nope, you lose the entire group in both cases because the remaining stones only have one eye. In the first problem, the remaining black group only has two liberties, so in two more moves white could take the entire thing. Even if black tried to force a ko fight by recapturing F5 white would eventually be able to take the group.
In the second problem it would take two more moves to capture E9, but there's nothing black can do to stop it from happening. So eventually the corner black group would also be reduced to only one eye and die.
For the first problem, he would go F5, then he would finish me off at A4 to capture them all? Or? This is the only part I get confused because it feels like you need a complete surrounding with whites to capture all the blacks.
I understand problem 2
Thanks!
I do need to practice some matches because after awhile; these scenarios make less sense since I didn't initiate myself into that situation.
In practice, white wouldn't actually go in and capture them because the stones are considered dead as soon as they're prevented from making a second eye. Dead groups are removed from the board at the end of the game and treated as though they were captured without having to spend the moves doing it.
Score is based on territory, but when you capture stones, don't you get to use them to reduce the number of 'blank spots' or territory of your opponent?
In the beginning it is best just to play a lot of games, that will give you a good grasp of the basic rules.
When I started playing I began with ~400 games against this 9x9 AI: http://www.heise.de/download/igowin-1120306.html However, I don't recommend playing it regularly or too often, since it is horribly bad. The best way to improve is to play the strongest opponents; this thing can only give you the earliest lessons very quickly.
You can just join the teamliquid channel on KGS (under rooms - social), everyone is friendly and you could always ask for a review or game.
On May 04 2014 08:01 Torte de Lini wrote: Is that in all versions?
Score is based on territory, but when you capture stones, don't you get to use them to reduce the number of 'blank spots' or territory of your opponent?
So wouldn't you use your extra move?
In your examples white would control the entire board, so there would be no advantage in forcing white to capture them, the score remains the same.
What I enjoy most about the game is that you play essentially to the flow of your opponent. My naivety may be showing but I played a few and she was really aggressive; so I play defensively and ended up gaining more territory over her over-eagerness to take over.
I'll install KGS tomorrow, I want to finish this interactive lesson (I'm on 23 out of 36). But I'm excited to play and meet new people!
Is it best to start 9x9 or just go for the standard board?
In the second problem, wouldn't it take two stones to capture E9?
Nope, you lose the entire group in both cases because the remaining stones only have one eye. In the first problem, the remaining black group only has two liberties, so in two more moves white could take the entire thing. Even if black tried to force a ko fight by recapturing F5 white would eventually be able to take the group.
In the second problem it would take two more moves to capture E9, but there's nothing black can do to stop it from happening. So eventually the corner black group would also be reduced to only one eye and die.
For the first problem, he would go F5, then he would finish me off at A4 to capture them all? Or? This is the only part I get confused because it feels like you need a complete surrounding with whites to capture all the blacks.
I understand problem 2
Thanks!
I do need to practice some matches because after awhile; these scenarios make less sense since I didn't initiate myself into that situation.
Did you make an account on KGS? You need java, then there is a webapp or you can download the java client. There is a teamliquid room there: Rooms->Room List->Social->teamliquid. I'm yoyoma there, several other people answering your questions are there too.
The smaller board sizes are only good for beginners and young children, because there is less to be confused about. Play 19x19 as soon as you want or when you feel confident in your understanding of the rules.
Ah man I saw some threads about the on a Go site.... I have just not updated my Java because of things like this... I think there may be a workaround in this thread: http://lifein19x19.com/forum/viewtopic.php?f=24&t=10180
Or you could try http://online-go.com/ instead. I'm KillerDucky there. The main drawback to online-go is the live player base is much smaller, although there are a good amount in the beginner range.
I used Online Go to play with my friend though, I liked it, but it also showed me moves I could and could not do; so I felt that wouldn't help me as I wouldn't be able to see when something is legal/illegal
On May 04 2014 08:32 Torte de Lini wrote: I just got into KGS!
I used Online Go to play with my friend though, I liked it, but it also showed me moves I could and could not do; so I felt that wouldn't help me as I wouldn't be able to see when something is legal/illegal
The only illegal moves are when you put a stone somewhere it would be captured immediately without the opponent doing anything (like in an eye), and doesn't capture a group itself. Or if you put the board in the exact same position as last move (Ko). You should be able to see those. The rest are just useless stones which suicide, it's dumb but not illegal (think a stone in a 2*1 eye)
What is really important and difficult to do is find where are the weaknesses in your stones, and the opponent's. Then you can take the initiative and ignore a small attack to press the issue somewhere vital. Gauging the importance of areas and potential moves is very hard though. Most of the times you'll get attacked and only then realise you're in DEEEEP shit. :-p
I'm too competitive to actually play Go against humans... I hate to be bad at the game and make glaring mistakes, so I'll wait a few more years to play online... (Already 3/4 years lol though, I should be around 20kyu maybe ?) Just learning joseki, life and death and yose with several books and apps. (yeah, I know, it's the only way to truly progress, but I can't involve the necessary time, same problem with WC3 and SC2 and DOTA2, that's why I don't play them.)
I'd advise one book which I found extremely interesting : http://senseis.xmp.net/?MakingGoodShape http://www.bengozen.com/book-review-making-good-shape/ Gives a lot of good advices on the strengths and weaknesses of many shapes of stones, helping you find the right move in a lot of situations, even for beginners. It might not be perfect since it was written by amateurs, but it's much easier to comprehend than most. It's hard to get in the mind of a top-pro :-p (exemple, this one : http://www.bengozen.com/book-review-basic-techniques-go/ ) It deals with the way to make your moves useful and with a purpose, answers the question "Why should I play that move ?", instead of just telling you "play that and you're alive".
For a start though, something like that : http://www.bengozen.com/book-review-graded-go-problems-for-beginners-vol-1/ is perfect, showing you usual moves, shapes and obvious errors, in a wide range of situations. To familiarize yourself with things you will be encountering and knowing how to deal with them. (and vol 2/3/4...)
I can scan a few sample pages of these books if anyone is interested, to get an idea if you're interested in buying them (not more though, copyright etc)
Is there a way to upload a replay? I played someone today that went very poorly. It was my first match but I got frustrated with how anxious and narrow-minded I got trying to get territories.
Just giving it a first look, my guess is you lost the match in thr first 10 moves. You played two moves for each corner, giving your opponent the other two AND the sides.
The way a game is played is by first taking each corner (worth more since you only need to enclose them on two sides, less stones needed and easier to defend), then advancing on the sides of the board -need 3 sides to enclose-, then the battle starts spreading around the center.
Your opening gave him a huge headstart on the sides, and while yours was solid ie. he couldnt attack, he didn't need to since he had way more potential territory right from the beginning.
Tldr: play on two corners first.
Rest later
Edit : move 71 NEEDS to be on H6. Try to understand why.
On May 05 2014 17:43 Nouar wrote: Just giving it a first look, my guess is you lost the match in thr first 10 moves. You played two moves for each corner, giving your opponent the other two AND the sides. .
so much nonsense... Yes, it certainly may not be the most efficient opening ever created... But it isn't straight up losing. You can roll a dice for your first 5 moves and still play a decent game afterwards... at least as long as you aren't pro, and even then People like Gan Siyang or Ohashi Hirofumi prove us day by day, that almost everything is playable, if you know what you are doing. And at tortes level games are certainly not decided by this.
(btw, I'm as MaLa 3d in tl-room, whisper me, if you like)
On May 05 2014 17:43 Nouar wrote: Just giving it a first look, my guess is you lost the match in thr first 10 moves. You played two moves for each corner, giving your opponent the other two AND the sides. .
so much nonsense... Yes, it certainly may not be the most efficient opening ever created... But it isn't straight up losing. You can roll a dice for your first 5 moves and still play a decent game afterwards... at least as long as you aren't pro, and even then People like Gan Siyang or Ohashi Hirofumi prove us day by day, that almost everything is playable, if you know what you are doing. And at tortes level games are certainly not decided by this.
(btw, I'm as MaLa 3d in tl-room, whisper me, if you like)
e: of course i totally agree about h6 ^^
Yeah of course, but I just took a look at the first moves and the ending, and well, the territories at the end (apart from the black group at the top) were mostly defined by these first moves. 1/3 for black on the lower side, and 2/3 top for white. White is perfectly happy with creating walls around those 2/3 in the next 20/30 moves, before the move to the center. Then again, I'm a newbie myself, and nearly only studied theory ;-) I just wanted to explain what were the easiest areas of the board to take, which gave access to the most potential territory easily, and explain why 90% of games evolve from corners > sides > center.
When you're bad, you're not good at attacking, so giving your opponent 2/3 of the board right away isn't the best start... A good player might be able to come back, but if the opponent doesn't make obvious mistakes (I have yet to see what happened top), it's still hard to come back imo. Obviously being a 3d, your advice is more useful :-p
Around move 180 : you need to realise your whole group in the topleft needs two eyes to survive. If this group survives, you completely destroy white territory in the top left. You have one already in F13, but you need another one. At any moment around that move, instead of attacking F18 etc, if you just play around C18 or B18 to secure the corner with some space, you create another eye, your group is alive, and you probably win.
H6 on turn 62-64? I see it, and it is obviously to connect the two, but wouldn't he have cut it anyways? I feel it was cut one way or the other and H6 would have given him both options anyways ):
I'm not yet able to see when to make two eyes or how to yet. Is it possible to make two eyes even if there is an opposing stone in the mix?
On May 05 2014 19:27 Torte de Lini wrote: H6 on turn 62-64? I see it, and it is obviously to connect the two, but wouldn't he have cut it anyways? I feel it was cut one way or the other and H6 would have given him both options anyways ):
I'm not yet able to see when to make two eyes or how to yet. Is it possible to make two eyes even if there is an opposing stone in the mix?
No, on move 71. It would have cut white in two, making you able to (maybe) isolate and capture his 7 stones in the middle. If he plays there, he connects with his stones on the left and isolates all your stones on the top-left (G10 group), leaving them in a very bad spot. In short, in one move, you go from solidly invading his territory and working the center, to losing the center, and struggling to keep a big group of stones alive (which you ultimately lost). That's how important connecting your groups are (well, and thinking in terms of groups of stone) If you can connect some of your groups, you only need two eyes for the whole. You need to see at any given point, if a group can be connected to others, or if it's alone. In that case, it needs two eyes for itself.
- The top left stones are connected in diagonal because if black plays in one opening, you can play in the other and effectively wire them. - On the top right, they are not connected. If it's your turn, you can connect, if it's not, black can cut you. (That's why it's important to have the initiative. If you play somewhere not really important, a move that doesn't threaten the opponent, he can ignore your move and go play where it hurts) - In the bottom left (shitty example -_-), the white stones are strongly connected, in a line. - Bottom right, white is in fact connected, because if black tries to attack P1 or R1, his stones would already be in atari, you can capture his attacking stone on the next move. It is easier to connect stones near the edge of the board.
- In the center, looking at the previous examples, where would you play to secure white's position in one move ?
On May 05 2014 19:27 Torte de Lini wrote: H6 on turn 62-64? I see it, and it is obviously to connect the two, but wouldn't he have cut it anyways? I feel it was cut one way or the other and H6 would have given him both options anyways ):
I'm not yet able to see when to make two eyes or how to yet. Is it possible to make two eyes even if there is an opposing stone in the mix?
No, on move 71. It would have cut white in two, making you able to (maybe) isolate and capture his 7 stones in the middle. If he plays there, he connects with his stones on the left and isolates all your stones on the top-left (G10 group), leaving them in a very bad spot. In short, in one move, you go from solidly invading his territory and working the center, to losing the center, and struggling to keep a big group of stones alive (which you ultimately lost). That's how important connecting your groups are (well, and thinking in terms of groups of stone) If you can connect some of your groups, you only need two eyes for the whole. You need to see at any given point, if a group can be connected to others, or if it's alone. In that case, it needs two eyes for itself.
- The top left stones are connected in diagonal because if black plays in one opening, you can play in the other and effectively wire them. - On the top right, they are not connected. If it's your turn, you can connect, if it's not, black can cut you. (That's why it's important to have the initiative. If you play somewhere not really important, a move that doesn't threaten the opponent, he can ignore your move and go play where it hurts) - In the bottom left (shitty example -_-), the white stones are strongly connected, in a line. - Bottom right, white is in fact connected, because if black tries to attack P1 or R1, his stones would already be in atari, you can capture his attacking stone on the next move. It is easier to connect stones near the edge of the board.
- In the center, looking at the previous examples, where would you play to secure white's position in one move ?
Ah, I didn't even see it! I was too focused on making sure I had exits!
I understand, in context, it makes sense; but when I'm playing I lose sight of it because of everything that is happening!
K12? Why? It's too far to be a connection. What happens if I play j12 (then either h12 or j11), or h12 (either capture with j11 or extend with h13 on the next move depending on your answer). No, the correct shape is playing in j12 yourself, effectively protecting both holes. If the opponent tries to play in one of them, he is captured in answer. This shape, you will se often.
On May 06 2014 02:02 Torte de Lini wrote: It was automatch, but it gave him two stones for free T_T
I will chekc our Nouar's comment after laundry and KillerDucky :D
You can reduce the number of sandbaggers (stronger players trolling/smurfing lower ranks) by turning off "Unranked Opponent OK". This prevents you from getting matched with guests, who are much more likely to be sandbagging.
On May 06 2014 02:02 Torte de Lini wrote: It was automatch, but it gave him two stones for free T_T
I will chekc our Nouar's comment after laundry and KillerDucky :D
You can reduce the number of sandbaggers (stronger players trolling/smurfing lower ranks) by turning off "Unranked Opponent OK". This prevents you from getting matched with guests, who are much more likely to be sandbagging.
On May 05 2014 21:35 Nouar wrote: K12? Why? It's too far to be a connection. What happens if I play j12 (then either h12 or j11), or h12 (either capture with j11 or extend with h13 on the next move depending on your answer). No, the correct shape is playing in j12 yourself, effectively protecting both holes. If the opponent tries to play in one of them, he is captured in answer. This shape, you will se often.
Oops, I meant J12 as you can block both areas with double atami no?
On May 06 2014 01:15 KillerDucky wrote: I added just a few comments to your last game: http://eidogo.com/#7jzBCJoW
What servers do you guys play on? I used to play kgs a lot, recently wbaduk and then a polish friend suggested tygemgo (which actually looks to be suspiciously similar to wbaduk room wise)
Torte i have many accounts for various levels of shitfacedness. I lose against 15k koreans playing blitz when hammered (tbh that game i should never of lost - i usually am about 8k in those games) yet beat 1d in slow games in pubs in uk (on good days)
omg one game of go and you get trolled. He was stronger by quite a bit, v hard to tell how much i guess > 9 stones. you are right he was rude, game is over when both pass. When i get into games where i am clearly stronger, its over and i want it to end i just resign and review the game.
Key thing about go: Its all about the 4th lines to the edge the bits further into the middle are important but in terms of getting better fast (ie to 20k) forget it.
Play a ton of 9x9 games. Igowin is a decent program - the key is play them fast. You learn more from 'oh yeah' and analysing later than reading. Reading is hugely important but you need to train yourself to see the good stuff from the noise of the rest of the moves. That comes from doing problems.
In fact i wrote tons deleted it. Just play 100 9x9 games asap. Get it over with. Will help you a ton. Treat it like speed minesweeper.
That will teach you lots about the 3rd line towards the edge - connecting groups and maybe some life and death. When you play 19x19 just play 4-4 points. The 3-4 point joseki get insanely complex and you really don't need that. I got to 3k never playing them. Learn some simple 3-4 move joseki - they will still be good when you are 9p and you will see them in nearly ever pro games played. Then id say the real game begins
On May 05 2014 21:35 Nouar wrote: K12? Why? It's too far to be a connection. What happens if I play j12 (then either h12 or j11), or h12 (either capture with j11 or extend with h13 on the next move depending on your answer). No, the correct shape is playing in j12 yourself, effectively protecting both holes. If the opponent tries to play in one of them, he is captured in answer. This shape, you will se often.
Oops, I meant J12 as you can block both areas with double atami no?
On May 06 2014 01:15 KillerDucky wrote: I added just a few comments to your last game: http://eidogo.com/#7jzBCJoW
How do I see yoru comments O:
Yes about j12.
Lower right, click on move numbers around where there are variations (line branching out) and you'll see his comments and suggestions.
I agree, for the second game, at what moment did you feel you were in charge ? 0_o Going through it fast, it seems he always has your number and is countering you at every corner, he feels much stronger, I don't get why he got handicap stones. Don't be sad about that game, you didn't stand a chance.
On May 06 2014 03:53 MrTortoise wrote: cool go players
shall we start a team liquid room on kgs?
What servers do you guys play on? I used to play kgs a lot, recently wbaduk and then a polish friend suggested tygemgo (which actually looks to be suspiciously similar to wbaduk room wise)
Torte i have many accounts for various levels of shitfacedness. I lose against 15k koreans playing blitz when hammered (tbh that game i should never of lost - i usually am about 8k in those games) yet beat 1d in slow games in pubs in uk (on good days)
There's already a room on KGS! Look under Social :DDDD
Torte: Scroll through the game, I put comments and variations. When there is a variation it will show #s on the board, and you can see them in the game tree in the lower right. Last comment is move #106 btw so you don't have to hunt for more after that.
MrTortoise: There is already a teamliquid room on KGS. See the OP.
Btw, when he says "slow" it means the move doesn't accomplish a lot in the grand scheme of things, and doesn't force your opponent to react, thus losing initiative. I won't join you on kgs these days, sorry, got too little time with work and evenings are busy :s
After move 72, it's basically useless playing on your group in the lower right. It's completely cut off and has no space to survive. There's a huge group just atop of it, and at the lower left, and on the right, the opponent's stones are far too close for you to be able to survive in that small space. He can close the gaps very easily.
A chain of stones alone in the middle of the board is useless (or with chance can create aji -potential threats, weaknesses in the opponent's shape-), they must have a root, be linked to another group, preferrably on the sides, or you have to work in creating two eyes for it to survive. If it just stays there alone, it will die. And dead stones are worth 2 points each for your opponent : 1 for the territory, 1 for the stone. See comment at move #100 and #106
Your left group never had a chance after move 128. It cuts it off from the top, and the opponent carefully played before that on line 8 so that you don't have the room to create eyes on the bottom.
On May 06 2014 03:53 MrTortoise wrote: cool go players
shall we start a team liquid room on kgs?
What servers do you guys play on? I used to play kgs a lot, recently wbaduk and then a polish friend suggested tygemgo (which actually looks to be suspiciously similar to wbaduk room wise)
Torte i have many accounts for various levels of shitfacedness. I lose against 15k koreans playing blitz when hammered (tbh that game i should never of lost - i usually am about 8k in those games) yet beat 1d in slow games in pubs in uk (on good days)
There's already a room on KGS! Look under Social :DDDD
I play very slowly ):
I dont go on that much, but when i am ill give you a game or 2
On May 05 2014 21:35 Nouar wrote: K12? Why? It's too far to be a connection. What happens if I play j12 (then either h12 or j11), or h12 (either capture with j11 or extend with h13 on the next move depending on your answer). No, the correct shape is playing in j12 yourself, effectively protecting both holes. If the opponent tries to play in one of them, he is captured in answer. This shape, you will se often.
Oops, I meant J12 as you can block both areas with double atami no?
On May 06 2014 01:15 KillerDucky wrote: I added just a few comments to your last game: http://eidogo.com/#7jzBCJoW
How do I see yoru comments O:
Yes about j12.
Lower right, click on move numbers around where there are variations (line branching out) and you'll see his comments and suggestions.
I agree, for the second game, at what moment did you feel you were in charge ? 0_o Going through it fast, it seems he always has your number and is countering you at every corner, he feels much stronger, I don't get why he got handicap stones. Don't be sad about that game, you didn't stand a chance.
That is so fucking cooooooooool!
Turn 37 I felt had a good handle on things and then I ventured to his corner. What also tricks me up is that sometimes I feel like he's placing random stones and I can't understnad what they are for until it is way too late!
Have a look at the board. You have about 12 points of secure territory on the upper side, another group at d12 which does not secure any points but influences the left side and the center. Realistically you cannot expect to profit much from the d12 group because it is missing a stone at b10 which makes it hard to build territory on the left side and black already encroached into your sphere of influence in the center with f15. Note that there is no way for you to cut off f15. Black has about 7 points of territory at the left side and two captured stones, 9 points in total. However he completely dominates the right side and the bottom side of the board. None of this positions are already secure territory bu he will be able to make far more points in these areas than you. Meanwhile you aspirations in the center will come to nothing due to f15.
As mentioned above your opponent seems to be far stronger than you, probably a sandbagger.
On May 06 2014 05:27 Hagen0 wrote: At turn 37 the game was already hopeless for you.
Have a look at the board. You have about 12 points of secure territory on the upper side, another group at d12 which does not secure any points but influences the left side and the center. Realistically you cannot expect to profit much from the d12 group because it is missing a stone at b10 which makes it hard to build territory on the left side and black already encroached into your sphere of influence in the center with f15. Note that there is no way for you to cut off f15. Black has about 7 points of territory at the left side and two captured stones, 9 points in total. However he completely dominates the right side and the bottom side of the board. None of this positions are already secure territory bu he will be able to make far more points in these areas than you. Meanwhile you aspirations in the center will come to nothing due to f15.
As mentioned above your opponent seems to be far stronger than you, probably a sandbagger.
I think the biggest problem I am having is understanding what is my territory, the balance between ensuring I am securing territory over capturing stones.
I feel like I am flipping between the two. On a smaller scale, it is very obvious, but on a larger scale I get confused.
Looking at your games, I think, you look too far in your search for mistakes Let's say, there is a scale of how good a move is. Pass is 0, a move perfectly achieving a single purpose is worth 1
There are not many moves worse then pass(so they have a negative value), but some are... And they can also be found in your games a few times. They are pretty easy to avoid, once you know about them, and it immediately makes you a fair bit stronger... Then the next step is avoiding all those moves with a value around 0. This can be easy in some cases, but difficult in others... still.. Again with some explanations it is easy to reduce the ratio drastically.
And if this done... Then the real game starts... Because then it is all about finding the best moves for a certain purpose, finding the most important purposes, and later find moves that actually help with several purposes But first master the basics, before diving too deep here
On May 06 2014 23:19 mahrgell wrote: Looking at your games, I think, you look too far in your search for mistakes Let's say, there is a scale of how good a move is. Pass is 0, a move perfectly achieving a single purpose is worth 1
There are not many moves worse then pass(so they have a negative value), but some are... And they can also be found in your games a few times. They are pretty easy to avoid, once you know about them, and it immediately makes you a fair bit stronger... Then the next step is avoiding all those moves with a value around 0. This can be easy in some cases, but difficult in others... still.. Again with some explanations it is easy to reduce the ratio drastically.
And if this done... Then the real game starts... Because then it is all about finding the best moves for a certain purpose, finding the most important purposes, and later find moves that actually help with several purposes But first master the basics, before diving too deep here
Mm, you're right. Maybe I am a bit too critical. I don't see problems as gradual steps; but key areas that need "fixing" checkmarks of reminders. I should probably change it to more of a scaling point.
I just won my first game, but I had help from my boss: Mr. Lee. He played amateur GO when he was younger and he says he was about a 1 kyu (geup) online and 6 kyu in live matches.
He says his father is a professional Go teacher also, he was a 5dan.
Here's the replay. Can you guess at which move he started to teach me?
On June 09 2014 05:08 ToSs.Bag wrote: Torte de lini, do you still play? I picked up this game right about the same time you did. I would love to play you somehow. Cheers!
hey!
Yes, I still play albeit not recently due to more pressing issues! I intend to get back into the swing of things this week :D
Wow, this thread is randomly active. I still drop in on KGS every now and again to observe games, but don't play enough of it. Still probably somewhere around 10k-1k
I love OGS. Browser based means I can play on any computer from anywhere, and the player base is great for beginner-intermediate range. When I first started playing, I started on KGS and the skill level was too prohibitive and the interface is awkward, in my opinion.
Hey thanks for the shout on online-go.com. I've been meaning to learn this game for a while now and that site has some pretty good resources.
Also, only played one human so far and got wrecked pretty hard, but they were really friendly and played another match with me and talked me through the game step-by-step telling me the basics.
On October 22 2014 21:33 jtype wrote: Hey thanks for the shout on online-go.com. I've been meaning to learn this game for a while now and that site has some pretty good resources.
Also, only played one human so far and got wrecked pretty hard, but they were really friendly and played another match with me and talked me through the game step-by-step telling me the basics.
yay!
Which board did you play? 9x9 or something larger? I've been playing a bunch recently and I win about 50% of the matches! I used to play 19x19 boards when I first started, but 9x9 is actually a bit more comfortable as you learn to see scenarios (which you catch on after 20-30 games I would say).
There's tons of good TL Go players here, they're very nice and mature; it's awesome!
Just played a match!
I felt Black was rude by not passing and trying to minimize the difference of territory even though I would have clearly captured his black stones (around ~J6), but maybe it's not rude; just part of the game.
But having said that, I say 'playing', when really I've played one real game vs a human opponent and the rest of the time I've been reading and doing the puzzle challenges.
Hehe, one of the things that stood out to me during reading some of the strategy resources was that it is considered rude (or dishonourable) to extend a game out when you've clearly lost.
On October 22 2014 21:39 jtype wrote: Yea I've only been playing on 9x9 boards.
But having said that, I say 'playing', when really I've played one real game vs a human opponent and the rest of the time I've been reading and doing the puzzle challenges.
Hehe, one of the things that stood out to me during reading some of the strategy resources was that it is considered rude (or dishonourable) to extend a game out when you've clearly lost.
9x9 is definitely a nice start, but others may disagree as it's maybe playing with a bitch of crutch (shorter games too )
The puzzle challenges are great, The Eye concept is my favourite thing to try and pull off. But I'm very new as well. What I found in playing is that it is better to play straight-forward and focus on capturing territories/sectioning off parts of the board over capturing stones or trying to reduce his ability.
Also knowing when stones are lost helps a lot!
edit: OH ALSO, the game auto-scores your points, so that's nice, but be sure to properly section off all areas that are yours but you never finished because it only takes one move to clean up a territory (the calculations don't know that, so do it before you end the game so it auto-scores that area clearly yours. You can see it in the screenshot above that there's a section he forgot to claim at the top-left that could have helped him).
Yeah, I just spoke to some mods and they told me it was pretty rude to keep playing in a territory very clearly captured. But we're all a bit new, so maybe he thought he could (though to be honest, there is almost nothing to do there).
Nonetheless, still won and I told him he was rude!
I created a group called Tortellini (lol), but maybe we should start a Team Liquid group!
Continuing to play when you have clearly lost is not really rude at lower ranks. Many players simply won't know what exactly is going on, so they keep on playing but mean no harm. I understand how it can be annoying, but I know I've continued plenty of games that were lost because I didn't know it's over. Also, it can be helpful to continue playing the game to the end to practice endgame or whatever, so some people don't resign even though they know they've lost. That's perfectly fine in my opinion, as long as it's not a 60 point lead and they keep playing honest moves.
Remember, it's supposed to be an honorable game. You thank your opponent for the game regardless of the outcome and you don't get mad at people for playing moves that appear weak to you. http://senseis.xmp.net/?Etiquette
[...] you should not wait to resign until the game is about to end. But one should not resign too early either. When suffering a big loss, it is wise not to resign immediately. Continue for several moves until your anger or disappointment has cooled down, and you are able to make a positional judgment. Only if this judgment shows that the game is really lost, is resignation appropriate.
Apparently, professional players apply the latter issue with utmost sincerity. We should not be so hard on each other as demanding our opponent to master the professional's art of resigning.
On October 23 2014 02:28 nimbim wrote: Continuing to play when you have clearly lost is not really rude at lower ranks. Many players simply won't know what exactly is going on, so they keep on playing but mean no harm. I understand how it can be annoying, but I know I've continued plenty of games that were lost because I didn't know it's over. Also, it can be helpful to continue playing the game to the end to practice endgame or whatever, so some people don't resign even though they know they've lost. That's perfectly fine in my opinion, as long as it's not a 60 point lead and they keep playing honest moves.
Remember, it's supposed to be an honorable game. You thank your opponent for the game regardless of the outcome and you don't get mad at people for playing moves that appear weak to you. http://senseis.xmp.net/?Etiquette
[...] you should not wait to resign until the game is about to end. But one should not resign too early either. When suffering a big loss, it is wise not to resign immediately. Continue for several moves until your anger or disappointment has cooled down, and you are able to make a positional judgment. Only if this judgment shows that the game is really lost, is resignation appropriate.
Apparently, professional players apply the latter issue with utmost sincerity. We should not be so hard on each other as demanding our opponent to master the professional's art of resigning.
The biggest appeal about Go to me is the feeling of honor and etiquette. Something about it really makes me attracted to the game.
I also was a bit annoyed by him continuing to play because I felt he could have actually captured my entire territory when I passed seeing as I had no longer any moves. So essentially, I was done because putting stones in my territory would have netted me less points; so I passed. He didn't and took advantage of my passed turn to create an initiative that was weak and could not be won.
On October 23 2014 02:28 nimbim wrote: Continuing to play when you have clearly lost is not really rude at lower ranks. Many players simply won't know what exactly is going on, so they keep on playing but mean no harm. I understand how it can be annoying, but I know I've continued plenty of games that were lost because I didn't know it's over. Also, it can be helpful to continue playing the game to the end to practice endgame or whatever, so some people don't resign even though they know they've lost. That's perfectly fine in my opinion, as long as it's not a 60 point lead and they keep playing honest moves.
Remember, it's supposed to be an honorable game. You thank your opponent for the game regardless of the outcome and you don't get mad at people for playing moves that appear weak to you. http://senseis.xmp.net/?Etiquette
[...] you should not wait to resign until the game is about to end. But one should not resign too early either. When suffering a big loss, it is wise not to resign immediately. Continue for several moves until your anger or disappointment has cooled down, and you are able to make a positional judgment. Only if this judgment shows that the game is really lost, is resignation appropriate.
Apparently, professional players apply the latter issue with utmost sincerity. We should not be so hard on each other as demanding our opponent to master the professional's art of resigning.
The biggest appeal about Go to me is the feeling of honor and etiquette. Something about it really makes me attracted to the game.
I also was a bit annoyed by him continuing to play because I felt he could have actually captured my entire territory when I passed seeing as I had no longer any moves. So essentially, I was done because putting stones in my territory would have netted me less points; so I passed. He didn't and took advantage of my passed turn to create an initiative that was weak and could not be won.
There is still something you could have played. If White plays D9 first, the area on top will be worth 3 points for Black, but if Black goes first it's 4 points. In the end both players neglected to play there, so it's at least 3 points lost for Black.
On October 23 2014 02:28 nimbim wrote: Continuing to play when you have clearly lost is not really rude at lower ranks. Many players simply won't know what exactly is going on, so they keep on playing but mean no harm. I understand how it can be annoying, but I know I've continued plenty of games that were lost because I didn't know it's over. Also, it can be helpful to continue playing the game to the end to practice endgame or whatever, so some people don't resign even though they know they've lost. That's perfectly fine in my opinion, as long as it's not a 60 point lead and they keep playing honest moves.
Remember, it's supposed to be an honorable game. You thank your opponent for the game regardless of the outcome and you don't get mad at people for playing moves that appear weak to you. http://senseis.xmp.net/?Etiquette
[...] you should not wait to resign until the game is about to end. But one should not resign too early either. When suffering a big loss, it is wise not to resign immediately. Continue for several moves until your anger or disappointment has cooled down, and you are able to make a positional judgment. Only if this judgment shows that the game is really lost, is resignation appropriate.
Apparently, professional players apply the latter issue with utmost sincerity. We should not be so hard on each other as demanding our opponent to master the professional's art of resigning.
The biggest appeal about Go to me is the feeling of honor and etiquette. Something about it really makes me attracted to the game.
I also was a bit annoyed by him continuing to play because I felt he could have actually captured my entire territory when I passed seeing as I had no longer any moves. So essentially, I was done because putting stones in my territory would have netted me less points; so I passed. He didn't and took advantage of my passed turn to create an initiative that was weak and could not be won.
There is still something you could have played. If White plays D9 first, the area on top will be worth 3 points for Black, but if Black goes first it's 4 points. In the end both players neglected to play there, so it's at least 3 points lost for Black.
So if I'm ahead and I clearly don't have any more purpose to play, what do I do if I know he won't pass? If I initiate a pass even though I'm winning he may use it to his advantage and I'm stuck losing points by his own short-sightedness.
On October 23 2014 02:28 nimbim wrote: Continuing to play when you have clearly lost is not really rude at lower ranks. Many players simply won't know what exactly is going on, so they keep on playing but mean no harm. I understand how it can be annoying, but I know I've continued plenty of games that were lost because I didn't know it's over. Also, it can be helpful to continue playing the game to the end to practice endgame or whatever, so some people don't resign even though they know they've lost. That's perfectly fine in my opinion, as long as it's not a 60 point lead and they keep playing honest moves.
Remember, it's supposed to be an honorable game. You thank your opponent for the game regardless of the outcome and you don't get mad at people for playing moves that appear weak to you. http://senseis.xmp.net/?Etiquette
[...] you should not wait to resign until the game is about to end. But one should not resign too early either. When suffering a big loss, it is wise not to resign immediately. Continue for several moves until your anger or disappointment has cooled down, and you are able to make a positional judgment. Only if this judgment shows that the game is really lost, is resignation appropriate.
Apparently, professional players apply the latter issue with utmost sincerity. We should not be so hard on each other as demanding our opponent to master the professional's art of resigning.
The biggest appeal about Go to me is the feeling of honor and etiquette. Something about it really makes me attracted to the game.
I also was a bit annoyed by him continuing to play because I felt he could have actually captured my entire territory when I passed seeing as I had no longer any moves. So essentially, I was done because putting stones in my territory would have netted me less points; so I passed. He didn't and took advantage of my passed turn to create an initiative that was weak and could not be won.
There is still something you could have played. If White plays D9 first, the area on top will be worth 3 points for Black, but if Black goes first it's 4 points. In the end both players neglected to play there, so it's at least 3 points lost for Black.
So if I'm ahead and I clearly don't have any more purpose to play, what do I do if I know he won't pass? If I initiate a pass even though I'm winning he may use it to his advantage and I'm stuck losing points by his own short-sightedness.
It sounds like your describing an endgame situation where some of your shapes are not technically alive (though probably de facto alive). In such situations, some opponents may try to expose weaknesses in your shapes, hoping that you won't protect them. If your opponent makes such a move - they are throwing away 1 point to do so. This allows you to respond to the threat without a net loss in points. If somehow you end up losing points in the exchange, then your shape was missing a move to begin with.
You and your opponent are both beginners. Beginners don't have enough experience to know what sort of invasions can or cannot work. He just tried a few moves in there to see what happens, not a big deal IMO. Honestly look at yourself -- you passed and told your opponent to pass, yet the game was not finished. At the time the B2 area was not resolved, and D9 was never played.
I don't understand what you're saying about being stuck losing points. If he invades, that is giving you 1 point. If you answer, you lose 1 point. Thats a net 0 difference so the score does not change if you just answer every invasion stone. If he passes, usually that means he is admitting the invasion was unsuccessful, so you pass too.
If he passes but then tries to claim some clearly doomed stones are alive, then you have to convince him (verbally/chat) that they're dead. If you can't reach an agreement, call an admin.
On October 23 2014 02:28 nimbim wrote: Continuing to play when you have clearly lost is not really rude at lower ranks. Many players simply won't know what exactly is going on, so they keep on playing but mean no harm. I understand how it can be annoying, but I know I've continued plenty of games that were lost because I didn't know it's over. Also, it can be helpful to continue playing the game to the end to practice endgame or whatever, so some people don't resign even though they know they've lost. That's perfectly fine in my opinion, as long as it's not a 60 point lead and they keep playing honest moves.
Remember, it's supposed to be an honorable game. You thank your opponent for the game regardless of the outcome and you don't get mad at people for playing moves that appear weak to you. http://senseis.xmp.net/?Etiquette
[...] you should not wait to resign until the game is about to end. But one should not resign too early either. When suffering a big loss, it is wise not to resign immediately. Continue for several moves until your anger or disappointment has cooled down, and you are able to make a positional judgment. Only if this judgment shows that the game is really lost, is resignation appropriate.
Apparently, professional players apply the latter issue with utmost sincerity. We should not be so hard on each other as demanding our opponent to master the professional's art of resigning.
The biggest appeal about Go to me is the feeling of honor and etiquette. Something about it really makes me attracted to the game.
I also was a bit annoyed by him continuing to play because I felt he could have actually captured my entire territory when I passed seeing as I had no longer any moves. So essentially, I was done because putting stones in my territory would have netted me less points; so I passed. He didn't and took advantage of my passed turn to create an initiative that was weak and could not be won.
There is still something you could have played. If White plays D9 first, the area on top will be worth 3 points for Black, but if Black goes first it's 4 points. In the end both players neglected to play there, so it's at least 3 points lost for Black.
So if I'm ahead and I clearly don't have any more purpose to play, what do I do if I know he won't pass? If I initiate a pass even though I'm winning he may use it to his advantage and I'm stuck losing points by his own short-sightedness.
It sounds like your describing an endgame situation where some of your shapes are not technically alive (though probably de facto alive). In such situations, some opponents may try to expose weaknesses in your shapes, hoping that you won't protect them. If your opponent makes such a move - they are throwing away 1 point to do so. This allows you to respond to the threat without a net loss in points. If somehow you end up losing points in the exchange, then your shape was missing a move to begin with.
How does he throw away a point? I feel like I'm forgetting a basic rule that states it but please remind me!
I find myself losing a lot to territory differences and not so much captured points. It really sucks when I think I got a huge chunk of the board but then I can't seem to deny or cripple his own side.
He contested it a bit at the bottom left but I mostly got that huge chunk to me.
But after that, I tried to play around the H-8 area and it fell apart and I lost because I can't materialize much defense (I was trying to create an eye) to deteriorate his % of territory.
He contested it a bit at the bottom left but I mostly got that huge chunk to me.
But after that, I tried to play around the H-8 area and it fell apart and I lost because I can't materialize much defense (I was trying to create an eye) to deteriorate his % of territory.
Do remember that a group with only one eye is still dead, unless you can somehow connect it to the rest of your stones (in which case it's fine even if that group doesn't have an eye. It's either 2 eyes or a connection to a live group of stones.
In a 9x9 board it's really easy to avoid an invasion, since it's so f***ing hard to create two eyes in such a small place by invading. He just has to reduce you and confine you, and he's fine if he doesn't screw up big time. That's why I don't like 9x9, it feels too cramped
He contested it a bit at the bottom left but I mostly got that huge chunk to me.
But after that, I tried to play around the H-8 area and it fell apart and I lost because I can't materialize much defense (I was trying to create an eye) to deteriorate his % of territory.
Do remember that a group with only one eye is still dead, unless you can somehow connect it to the rest of your stones (in which case it's fine even if that group doesn't have an eye. It's either 2 eyes or a connection to a live group of stones.
In a 9x9 board it's really easy to avoid an invasion, since it's so f***ing hard to create two eyes in such a small place by invading. He just has to reduce you and confine you, and he's fine if he doesn't screw up big time. That's why I don't like 9x9, it feels too cramped
Sometimes it feels cramped but I prefer right now as I am not very good at some basic key moves that I keep missing (KillerDucky just did some reviews with me, thank you so much!)
I felt like I had achieved something pretty good but in reality, I gave the board to him without him knowin g maybe.
On October 23 2014 10:07 Hesmyrr wrote: I once tried to learn Go after reading this thread but abandoned it due to lack of thread activity - hope it can stay more active this time.
On October 23 2014 10:07 Hesmyrr wrote: I once tried to learn Go after reading this thread but abandoned it due to lack of thread activity - hope it can stay more active this time.
On October 23 2014 04:56 Garnet wrote: http://www.lifein19x19.com/forum/ Probably the most "active" Go forum. Join them if you want!.
I love that site. Malkovich games are entertaining to read though most of the time their explanation flies right through my brain.
The players write hidden comments about their plans, so you can see what they're thinking. Observers write hidden comments also -- strong players offer their own opinions on the situation. Weaker players can ask questions about what is going on etc.
On October 23 2014 17:04 gTank wrote: Hi guys, would you rather suggest using gokgs or online-go to start playing. I am an absolute beginner to this but always wanted to ply.
Both have their merit. I prefer Online Go right now but KGS is a bit more robust if I recall.
On October 23 2014 10:07 Hesmyrr wrote: I once tried to learn Go after reading this thread but abandoned it due to lack of thread activity - hope it can stay more active this time.
On October 23 2014 04:56 Garnet wrote: http://www.lifein19x19.com/forum/ Probably the most "active" Go forum. Join them if you want!.
I love that site. Malkovich games are entertaining to read though most of the time their explanation flies right through my brain.
I'll see if I can keep this thread alive with the amount of questions and mistakes I make ^^ the veterans are usually on KGS or just circling this thread, waiting for some new pup to come in!
Hi, i just started playing. I've been through the tutorial in op, and gave done some exercises, but I'm don't really understand the malkovich game yet. Is it advisable to just keep playing exercises for a while until i feel familiar with the basics? Or are there other ways of learning to understand tactical/strategic play?
At goproblems.com I'm hovering around 24 kyu after some exercises, so i'm pretty bad.
On October 23 2014 19:24 Yorbon wrote: Hi, i just started playing. I've been through the tutorial in op, and gave done some exercises, but I'm don't really understand the malkovich game yet. Is it advisable to just keep playing exercises for a while until i feel familiar with the basics? Or are there other ways of learning to understand tactical/strategic play?
At goproblems.com I'm hovering around 24 kyu after some exercises, so i'm pretty bad.
Skimmed the last two pages so I'm sorry if it was already posted.
But there was a cool talk that was uploaded from the US Go Congress talking about Go and its place in the modern world. Some cool comparisons are made with StarCraft and League.
On October 23 2014 20:08 gTank wrote: Won my first game against the bots, can someone explain me why he lost all the stones in the right? https://online-go.com/game/1009968
Hey, I started Go about a year ago, sitting currently at 13k-15k (probably more 15k, I haven't played much in the last 2 month, but motivation is coming back currently) Since I'm currently trying my best NOT to write my thesis, if you have any questions/games you want to discuss/anything, just ask :D
I don't like 9x9 personally but the big board can be overwhelming. For a beginner 9x9 is probably better until they have a grasp of the core ideas. Then I'd move up to 19x19 which is more interesting and more fun.
My main issue and KillerDucky said this, is that I don't know how to be aggressive and follow-through on that aggression (they'll usually block it and then I have no idea what to do after).
9x9 games need not necessarily be aggressive. There are games in which both players just take a side and it comes down to the endgame.
It depends on the way the game starts. If black plays in the middle first a typical way to be aggressive for white is to build a position on one side first then take a 33-point on the other side trying to get ahead in territory. Sometimes you can cut their position which typically leads to an all-or-nothing fight.
If both sides start with a 33-point each there is no real way to get agressive. Both sides have settled positions already. It comes down to the better endgame.
If you have a KGS account we can discuss this during a game. It's easier to explain if you have a example.
https://online-go.com/game/1010365 -- everytime I play aggressive, I lose. I take the outer-limits and try to be cross and I always lose more than I have the chance of winning.
When it feels like I have the start of a good surrounding, I shoot myself in the foot and lose any advantage I have. It's beyond frustrating because I don't know how to take any solid wall I have and push with it.
Every game, I go: "Okay, let me try and take advantage of my strong wall" and then I try and he just traps the one stone I tried to cross with and lose.
I think starting out with a lot of 9x9 games is the way to go. 9x9 really concentrates on fighting and getting a feeling for the basics of shapes. I think these are the most important aspects a complete beginner can learn. Tsumegos are another great way of training this. As soon as playing 9x9 evolves into stale barrier-building (which somehow inevitably happens with new players at some point) it is important to move on, and maybe come back later. ( I currently enjoy 9x9 really much again, but I play it more in the style where I drop my first 2-3 moves pretty random on the board and try to make the best out of the situation) Playing 13x13 starts concentrating on the strategical sense which is not used much in 9x9, where and when to take fights, etc. The problem is this builds on the basics taught in 9x9. So moving on to quickly leads to playing moves for the wrong reason. Still I don't think 13x13 has that many benefits and 19x19 is better. 19x19 is the most beautiful game and I think you should aim to play 19x19 as soon as you can comprehend what is going on on the big board. The other consideration might be practical: A 19x19 game just takes that much longer than a 9x9 and I think it is important to play as many games as possible.
Edit: If you want to play aggressive you need to start out with more aggressive moves. Your first 3 moves are really passive and small. Try playing 4-4, 3-4 or even more middle-ish moves. In the game you linked I actually thought he was the one playing the more aggresive-boardcontrol game and you played the passive moves to get territory. Edit2: Also there exist no strong walls in 9x9. The space is just soo small to take any advantage of walls.
On October 23 2014 23:03 giecha wrote: I think starting out with a lot of 9x9 games is the way to go. 9x9 really concentrates on fighting and getting a feeling for the basics of shapes. I think these are the most important aspects a complete beginner can learn. Tsumegos are another great way of training this. As soon as playing 9x9 evolves into stale barrier-building (which somehow inevitably happens with new players at some point) it is important to move on, and maybe come back later. ( I currently enjoy 9x9 really much again, but I play it more in the style where I drop my first 2-3 moves pretty random on the board and try to make the best out of the situation) Playing 13x13 starts concentrating on the strategical sense which is not used much in 9x9, where and when to take fights, etc. The problem is this builds on the basics taught in 9x9. So moving on to quickly leads to playing moves for the wrong reason. Still I don't think 13x13 has that many benefits and 19x19 is better. 19x19 is the most beautiful game and I think you should aim to play 19x19 as soon as you can comprehend what is going on on the big board. The other consideration might be practical: A 19x19 game just takes that much longer than a 9x9 and I think it is important to play as many games as possible.
Edit: If you want to play aggressive you need to start out with more aggressive moves. Your first 3 moves are really passive and small. Try playing 4-4, 3-4 or even more middle-ish moves. In the game you linked I actually thought he was the one playing the more aggresive-boardcontrol game and you played the passive moves to get territory. Edit2: Also there exist no strong walls in 9x9. The space is just soo small to take any advantage of walls.
I want to play passive and defensive, but the problem is that they always take advantage of it to eventually get more territory (usually the center to their side) so I lose in the end.
I will play a passive game and show you what happens.
I had a look at your game (edit: the first one https://online-go.com/game/1010365). You lost it not because a lack of aggression but because of a lack of ability. Sorry to be so frank. :p
You start with g7 and g3 which are reasonable moves but very solid and defensive. If you want to play a fighting game this opening is not the best choice (try tengen, e5, for instance). White e3 is a mistake. (It allows you to take the third corner with c7 or d7. If he invades your position with g5 you can either connect with h5 or fight f4. The latter is recommended. It leads to a more active game and a fight in which you are advantaged since you have more stones at the scene of battle.)
Returning to the game your fifth move h5 is really bad. You are reinforcing two strong stones with a very low and passive move. If you want to defend your position on the right side a good idea would be something like f4. Notice the huge difference in prospective territory between h5 and f4.
White e7 is also bad since it allows you to comeback with c7. c7 is both aggressive and a strong move.
After white c6 you make another very weak and defensive move in e8. (Notice the similarity to h5.) If I interpret this move correctly you try to connect to your stones on the right side. This is not good enough. What you should do is try to live in the top left corner and neutralize as much as possible of white's left side. Namely, I'd play b6. This is a dangerous gamble but necessary and leads to the kind of fight you seem to want.
Through move 29 you are defending your position against white's sente. Almost every move you make is weak and submissive. Some of the defensive are necessary but several were not. You make these moves because you can not read the continuation if you play elsewhere. That's a matter of reading ability which comes with practice. If you want to play a fighting game it is essential to have a strong grasp of tactics and good reading to back it up. Otherwise you will get crushed.
Have a look at the board after turn 30. You have been pressed into a low position on the right and top sides. Does this really look like the game of an aggressive player? In fact contrary to your assertion your play is submissive and defensive in general.
Sorry for the harsh words. I'm not trying to be mean but it seems you harbor misconceptions about your own game which makes it hard to get stronger.
Edit2: Forgot my advice: Try soving some tsumego's (reading exercises), goproblems.com is a good site. There are also books full of them. The best way to get stronger early on is to improve your reading ability. Nothing else even comes remotely close in effectiveness if you want to improve.
I think you would have won that with a different move 39. Playing move 39 on H7-J7-J8-H8-C2-B2-C1-B1-D10 would have given you 25 points, him 21. You were leading at that point because his aggressive moves were really clumsy (What's up with that rating anyway, there is no way he is 17k...^^)
https://online-go.com/review/36201 Move 52 you passed before the game was over. Same thing on move 66 at the very end. BTW also notice the computer auto-scoring marked two black stones dead (they are translucent), but didn't give white a point for territory. That's because the computer is a bit confused about this not-complete game. BTW the computer also sometimes makes mistakes even in complete games, and in that case you have to correct it by clicking on the stones to toggle their dead/alive state.
Most games are going to be decided on tactics like the move 33 ladder. Playing 9x9 is a good way to get more games, face more tactical situations, and try to get better at them. If you really want to move on try 13x13 next. On 19x19 you will spend too much time being lost in an ocean of confusion. Better to improve in a pond of confusion.
On October 23 2014 23:42 Hagen0 wrote: I had a look at your game (edit: the first one https://online-go.com/game/1010365). You lost it not because a lack of aggression but because of a lack of ability. Sorry to be so frank. :p
You start with g7 and g3 which are reasonable moves but very solid and defensive. If you want to play a fighting game this opening is not the best choice (try tengen, e5, for instance). White e3 is a mistake. (It allows you to take the third corner with c7 or d7. If he invades your position with g5 you can either connect with h5 or fight f4. The latter is recommended. It leads to a more active game and a fight in which you are advantaged since you have more stones at the scene of battle.)
Returning to the game your fifth move h5 is really bad. You are reinforcing two strong stones with a very low and passive move. If you want to defend your position on the right side a good idea would be something like f4. Notice the huge difference in prospective territory between h5 and f4.
White e7 is also bad since it allows you to comeback with c7. c7 is both aggressive and a strong move.
After white c6 you make another very weak and defensive move in e8. (Notice the similarity to h5.) If I interpret this move correctly you try to connect to your stones on the right side. This is not good enough. What you should do is try to live in the top left corner and neutralize as much as possible of white's left side. Namely, I'd play b6. This is a dangerous gamble but necessary and leads to the kind of fight you seem to want.
Through move 29 you are defending your position against white's sente. Almost every move you make is weak and submissive. Some of the defensive are necessary but several were not. You make these moves because you can not read the continuation if you play elsewhere. That's a matter of reading ability which comes with practice. If you want to play a fighting game it is essential to have a strong grasp of tactics and good reading to back it up. Otherwise you will get crushed.
Have a look at the board after turn 30. You have been pressed into a low position on the right and top sides. Does this really look like the game of an aggressive player? In fact contrary to your assertion your play is submissive and defensive in general.
Sorry for the harsh words. I'm not trying to be mean but it seems you harbor misconceptions about your own game which makes it hard to get stronger.
Edit2: Forgot my advice: Try soving some tsumego's (reading exercises), goproblems.com is a good site. There are also books full of them. The best way to get stronger early on is to improve your reading ability. Nothing else even comes remotely close in effectiveness if you want to improve.
At 21kyu, it goes without saying I lack ability.
I always open defensive and then turn 5, I go offensive to see if I can make anything happen.
I don't want to play aggressive, I just don't want to be passive and defensive all the time as it never leads to a victory, it always leads me to defeat as he ends up taking more territory. Your advice is really scrutinized but not much to take away except that I played passive and defensive, how do I draw anything from that except for the very problem I already underlined: I'm too defensive and not aggressive enough.
I don't want my games to be an all-out attack, I want to be an imprenetrable wall that blocks him out and lets me slowly advance for further territory gain (either through small captures or simply a good contour around his attempts).
On October 23 2014 23:43 giecha wrote: I think you would have won that with a different move 39. Playing move 39 on H7-J7-J8-H8-C2-B2-C1-B1-D10 would have given you 25 points, him 21. You were leading at that point because his aggressive moves were really clumsy (What's up with that rating anyway, there is no way he is 17k...^^)
I don't see how the B-C-D lower area is something I could easily defense or even claim remotely without losing those stones.
On October 23 2014 23:53 KillerDucky wrote: https://online-go.com/review/36201 Move 52 you passed before the game was over. Same thing on move 66 at the very end. BTW also notice the computer auto-scoring marked two black stones dead (they are translucent), but didn't give white a point for territory. That's because the computer is a bit confused about this not-complete game. BTW the computer also sometimes makes mistakes even in complete games, and in that case you have to correct it by clicking on the stones to toggle their dead/alive state.
Most games are going to be decided on tactics like the move 33 ladder. Playing 9x9 is a good way to get more games, face more tactical situations, and try to get better at them. If you really want to move on try 13x13 next. On 19x19 you will spend too much time being lost in an ocean of confusion. Better to improve in a pond of confusion.
I want to stay with 9x9 because it is the most uncomfortable for my play, but it is also the most frustrating because the crucial moves I can't see end up me losing the game.
I'm going to take everyone's advice and maybe play some goproblems.com before trying real games as I don't recognize ladder situations and such
I did give you examples of alternate lines that would have been more proactive. Moreover, you can't reasonably expect to be able to attack while having an impenetrable position yourself.
Let me iterate my advice. Improve your reading ability by solving problems. This is the single most effective way to get better at Go. You can easily get to around 5k without bothering with niceties such as strategy or playstyle (defensive, aggressive).
A note regarding Goproblems.com. Unfortunately, the problems vary in accessibility. Some of them are well documented. Others have no text at all and one is expected to find the best move in general. It is hard for a beginner to tackle problems without direction or maybe a hint. There are problem books for beginners, the most famous being Graded Go Problems for Beginners by Kano Yoshinori. They should be on the internet as pdf if you don't want to buy them.
On October 24 2014 00:46 Hagen0 wrote: I did give you examples of alternate lines that would have been more proactive. Moreover, you can't reasonably expect to be able to attack while having an impenetrable position yourself.
Let me iterate my advice. Improve your reading ability by solving problems. This is the single most effective way to get better at Go. You can easily get to around 5k without bothering with niceties such as strategy or playstyle (defensive, aggressive).
A note regarding Goproblems.com. Unfortunately, the problems vary in accessibility. Some of them are well documented. Others have no text at all and one is expected to find the best move in general. It is hard for a beginner to tackle problems without direction or maybe a hint. There are problem books for beginners, the most famous being Graded Go Problems for Beginners by Kano Yoshinori. They should be on the internet as pdf if you don't want to buy them.
I definitely agree that I lack reading of scenarios and situations as I see them come up but no idea how to overcome them adequately (even if they are to my advantage). So I'll take your advice and try some Go Problems because right now I'm a little frustrated by my own lack of foresight.
For Go problems, should I just start at the beginning or?
On the main page there is a green button labelled Go to Problems. IIrc you start with the easiest problems, so that shoudl be OK. You can also chos time trial on the right of the page. This allows you to chose the difficulty level.
When I was in high school I tried to learn this game but got bored before I could wrap my head around it. I got my butt kicked by that Aya program a hundred times and then I quit, ha. Now I'm doing The Interactive Way To Go and it makes it so easy to figure out, so I'm giving it another shot. What free programs do you guys recommend for beginners, for watching Go games, and playing against AI? And where is a good place to find game reviews, preferably for 9x9?
gokgs.com is the most used goserver in the west. You can do pretty much all you asked for there. There are some rooms designated for gamereviews, like check out KGS Teaching Ladder. And for playing: I recommend playing against humans, and there are enough of your level on KGS. If you feel just like watching games, for your level whatever game is on top on KGS should be good enough to watch.
On October 24 2014 01:56 mahrgell wrote: gokgs.com is the most used goserver in the west. You can do pretty much all you asked for there. There are some rooms designated for gamereviews, like check out KGS Teaching Ladder. And for playing: I recommend playing against humans, and there are enough of your level on KGS. If you feel just like watching games, for your level whatever game is on top on KGS should be good enough to watch.
I second those advices (but as a former KGS admin I could be biased). I play as explo there, all the beginners are free to ask me for a game or a review when I'm online.
All right, thanks. I'll check it out! Though I don't have internet at my apartment yet and don't plan on having it for a while. What about stuff I can do offline while I'm not at a library?
KGS also comes with handy review tools, the ability to save reviews and a lot of people to play against. I don't play very often anymore unfortunately, but usually everyone in the teamliquid channel is very nice and will help beginners. I met my teacher there and he basically carried me from 25k to 1k with countless reviews. The go community is just great, because many strong players are interested in teaching the game to spread it.
For times without internet you can download a .sgf app (sgf is the file format for kifus) for your smartphone etc and download pro games to watch. When I started playing I looked over about 1000 pro games and ended up with a feeling for shape without ever reading anything about it. Generally, the stronger you get the more you can actually learn from a pro game, but if you just look at a few games to see how professionals play, it will help you. Best source for kifu imo is here: http://igokisen.web.fc2.com/news.html There are a number of different sites to get games.
I got the right answer to this problem by using an intuitive argument (I tried to prevent black from getting an eye on L11 by putting a stone on L10), but to be honest, I really don't understand it. The comments didn't really help me either. Can anyone explain what the point of the problem is exactly? Sorry for the question, but I think Im missing some basic stuff here ):
I got the right answer to this problem by using an intuitive argument (I tried to prevent black from getting an eye on L11 by putting a stone on L10), but to be honest, I really don't understand it. The comments didn't really help me either. Can anyone explain what the point of the problem is exactly? Sorry for the question, but I think Im missing some basic stuff here ):
A group with 1 eye has the advantage vs a group with 0 eyes in a capturing race. So the point is to change it into a 0 eyes vs 0 eyes situation. Then it's a seki. If black makes en eyes, he wins the capturing race.
I got the right answer to this problem by using an intuitive argument (I tried to prevent black from getting an eye on L11 by putting a stone on L10), but to be honest, I really don't understand it. The comments didn't really help me either. Can anyone explain what the point of the problem is exactly? Sorry for the question, but I think Im missing some basic stuff here ):
A group with 1 eye has the advantage vs a group with 0 eyes in a capturing race. So the point is to change it into a 0 eyes vs 0 eyes situation. Then it's a seki. If black makes en eyes, he wins the capturing race.
Ah, in that context it makes a lot more sense indeed. Thanks!
Done I'm not quite comfortable enough to start playing against players yet though, I first of all want to just do exercises and get a feel for the game.
On October 24 2014 05:40 nimbim wrote: KGS also comes with handy review tools, the ability to save reviews and a lot of people to play against. I don't play very often anymore unfortunately, but usually everyone in the teamliquid channel is very nice and will help beginners. I met my teacher there and he basically carried me from 25k to 1k with countless reviews. The go community is just great, because many strong players are interested in teaching the game to spread it.
This is my big appeal as well; for online competitive games like CSGO/LoL, it can be a bit difficult to find some people to help or play with/against + the communities can get abrasive.
With Go, there's a lot of pride and interest in teaching new players and almost a sense of duty and responsibility to breed newcomers into proper players. That and the undertone of honor and politeness is something that makes me feel really sage and relaxed when playing. I get frustrated when I lose, but it is a different frustration where I don't feel the bitterness of my opponent winning, implicitly smug (for some reason) but rather just disappointed in my own lack of ability and that inner-desire to persevere!
I don't know, strange mix of feelings playing Go that make me really calm.
Ugh... I can't run gokgs. My computer's clock is fucked up and resets itself to the wrong time every few hours (yeah, I need a new computer). This causes Java to block some Java apps, like gokgs. I'll try to mess with Java settings... Some day. But for now, what's the next best online and offline stuff for playing and watching go?
EDIT: Well. Got the browser client working. I think? Won't let me create an account. And about gobase.org: I need to donate 20 euros to see the archived games on that website?
On October 25 2014 11:05 Capricis wrote: Ugh... I can't run gokgs. My computer's clock is fucked up and resets itself to the wrong time every few hours (yeah, I need a new computer). This causes Java to block some Java apps, like gokgs. I'll try to mess with Java settings... Some day. But for now, what's the next best online and offline stuff for playing and watching go?
You can play on Tygem http://www.tygemgo.com/ Mostly Koreans and Chinese are there, but you can watch Tygem 9D vs 9D games pretty much all day. I don't recommend this server for beginners, the lowest rank there is 18k I believe and the kyu ranks are full of sandbaggers.
I played my first game against an opponent and got smashed in a 9x9 game. Nothing weird, but I'd like to ask your opinions on the situation at the start of the game as well as your opinions on my considerations. This is how the game was played the first few moves, starting with black, my opponent (i was white) : D-6 -> F-4 -> F-6 -> C-4 -> D-4 -> D-3
Ok, so my opponent began with D-6, which wasnt really what I expected. This move seems somewhat aggressive, aiming towards the centre immediately. Is this a standard opening? If so, what are the considerations in the continuation? Is it an overextension? How am I supposed to punish it? I've (in my head) been playing around with C-7 as my next move, going for that corner, so his play feels 'in the middle' not doing anything, unless he commits to an attack. I'm not sure how that will play out, though.
I played F-4 as my first move; I just copied my opponent. Would G-3 have been a better choice? Or am I giving my opponent too much of a space advantage that way? Suppose we played F-4 -> G-3 -> G-7 -> C-3, would I be able to put enough pressure on the left side to counter his presence in the middle?
My opponent played F-6, and I decided to 'just do something' and played C-4. My opponent decided to step in and play D-4. I went defensive and played D-3. C-4 probably is just bad in those situations. I was wondering though, is playing D-3 is too defensive? I tried to think about playing D-5 instead of D-3, but I couldn't really find anything good after black's E-5. I have thought about E-5 after black's D-4, and I think I can at least seperate one of blacks stones, preventing they form one big group, although I'm not sure about the drawbacks that may have.
I'm pretty insecure about anything I've said so far. Can some of you point me in the right direction regarding general principles? In case anyone's interested, link to the entire game: http://online-go.com/game/1014407
Take what I say with a grain of salt. My experience on the 9x9 board is lacking.
The opening through D3 is reasonable for both sides, though probably not optimal. D3 is necessary, otherwise your position will desintegrate. It's a good move.
Typical opening moves are c7 or c6 as well as just taking the middle with e5. Your opponents move in the game D6 seems a half-measure. It doesn't really commit to controlling the centre like e5 does, but still leaves white with the opportunity to invade the top left corner with c7 later.
If I were in your position I'd first build a solid position with 2 (or 3 possibly, depending on blacks moves) stones on the lower sight of the board and the invade at c7 to neutralize the top side.
On October 25 2014 22:37 Hagen0 wrote: Take what I say with a grain of salt. My experience on the 9x9 board is lacking.
The opening through D3 is reasonable for both sides, though probably not optimal. D3 is necessary, otherwise your position will desintegrate. It's a good move.
Typical opening moves are c7 or c6 as well as just taking the middle with e5. Your opponents move in the game D6 seems a half-measure. It doesn't really commit to controlling the centre like e5 does, but still leaves white with the opportunity to invade the top left corner with c7 later.
If I were in your position I'd first build a solid position with 2 (or 3 possibly, depending on blacks moves) stones on the lower sight of the board and the invade at c7 to neutralize the top side.
Thanks for the reply!
Would you think building the solid position in the bottom should be started at the third row? So instead of F-4 as a first move, i maybe should've answered G-3 going into C-3 and E-4 for example? Dependent of course on opponent's play, but would that shape be all right for the purpose of getting a solid position? Or are you talking about after the D-3 move?
I think this game shows the character of 9x9 pretty well... When playing the typical wallbuilding game, you can easily lose in the first 4 moves on 9x9... But in fighting games, even on move 15+ the result can be unclear. Playing 4-4 (D6) on first move is rather unusual on 9x9 as it is usually losing to G3. When he jumps too far on his third move you can always invade on 3-3 under him, if he jumps close, White can expand farther and win by komi. But F4 basically copies the mistake, and the way Black played it afterwards was quite good. One could argue, that your move 4 could be played elsewhere, but I feel it is quite difficult at this point already. If you invade 3-3 now in the topleft, you will end in gote and then suffer in the lower right, as you have no secured base there. The sequence that happens afterwards is pretty standard until move 15, but it is a straight path to losing for white. No point in discussing what happened afterwards
But all I said has to be discounted a bit by the fact, that the players here were double digit kyus. This makes the oucome of such moves much less predictable. Only because for Dan players the game can be easily over by move 4, it does not mean, a beginner should think "uh, first 4 moves, gg!" And as I said above... you can easily get into fighting games, that are difficult and undecided for quite some time so 9x9 wins back some appeal once stronger.
White still may have a chance after answering D4 D3 C5 with the atari at E4. If black defends at D5 white's lower side is now more stable due to E4 and white can invade at G7. If black plays something like G5 instead he probably has the game but he has to play in a surprisingly precise way to force the win.
Thanks again for the replies, i felt i learned a lot by reading your replies and thinking them over.
@mahrgell: a question: I'm afraid I don't really get this sentence: "When he jumps too far on his third move you can always invade on 3-3 under him, if he jumps close, White can expand farther and win by komi." I assume by 'his third move', you mean move 3 of the game, moved by my opponent (or his second move)? What do exactly mean by jumping too close or too far in this context?
On October 26 2014 06:40 Yorbon wrote: Thanks again for the replies, i felt i learned a lot by reading your replies and thinking them over.
@mahrgell: a question: I'm afraid I don't really get this sentence: "When he jumps too far on his third move you can always invade on 3-3 under him, if he jumps close, White can expand farther and win by komi." I assume by 'his third move', you mean move 3 of the game, moved by my opponent (or his second move)? What do exactly mean by jumping too close or too far in this context?
I meant move3 of the game, counting total number of moves and not moves per player is pretty much standard in Go.
About your question: When he one space jumps (or one off to the side) in any direction, a 3-3 invasion in the topleft becomes more difficult and less attractive and his corner is more secure. But White can simply win on the remaining board, as he moves too slow to keep up. If he decides to 2 space jump, one off from it or something completely different... a 3-3 invasion remains easily possible.
On October 26 2014 06:40 Yorbon wrote: Thanks again for the replies, i felt i learned a lot by reading your replies and thinking them over.
@mahrgell: a question: I'm afraid I don't really get this sentence: "When he jumps too far on his third move you can always invade on 3-3 under him, if he jumps close, White can expand farther and win by komi." I assume by 'his third move', you mean move 3 of the game, moved by my opponent (or his second move)? What do exactly mean by jumping too close or too far in this context?
I meant move3 of the game, counting total number of moves and not moves per player is pretty much standard in Go.
About your question: When he one space jumps (or one off to the side) in any direction, a 3-3 invasion in the topleft becomes more difficult and less attractive and his corner is more secure. But White can simply win on the remaining board, as he moves too slow to keep up. If he decides to 2 space jump, one off from it or something completely different... a 3-3 invasion remains easily possible.
Clear?
I was a bit confused by the words far and close, but put this way it's quite clear. Thanks again.
@ Hesmyrr: I'm afraid I'm not good enough to comment on your game the way you'd like, but I was wondering about one thing. Move number 27 was played at H4 probably (?) to prevent cuts at that location. However, I felt more for a(n in my eyes) more dynamic defense at J4, indirectly defending both H4 and J3. What do you guys think, would that work out all right?
On October 27 2014 09:20 Hesmyrr wrote: Can someone take a look at my 9x9 game and comment? I'm pretty sure I made a over-play here and wanted to ask how my play could have been punished.
Until move 24 there is not much I would argue about on your level. Pretty solid play from both sides there. (I don't say it couldn't be done better, but that would require more analysis then useful here for learning purposes)
Move 25 is the first major mistake. Just H4 without doing that exchange would be better here. You are not exactly making it easier to kill him here by playing those moves and he can't make 2 eyes locally anyway (but the cut at E4 makes the situation very interesting.)
His move 36 at J1 is really bad, because he only focusses on making life via Ko. But when it is Ko anyway... It would be much better for him to create the Ko in a way you can also not afford to lose it. So he should just cut E4 and Black would be in huge trouble after E4-D3-F5. (it would turn into a Ko, that is unwinnable for Black, as White has local threats, Black doesn't) But he gives you a chance. If you thought carefully about my last comment, you know which move is a MUST on move 37. It is not the move you played...
If you just connect E4 here, without even taking the Ko first (if you do, he can play it as threat), his group is still not alive, he still has to win the Ko. But even worse. In fact his entire center is simply dead. You could cut at F6, or just play B4 to show him that he has no eyespace while you are perfectly connected. J3 is simply bad. To win the race you have to play Ko anyway. Spending a move that pretends to avoid a Ko which is still there is wasted. Now in the resulting Ko you waste another move. D1 doesn't do anything. It should be at F1/2. Just imagine taking his stones via the Ko. For that you have to play F1 and F2 anyway... If you capture 2 more stones with D1 does not matter. Also you can't shortcut by getting in from the other side. It really just wastes another move. But if properly played, after playing F1 White still wins the Ko, as he wins by one threat. Black could make life in the bottom left with his final threat, but White would win with about 5-8 points (so komi +- a few)
Thanks for the analysis. I was actually white but you comprehensively covered both sides anyway. Move 36 is reflection of my poor understanding regarding Ko. I know how to seize the opportunity when such situation end up happening, but I can't - or likely even don't know how to - anticipate or plan for one in advance. Due to time constraint I suspected it might be something else but now I see it's not due to the fact that I have guaranteed eye. I also picked H5 over F5 because of worry about black H6/J6, another concern which I realize is unfounded after more critical look under no time constraint.
The reason why I posted the game is because of move 20 at G3. I recognized the need to play down there but wanted to extend upper-right stones one step further because it didn't feel "secure enough", which in retrospect was too passive of me. When black played at H3, I panic invaded with little expectation it was going to live (invasions are another thing I am clueless about). So I wanted to know if it would have died assuming black play perfectly... but after some analysis looks like I could have made Ko no matter what I think? (If black H4, E2-D2-E4-D3-F1 for one eye space in gote) I guess that's what you mean by E4 cut, since if I E4 right after hypothetical black H4, black E2 kills.
One thing I also wanted to ask about is my strategy. I am pretty sure I lost after move 5 to 16. Two black stones are way too close for me to successfully utilize my wall. If black decided to make me leave small on the corner... for example, play E4 as move 33 instead of J2, I think my center group outright dies (never mind the need to save the corner in gote) which means my G3 invasion wasn't good idea to begin with regardless of life-or-death? What should I have played as move 18 to better myself, or was my response to black E7 wrong from the very start?
Huh, looking at the reply I actually repeat most of your points (I must have missed them since I was exclusively thinking about right-bottom stones during initial reading) but fortunately my question still stands because of color misunderstanding. I wish I was as good as black on fuseki
@hesmyrr G3 is reasonable here. I feel like you are overstrategizing 9x9 a bit. Honestly, 9x9 is just a tactical battle, where for most moves there is no strategic guideline, that tells you if they are good or bad, but simple pure tactical reading decides, whether those moves work or don't. In this situation White has to do something like that, G3 would be among my first ideas as well, and I don't see a super simple refutation, so it is a fine choice.
Talking about perfect play... Uh well, I don't know perfect play, and honestly, this situation is quite difficult with lots of possibilities. By my own judgement White won't die unconditionally and easy. But wouldn't dare to say, who wins here. And yes, by E4 cutting I of course meant E2-D2-E4.
And about move5-16 and general strategy. When you played move 4 you basically decided to play that kind of game. If you would hve wanted to prevent that, you could have D6 instead of C6, mirroring Black, but I'm no expert on 9x9 openings to tell you how this would turn out in first move advantage vs komi value. After that your sequence seems pretty straight forward. Also as we could see in the game... The wall is in fact working, even with an additional move at H3 there are problems for him, that are not easily solved. (I'm against using the argument of "but it worked" because that is often only due to opponents making shitmoves... but here the situation later is so difficult and G3, that it obviously still everyones game, even with better play) Even though Black playing H3 at G4 would have been much better for him. About move 18... Difficult. You really want to play where you played... but G4 can probably win for Black. but not playing there also hurts a lot...You could now either backtrack to move 4 and call it bad... Or just try what happens with stuff like attaching to F4 from either side, which again is a lot to read and fight and difficult to predict:D
On October 27 2014 19:28 mahrgell wrote: @hesmyrr G3 is reasonable here. I feel like you are overstrategizing 9x9 a bit. Honestly, 9x9 is just a tactical battle, where for most moves there is no strategic guideline, that tells you if they are good or bad, but simple pure tactical reading decides, whether those moves work or don't. In this situation White has to do something like that, G3 would be among my first ideas as well, and I don't see a super simple refutation, so it is a fine choice.
Talking about perfect play... Uh well, I don't know perfect play, and honestly, this situation is quite difficult with lots of possibilities. By my own judgement White won't die unconditionally and easy. But wouldn't dare to say, who wins here. And yes, by E4 cutting I of course meant E2-D2-E4.
And about move5-16 and general strategy. When you played move 4 you basically decided to play that kind of game. If you would hve wanted to prevent that, you could have D6 instead of C6, mirroring Black, but I'm no expert on 9x9 openings to tell you how this would turn out in first move advantage vs komi value. After that your sequence seems pretty straight forward. Also as we could see in the game... The wall is in fact working, even with an additional move at H3 there are problems for him, that are not easily solved. (I'm against using the argument of "but it worked" because that is often only due to opponents making shitmoves... but here the situation later is so difficult and G3, that it obviously still everyones game, even with better play) Even though Black playing H3 at G4 would have been much better for him. About move 18... Difficult. You really want to play where you played... but G4 can probably win for Black. but not playing there also hurts a lot...You could now either backtrack to move 4 and call it bad... Or just try what happens with stuff like attaching to F4 from either side, which again is a lot to read and fight and difficult to predict:D
Locally probably g1 is the losing move. First it's damezumari (meaning your play removes a liberty to your group), never a good idea. Second extending to d2 looks like winning the fight, but I'm too lazy to read all variations here
Wanted to try and learn a bit of Go. The very easy computer is really awful on that online-go website. It let me capture 24 stones with one move. I'm excited to get destroyed easily by my first real opponent though, it's such a great game
On October 27 2014 12:21 Hesmyrr wrote: One thing I also wanted to ask about is my strategy. I am pretty sure I lost after move 5 to 16. Two black stones are way too close for me to successfully utilize my wall. If black decided to make me leave small on the corner... for example, play E4 as move 33 instead of J2, I think my center group outright dies (never mind the need to save the corner in gote) which means my G3 invasion wasn't good idea to begin with regardless of life-or-death? What should I have played as move 18 to better myself, or was my response to black E7 wrong from the very start?
You are indeed correct. After move 17 and especially move 19 at H3 the game is essentially over. Your invasion at G3 doesn't work and and it could be difficult to even live with your middle group. For reference black can answer G3 with F3.
I'm not sure where the problem lies with your opening. It's possible that c6 was already an overplay. More likely though your handling of the invasion was not correct. E7 E6 D7 D6 is beyond reproach but after C7 it is better to play F7. blacks corner is not alive yet. He has to add an extra move which allows you attack the lower side. It will likely come down to an all or nothing fight.
https://online-go.com/game/1031792 -- Had a very successful game today. He played a very aggressive opener and I managed to curb his start pretty hard. I almost lost as we moved up North but I ultimately succeeded by his own mistakes and my own luck.
On October 30 2014 01:28 nimbim wrote: I reviewed that game for you https://online-go.com/review/36968 Feel free to ask if anything is unclear.
Move 60: That False Eye really had me scared, I think he and I both thought that was two eyes!
Move 65: A1 -- I thought two eyes had to be connected directly together?
Move 71: E9 - I definitely felt I was in a similar situation where I could make one false move and lose that massive capture
Move 64: H1 - I felt that entire string of play would be something someone much more confident could do. It's smart move, but way beyond my ability to foresee how it'd turn into my favour, no?
I understand the overall lesson. I was surprised how quickly the initial skirmish spiraled.
On October 30 2014 01:44 Torte de Lini wrote: Move 65: A1 -- I thought two eyes had to be connected directly together?
That string of black stones is connected. You can have 1 eye in the upper left corner and 1 in the bottom right, as long as they belong to the same group.
Move 64: H1 - I felt that entire string of play would be something someone much more confident could do. It's smart move, but way beyond my ability to foresee how it'd turn into my favour, no?
Right now you probably can't read these moves in a game, but it is a very common shape. You will see it very often in your games and eventually know what you can do.
I try to play against player 1~2k better than me frequently You can also see progress of your rank on the profile page, I think. Seems like it's the blue bar thing.
https://online-go.com/game/1033281 -- how did he win at the bottom-left? I felt it was a "whoever sticks their stone to surround, captures. but I learned that it didn't matter in the end, yet, I still lost. I'm so confused.
So frustrating, I thought I won but I didn't close it when I could have.
On October 30 2014 08:00 Hesmyrr wrote: My first instinct is D2 over A3 but I am not sure if I am being too defensive.
That's my biggest fear too. I realize now that my over-defensiveness is also linked to my lack of readability as everyone has kept telling me. I think with time, I'll start to branch out more and do more crosses when I can see ahead. Right now, I can read maybe 5 moves (MAX) ahead.
Fran's Library on OGS is pretty good for a complete beginner, as it plays out all possible variations with some explanation. It's really helpful to have the rationale to dealing with certain types of problems explained (reduce eyespace, stealing eyes), or i'd be completely lost.
Fran's Library on OGS is pretty good for a complete beginner, as it plays out all possible variations with some explanation. It's really helpful to have the rationale to dealing with certain types of problems explained (reduce eyespace, stealing eyes), or i'd be completely lost.
Fran's Library on OGS is pretty good for a complete beginner, as it plays out all possible variations with some explanation. It's really helpful to have the rationale to dealing with certain types of problems explained (reduce eyespace, stealing eyes), or i'd be completely lost.
I also do some on pdf, mainly to force myself to play out the variations in my head instead of relying on trial and error which is easily done on an app/computer. http://tsumego.tasuki.org/
On October 31 2014 21:54 Torte de Lini wrote: Damn, I dropped to 22kyu, I'm doing awful T_T
Don't worry about rank if you are just starting out. It's a bad trap that people shouldn't fall into, but it ends up happening. Mostly because it stops them from experimenting with different things.
Yeah, I try to play aggressive instead of safe moyo games when possible (so no unsupported moves that have no chance of working out in the first place) to train my reading skills. I lose some games due to this but I assume it's going to all be helpful experience later on, though of course I know that itself can create bad habit if I don't constantly check myself.
Also, most games 19x19 games I lose is due to having a large group killed or having multiple small groups killed (need to learn how to spot weak groups properly), which destroys my frameworks. I usually resign when that happens. Is there any value in continuing to play such games?
If you were black: Don't play 19x19. Try 9x9 or 13x13. If you were white: No point in reviewing the game, the difference in strength between both players is several magnitudes.
About continuing lost games: If the game is lost, because there is not enough area that could compensate for your losses, even if your opponent messes everything up: resign Otherwise: Try to "win that area". You most likely fail, then you can resign... And if it works out, good! But don't just continue games to "practice your endgame". There is no endgame practice in playing out decided games, as the leading players will most likely not care at all about what you do. Only when something is at stake for both players you can practice. (assuming your opponent is not a sparing partner you have some agreements with, but some random opponent)
You did better than your opponent For now you should just try to identify when your opponent's move does nothing and take something big for yourself. At your rank the game will most likely not be decided by 1 dead group. Often times neither side will actually know if a group is dead or alive and if you think there is still something useful to do on the board, then play it to get as much go experience as possible. A ddk (double digit kyu) who doesn't resign is not being rude.
On November 06 2014 23:14 mahrgell wrote: If you were black: Don't play 19x19. Try 9x9 or 13x13. If you were white: No point in reviewing the game, the difference in strength between both players is several magnitudes.
About continuing lost games: If the game is lost, because there is not enough area that could compensate for your losses, even if your opponent messes everything up: resign Otherwise: Try to "win that area". You most likely fail, then you can resign... And if it works out, good! But don't just continue games to "practice your endgame". There is no endgame practice in playing out decided games, as the leading players will most likely not care at all about what you do. Only when something is at stake for both players you can practice. (assuming your opponent is not a sparing partner you have some agreements with, but some random opponent)
You did better than your opponent For now you should just try to identify when your opponent's move does nothing and take something big for yourself. At your rank the game will most likely not be decided by 1 dead group. Often times neither side will actually know if a group is dead or alive and if you think there is still something useful to do on the board, then play it to get as much go experience as possible. A ddk (double digit kyu) who doesn't resign is not being rude.
Thanks for the review, that was actually really useful! I think the reason why I make small moves in the early game is because I have no confidence in my reading. If I don't reinforce them into why seems like a strong group (in my inexperienced eyes), they tend to get killed in the mid game.
Wow, thanks for posting links to online-go.com ... Had not heard about it before, but it really looks neat. Doesn't feel like it has not been worked on since ~2000, like KGS or IGS... : ]
e: Oh, and great to see this thread being this active. Yay, people play Go! ^_^
Yo, I figure I should meet my 1 post/year quota to do a little bit of shameful self-promoting - I recently started twitch & youtube channels to stream/record my games/reviews:
I also was just recently granted audio review powers on KGS, so if any newer/weaker players (I'm 3dan on KGS for reference) would like teaching games/reviews there, get in touch with me and I'd be happy to help! I'd also be happy to provide non-KGS games/reviews if that platform doesn't suit you.
Just won my first game in awhile! Very happy with how I played it out! Only 35 seocnds per turn, kept my cool, didn't relent and made very few mistakes
I've been playing go again after a long hiatus, for the past few days, and I'm noticing kgs is a lot less active than it used to be. Where are the happening places for online ranked go, now? I have such a long wait right now for an opponent at my terrible level lol
On December 12 2014 10:03 micronesia wrote: I've been playing go again after a long hiatus, for the past few days, and I'm noticing kgs is a lot less active than it used to be. Where are the happening places for online ranked go, now? I have such a long wait right now for an opponent at my terrible level lol
I don't remember your level, but http://online-go.com has more beginners (KGS DDKs) lately than KGS. And many strong players (KGS 5d+) have migrated to http://www.tygem.com .
- hikaru no go (raised awareness in the west) - professional scene, new professional system in US and Europe - the jubango between lee sedol and gu li (most epic match of this century so far) - computers vs human (see yoda norimoto vs computer : - michael redmond (only western pro to reach 9th dan - go seigen, kitani minoru, lee changho (legendary players, go seigen passed away at 100yo last year)
I'll be referencing Go for several articles. I'm doing one right now where the tools and accessibility of the game are areas that esports can either match or imitate, but the attitudes and values in addition to how integral the game is to mainstream culture in the Eastern hemisphere is something esports has yet to penetrate.
What do you guys think? I think the attitudes and values taught in Go are probably my favourite about this game. Teaching games, Tsumego, Analyze mode on online clients and the sensei/deshi dynamic. Professional Go schools are also very cool.
Any videos about the culture of Go, Go schools, etc. is very neat and interesting. I want to draw parallels to the dynamism (or difference of) in pro teams.
Started playing again recently after a few months of not playing at all. I'm around 3 kyu right now according to KGS (which seems to be around 1 dan on Tygem). Losing games every day because of horrible misreads due to lack of practice. If anyone wants to play shoot me a message, I'm nabya on KGS or crouga on OGS.
On February 04 2015 04:29 Magggrig wrote: I think esport and go are extremely different... Go is driven by passion, esport is driven by thirst for glory.
I think both embody the feeling of a competitive game. Go is very mature and its age shows what values it still holds to: honor, dedication, passion for the game. Esports is young, new and immature; it tries to reach for new heights because of how fast its growing and the general attitude of its enthusiasts.
Doesn't make them all that different and if anything, esports could learn a lot from Go
As for active servers i play on wbaduk and tygem now - when i play online i want fast games. BTW if anyone in manchester uk is reading this - head down to the shakespear on thursday evenings (although i probably wont be there as dealing with a kid atm!) there are at least 8 people there every week of strengths from 15k-2d.
But i mainly play on wbaduk because its hilarious being 7-10 stones weaker on them than i am at home (blitz lol) Also i get the feeling people on there have *much* *MUCH* better fundamentals.
w baduk also has a lot of pro lectures in english which will give you some of those fundamentals. I wish i had resource like that 10 year ago. Then it was only sensei which whilst a good resource is not structured or filtered enough.
goproblems is also excellent but runs slow as fuck these days imo (the guy who made it made star sontata ... sold that and then kinda gave up on everything as far as i can tell)
I think the attitude in go is incredible. There is nowhere else i can play games and see peoples attitudes naturally improve. It is a game of self improvement. The whole playing in a way to help your opponent improve if huge - especially when playing face to face. As a result a lot of attitudes in games sicken me
The only sad thing is that i think go and esports are in competition for players.
online go seems to be good up to 10k, i cant get a game on there though.
failing that let me pimp guo juans website (http://internetgoschool.com/). I have no affiliation just a ton of respect for her and mingjiu I learnt more through them in 6 months than i did in 10 years prior.
There are also some korean inseis offering structured lessons if you are stronger.
I think the Korean semipro/professional esports feeder structure is essentially derived from the Insei/Yeongusung system that have been established for Go schools for centuries. I think players at the professional level take their competition seriously, and culturally it has enough respect to have a well-established business side. "Star culture" is relevant in every field, and professional Go is no exception.
As a game I think it's strategically unmatched, and there are a lot of cool Art-of-War-esque proverbs that apply to games beyond itself.
I think the accessibility of Go is unmatched. It's rules are simple, there's no hierarchy like in Chess and its nature and surrounding welcomes old values that we forget: dedication, diligence, respect and honor. I get a lot of pride saying I know how to play Go/actively (I try to) trying to play. Maybe it's just me, but I'm not sure.
I feel like with Go, I'm part of something that inherits a lot of what I want in a game and an endless depth that mirrors that of Dota 2 or League as its meta changes (and the balance pushes for change).
I can write a lot about it and I will (: Thanks everyone
Amateur Go is a mix of two things, most people who play it don't approach it as a competitive thing, those who do pursue it certainly do so in a way that western esports does not come even close to doing. Individually it is much closer to chess in the game-sport duality.
The game is no less hierarchical than chess in much of its nature and I think so much of what you're posting is just a western hipster view
On February 04 2015 09:25 Kupon3ss wrote: Amateur Go is a mix of two things, most people who play it don't approach it as a competitive thing, those who do pursue it certainly do so in a way that western esports does not come even close to doing. Individually it is much closer to chess in the game-sport duality.
The game is no less hierarchical than chess in much of its nature and I think so much of what you're posting is just a western hipster view
Really? Because I was paraphrasing the point-of-view of a Chinese sports writer who described the history of the game being non-hierarchical. Chess has ranks and hierarchy of value for units.
Go has none of that and is simplified, stemming from the idea that it was originally played by lower folk with too much time on their hands before later being accepted into higher society.
The bottom-line of what I'm comparing the two from is still valid (in my opinion), labeling my enthusiasm for the other aspects as "western hipster" is just derogatory.
The game historically originated in the upper classes, most likely from divination and from warrior nobles The entire Dan system is based upon ranks and hierarchical designations, and titles that have been governing the game for thousands of years
Bottom line is that you don't know much about the history of Go or Chess or even esports and you're just filling in the blanks to fit your hipster conception of the game
On February 04 2015 09:37 Kupon3ss wrote: The game historically originated in the upper classes, most likely from divination and from warrior nobles The entire Dan system is based upon ranks and hierarchical designations, and titles that have been governing the game for thousands of years
Bottom line is that you don't know much about the history of Go or Chess or even esports and you're just filling in the blanks to fit your hipster conception of the game
With that kind of dismissal, there wouldn't be convincing you otherwise. From what I've read, seen and listened to, I learned otherwise. But if my only crime is your perception of a "hipster conception", then I'm fine with that as I hardly respect your views as is and it only serves to crush my interest, not to educate or inform.
edit: I have the stomach flu, so I couldn't get to work, but my co-worker sent me the files:
On February 04 2015 09:37 Kupon3ss wrote: The game historically originated in the upper classes, most likely from divination and from warrior nobles The entire Dan system is based upon ranks and hierarchical designations, and titles that have been governing the game for thousands of years
Bottom line is that you don't know much about the history of Go or Chess or even esports and you're just filling in the blanks to fit your hipster conception of the game
You discredit yourself with that kind of pathetic offensive post...
I think there are many aspects of Go, many different kind of people that play it in many different context. It is true that go is less hierarchical than other games, it doesn't matter that much what rank you are because you can put handicap stones, go players don't mind as much (talking from the go players I know) for ranking as videogames players do. But it is true that Go is hierarchical, it would be foolish to pretend ranks don't exist at all.
By the way I also read that Go was played by higher classes in Japan.
On February 04 2015 09:37 Kupon3ss wrote: The game historically originated in the upper classes, most likely from divination and from warrior nobles The entire Dan system is based upon ranks and hierarchical designations, and titles that have been governing the game for thousands of years
Bottom line is that you don't know much about the history of Go or Chess or even esports and you're just filling in the blanks to fit your hipster conception of the game
You discredit yourself with that kind of pathetic offensive post...
I think there are many aspects of Go, many different kind of people that play it in many different context. It is true that go is less hierarchical than other games, it doesn't matter that much what rank you are because you can put handicap stones, go players don't mind as much (talking from the go players I know) for ranking as videogames players do. But it is true that Go is hierarchical, it would be foolish to pretend ranks don't exist at all.
By the way I also read that Go was played by higher classes in Japan.
So I wasnt discredited before he put me down lol?
When I meant hierarchy, I meant within the game: like Chess has pawns and queens, but go is just stones. I should have clarifiee. The handicap stones aspect is something i didnt consider thanks!
Where did you read that? Ill read it too, it'l help improve my view
On February 04 2015 09:37 Kupon3ss wrote: The game historically originated in the upper classes, most likely from divination and from warrior nobles The entire Dan system is based upon ranks and hierarchical designations, and titles that have been governing the game for thousands of years
Bottom line is that you don't know much about the history of Go or Chess or even esports and you're just filling in the blanks to fit your hipster conception of the game
You discredit yourself with that kind of pathetic offensive post...
I think there are many aspects of Go, many different kind of people that play it in many different context. It is true that go is less hierarchical than other games, it doesn't matter that much what rank you are because you can put handicap stones, go players don't mind as much (talking from the go players I know) for ranking as videogames players do. But it is true that Go is hierarchical, it would be foolish to pretend ranks don't exist at all.
By the way I also read that Go was played by higher classes in Japan.
So I wasnt discredited before he put me down lol?
When I meant hierarchy, I meant within the game: like Chess has pawns and queens, but go is just stones. I should have clarifiee. The handicap stones aspect is something i didnt consider thanks!
Where did you read that? Ill read it too, it'l help improve my view
Can't remember where I read it. Probably senseis library, it really is the best website for go knowledge in english. I just found this page that could be interesting for you http://senseis.xmp.net/?CompareGoToChess/History I dont know about the writor though.
https://en.wikipedia.org/wiki/Four_Arts check this page. Really a few centuries ago, lower folks were all peasants and peasants didn't have the time for these I think.
PS : you are not discredited, I just meant the other guy is discrediting himself when he uses offensive words in an opinion discussion
On February 04 2015 09:37 Kupon3ss wrote: The game historically originated in the upper classes, most likely from divination and from warrior nobles The entire Dan system is based upon ranks and hierarchical designations, and titles that have been governing the game for thousands of years
Bottom line is that you don't know much about the history of Go or Chess or even esports and you're just filling in the blanks to fit your hipster conception of the game
You discredit yourself with that kind of pathetic offensive post...
I think there are many aspects of Go, many different kind of people that play it in many different context. It is true that go is less hierarchical than other games, it doesn't matter that much what rank you are because you can put handicap stones, go players don't mind as much (talking from the go players I know) for ranking as videogames players do. But it is true that Go is hierarchical, it would be foolish to pretend ranks don't exist at all.
By the way I also read that Go was played by higher classes in Japan.
So I wasnt discredited before he put me down lol?
When I meant hierarchy, I meant within the game: like Chess has pawns and queens, but go is just stones. I should have clarifiee. The handicap stones aspect is something i didnt consider thanks!
Where did you read that? Ill read it too, it'l help improve my view
Can't remember where I read it. Probably senseis library, it really is the best website for go knowledge in english. I just found this page that could be interesting for you http://senseis.xmp.net/?CompareGoToChess/History I dont know about the writor though.
https://en.wikipedia.org/wiki/Four_Arts check this page. Really a few centuries ago, lower folks were all peasants and peasants didn't have the time for these I think.
PS : you are not discredited, I just meant the other guy is discrediting himself when he uses offensive words in an opinion discussion
Oh my mistake, I apologize about the misunderstanding!
That link is honestly a massive godsend. I just wish the documentary would come out as well because I'd love to see how Go is in mainstream culture now!
Thanks a lot for the links! I'm very excited to finish writing this.
http://magazine.1337mag.com/issue1/ -- I've released an issue about Go. I unfortunately couldn't write about everything, so it's been heavily reconstructed. It's on page 66 of our first issue.
Apparently, the program AlphaGo managed to beat the European Champion Fan Hui.
Now, for the first time, a computer has beaten a human Go professional without the advantage of a handicap. AlphaGo, a program developed by Google’s London-based company DeepMind, bested European champion Fan Hui in five games out of five.
Is it normal for the commentators to be lagging behind the game? Wouldn't it make more sense for them to catch up to the game and then start talking about possible next moves?
does anyone know how much processing power the computer running alphago has? IBM's Watson took up an entire room. But I assume they're playing the match in South Korea, which means they would have had to move the bot over from Europe, so there must be an upper bound on the computer's size
Lol, commentating on the fact that its impossible for Sedol to do after-game analysis with his opponent. I fear how game would have turned out if AlphaGo played 100% perfect....
On March 09 2016 16:20 Pseudoku wrote: Is AlphaGo actually going to have enough time?
This was never an issue. After main time runs out it has 6 periods 30 seconds, which means as long as it makes a move in under 30 seconds it does not lose a period. So it basically has 30 seconds for each moves and 5 chances to take a bit more per move. The state the game was in this was entirely acceptable.
On March 09 2016 16:20 Pseudoku wrote: Is AlphaGo actually going to have enough time?
This was never an issue. After main time runs out it has 6 periods 30 seconds, which means as long as it makes a move in under 30 seconds it does not lose a period. So it basically has 30 seconds for each moves and 5 chances to take a bit more per move. The state the game was in this was entirely acceptable.
I've never watched or followed Go before and the game started off slow, but I learned a lot as the commentators went on and by the end it was pretty exciting. I like how they didn't know who was going to win (and seemingly nobody else?) until the very end. Really phenomenal effort by AlphaGo and I'll be paying close attention to tomorrow's match.
Yeah, I joined mid-game but now I see Sedol played some really dubious mistakes throughout the game. I guess AlphaGO being AI made us more focus on its mistakes, but its misplays during the mid-game only gave Sedol a chance to put the game into doubt again rather than being decisive failure.
It'll be interesting to see what happens on Game 2 when Sedol take this AI more respect from very first move.
On March 09 2016 16:57 Hesmyrr wrote: Yeah, I joined mid-game but now I see Sedol played some really dubious mistakes throughout the game. I guess AlphaGO being AI made us more focus on its mistakes, but its misplays during the mid-game only gave Sedol a chance to put the game into doubt again rather than being decisive failure.
It'll be interesting to see what happens on Game 2 when Sedol take this AI more respect from very first move.
Enlighten us about all those dubious moves. Can't wait to have my eyes opened. Because all the profesionals commenting on the various streams only marked one moved as really dubious (Q5). But you obviously have more insight than those 9p scrubs.
Turned late to the stream but is Alpha Go a sort sort of improved Takemiya??? Those moyos... Dont want to go to deep in this but that would make sense as Takemiya's strength is fighting and an AI strength must be in fighting too imo. Cant wait for an anlysis of this game.
On March 09 2016 16:57 Hesmyrr wrote: Yeah, I joined mid-game but now I see Sedol played some really dubious mistakes throughout the game. I guess AlphaGO being AI made us more focus on its mistakes, but its misplays during the mid-game only gave Sedol a chance to put the game into doubt again rather than being decisive failure.
It'll be interesting to see what happens on Game 2 when Sedol take this AI more respect from very first move.
Enlighten us about all those dubious moves. Can't wait to have my eyes opened. Because all the profesionals commenting on the various streams only marked one moved as really dubious (Q5). But you obviously have more insight than those 9p scrubs.
As I said before, I'm only beginner so definitely does not feel justified to do this epic game prosperity! I am only ahead in my ability to understand Korean and abbreviate their commentary for LR purposes.
However, I am pretty certain that other baduk communities will be able to point out all the weak points in this game which could've been improved. Since this is pretty significant match, lot of professional players will definitely analyze this match inside and out, and discover all the critical moves which influenced outcome of the game
Big picture move from AlphaGo which grabs top of the board rather than reinforcing the bottom group forces first time-intensive thinking from Lee Sedol.
Korean commentators talk about P10-Q12 variation and how previous success of AlphaGo forced Sedol to analyze this decision more. Has AlphaGo already foreseen a future where it'll sacrifice the right for solid middle?
On March 10 2016 14:09 Hesmyrr wrote: Can anyone more knowledgeable about the game commentate on whether Sedol's choice of block direction is correct?
Now I (4d) admit, that my first intuition was also to block off the side. But upon comparing the results of both blocks, I'm pretty convinced blocking upwards as played is correct. The main issue is the top side here. At the point it was played, it was really easy for B to press down on the left side of W. This means getting access from there is difficult for W. If W would have now also been pressed down (even if down here would mean 4th line, which is generally considered good) on the right side, any invasion on top would be a real difficulty and potentially get under severe attack. Remember that "under attack" does not mean it would die. But the profit for Black from attacking would outweight the potential gain, White could have taken earlier on the side.
Well... now that I wrote it, the players were so nice to actually show it on the board... Imagine the white stone peaking into the center from the right side would be Black, and in return White would have the entire right side cleared. The gain on the right side would be about 15-20 points. But the Black framework would be really impressive and threatening.
On March 10 2016 15:05 Furikawari wrote: Lee has too much secured territory imo, I don't see alphago winning this. B almost has nothing secured right now.
Top is now split and open on both sides, middle is open on 2 sides (and theres a cut on left even though a crude one), bottom right is open (even though it would cost in the corner obviously).
Black has no secured point right now, only prospects.
On March 10 2016 17:17 calh wrote: Fantasy gg timing?
When in Byoyomi, you will see much less resignations then when having lots of time left like yesterday. But of course, when you have a dozen pros chewing it for the couch spectator, it is easy to call everything and make bold judgements.
On March 10 2016 17:35 Draconicfire wrote: I think five games regardless. The games against Fan Hui were 0:5 so I'm assuming it will be the same case here.
Yeah. We were silly to think the difference in skill between Fan Hui Lee Sedol would actually matter. AlphaGo destroying them both.
On March 10 2016 17:35 Draconicfire wrote: I think five games regardless. The games against Fan Hui were 0:5 so I'm assuming it will be the same case here.
Yeah. We were silly to think the difference in skill between Fan Hui Lee Sedol would actually matter. AlphaGo destroying them both.
? In both games Lee made questionnable moves (to say the least). It's nowhere close to "destroying".
On March 10 2016 17:35 Draconicfire wrote: I think five games regardless. The games against Fan Hui were 0:5 so I'm assuming it will be the same case here.
Yeah. We were silly to think the difference in skill between Fan Hui Lee Sedol would actually matter. AlphaGo destroying them both.
? In both games Lee made questionnable moves (to say the least). It's nowhere close to "destroying".
I just meant if it does go 5-0 for AlphaGo. I'm pretty sure 5-0ing someone would be considered "destroying" even if the games are close.
I agree the final result is what matters the most. Lee Changho did not go for fancy wins with massive pt difference, but nobody can deny he "destroyed" in his highest point because he won things.
Also, probably switching to BadukTV commentary. They have been surprisingly accurate with predictions while SBS and KBS faltered.
It's undeniable now that AlphaGo is really really good, which is made worse by the fact that it seemingly plays "incomprehensible" moves. I fear what effect this might have on the Go scene.
Many of those still seem to play, so I guess it give you some info on who is the best. Also this has some info descriptions on the best players: https://en.wikipedia.org/wiki/List_of_Go_players
This was produced by Dr. Bae Taeil using repeat calculation(I am not sure what it means.). It is not official ranking but Korean Baduk association use this table to produce their Korean Ranking.
I just keep track of who holds the titles in their respective country, this site is incredibly good with all the tournament results of previous years and kifus for most games: http://igokisen.web.fc2.com/news.html
This was produced by Dr. Bae Taeil using repeat calculation(I am not sure what it means.). It is not official ranking but Korean Baduk association use this table to produce their Korean Ranking.
I just keep track of who holds the titles in their respective country, this site is incredibly good with all the tournament results of previous years and kifus for most games: http://igokisen.web.fc2.com/news.html
I know very little of the game, but I'm a huge fan of AI and how Google has applied neuronetworks in this particular challenge. This is akin to finally putting a "brain" behind a super computer.
I'm posting live impressions here b/c other thread seems more suited for actual discussion. Quite fascinating I3 from AlphaGo, should be fun to see how this battle turns out.
I'm actually looking forward to game4 and 5. The last game was really painfull, yeah. But I hope Lee Sedol will be under less pressure now. Everything he can get from here on out is a bonus. So I hope to see something along the lines of what game 2 offered. Just some games where he will be able to compete to begin with. And then maybe just does not slip up in the late middlegame/early endgame.
Looking forward to the last two matches as well as the press conference on tuesday. ^_^
On March 12 2016 16:38 usopsama wrote: This is too painful to watch. I won't be watching game 4 and 5.
Same, I got the point after game 3
I look at it like this...what's makes a game the best it can possibly be? I think it's when it's so neck and neck nobody can figure out who's going to win until it finally happens. The only drawback here is that we'll already know the winner -- the computer -- but isn't the journey of getting to the win just as amazing? The game is so conservative that it will do anything to keep a higher win %...behind the scenes what that looks like to us is the closest possible game (and arguably the "best")?
I guess what I'm saying is, in essence the AI will tend to dumb itself down just enough to be just the slightest bit better than whichever human faces it. It'd be like Flash vs. JD, both in their primes, every single game. I think by definition there couldn't even BE a "bad" game.
EDIT: Maybe that even means that to be a professional go player still means fighting your hardest to play your best. The only difference is, your dedication to playing an amazing game gives fans an opportunity to see a strategy even better than your best.
Really interesting game. AlphaGo will learn not to play itself into a corner on two fateful moves next time. It was also interesting to see it "regress" to more Hail Mary moves, like when the large rightmost group got taken and it effectively gave away a stone for no reason. Congratulations to Lee Sedol!
For anyone who was curious, the message that AlphaGo displays on the screen when it resigns is, "AlphaGo resigns" and "The result 'W+resign' was added to the game information."
For anyone who was curious, the message that AlphaGo displays on the screen when it resigns is, "AlphaGo resigns" and "The result 'Weresign' was added to the game information."
Minor correction: "The result 'W + Resign' was added to the game information."
Even though it started as a 3-0, if this series ends up 3-2 it certainly means alphago needs more practice before it can be considered best in the world, right? I'm going to bet lee sedong has lost best of fives before
if lee sedol wins the next game i would think that if you turned this into a best of 7 lee sedol would comeback to win 3-4. successive wins from lee sedol could mean he noticed some patterns within the program
On March 14 2016 11:51 evilfatsh1t wrote: if lee sedol wins the next game i would think that if you turned this into a best of 7 lee sedol would comeback to win 3-4. successive wins from lee sedol could mean he noticed some patterns within the program
All that would mean is AlphaGo would go back into its Hyperbolic time chamber for a little longer to improve some more.
A true comeback game. The commentators pretty much wrote Lee off until the "Hand of God" White 78 move. I actually stopped watching because I thought it was a sure defeat.
I'm glad he got one. No question that going up against 1200 CPUs is no small feat. Hopefully he can pull out some more God moves in the next one.
On March 14 2016 11:51 evilfatsh1t wrote: if lee sedol wins the next game i would think that if you turned this into a best of 7 lee sedol would comeback to win 3-4. successive wins from lee sedol could mean he noticed some patterns within the program
I think it's likely for Lee Sedol to win game 5 as well. The match is inherently imbalanced as the commentators once said. Alphago has been trained on existing matches, including those from Lee Sedol. But this version of Alphago has never been shown before.
On March 14 2016 11:51 evilfatsh1t wrote: if lee sedol wins the next game i would think that if you turned this into a best of 7 lee sedol would comeback to win 3-4. successive wins from lee sedol could mean he noticed some patterns within the program
I think it's likely for Lee Sedol to win game 5 as well. The match is inherently imbalanced as the commentators once said. Alphago has been trained on existing matches, including those from Lee Sedol. But this version of Alphago has never been shown before.
They mentioned in the press conference that all of the games that AlphaGo used to train were amateur matches pulled from online Go sites. No professional games, by Lee Sedol or otherwise, were included in its learning database.
On March 14 2016 11:51 evilfatsh1t wrote: if lee sedol wins the next game i would think that if you turned this into a best of 7 lee sedol would comeback to win 3-4. successive wins from lee sedol could mean he noticed some patterns within the program
I think it's likely for Lee Sedol to win game 5 as well. The match is inherently imbalanced as the commentators once said. Alphago has been trained on existing matches, including those from Lee Sedol. But this version of Alphago has never been shown before.
They mentioned in the press conference that all of the games that AlphaGo used to train were amateur matches pulled from online Go sites. No professional games, by Lee Sedol or otherwise, were included in its learning database.
I was under the impression that the network was trained with ancient traditional games, online games as well as professional games. But I don't know for sure.
From the abstract of the Nature article,
These deep neural networks are trained by a novel combination of supervised learning from human expert games, and reinforcement learning from games of self-play.
I would be interested to find out which games it has used to train on. If it really didn't use any 9-dan games then that's really impressive to reach the level that it has.
On March 14 2016 11:51 evilfatsh1t wrote: if lee sedol wins the next game i would think that if you turned this into a best of 7 lee sedol would comeback to win 3-4. successive wins from lee sedol could mean he noticed some patterns within the program
I think it's likely for Lee Sedol to win game 5 as well. The match is inherently imbalanced as the commentators once said. Alphago has been trained on existing matches, including those from Lee Sedol. But this version of Alphago has never been shown before.
They mentioned in the press conference that all of the games that AlphaGo used to train were amateur matches pulled from online Go sites. No professional games, by Lee Sedol or otherwise, were included in its learning database.
I was under the impression that the network was trained with ancient traditional games, online games as well as professional games. But I don't know for sure.
These deep neural networks are trained by a novel combination of supervised learning from human expert games, and reinforcement learning from games of self-play.
I would be interested to find out which games it has used to train on. If it really didn't use any 9-dan games then that's really impressive to reach the level that it has.
They used a very large database of KGS 4d-9d games. Can I now say that I was part of Alphagos development, because some of my games were in that database? :D
On March 14 2016 11:51 evilfatsh1t wrote: if lee sedol wins the next game i would think that if you turned this into a best of 7 lee sedol would comeback to win 3-4. successive wins from lee sedol could mean he noticed some patterns within the program
I think it's likely for Lee Sedol to win game 5 as well. The match is inherently imbalanced as the commentators once said. Alphago has been trained on existing matches, including those from Lee Sedol. But this version of Alphago has never been shown before.
They mentioned in the press conference that all of the games that AlphaGo used to train were amateur matches pulled from online Go sites. No professional games, by Lee Sedol or otherwise, were included in its learning database.
can anyone confirm this? it would be absolutely ridiculous if something (human or bot) could get good enough to beat lee sedol without learning off any high level games. that idea is just ludicrous for any game
On March 14 2016 11:51 evilfatsh1t wrote: if lee sedol wins the next game i would think that if you turned this into a best of 7 lee sedol would comeback to win 3-4. successive wins from lee sedol could mean he noticed some patterns within the program
I think it's likely for Lee Sedol to win game 5 as well. The match is inherently imbalanced as the commentators once said. Alphago has been trained on existing matches, including those from Lee Sedol. But this version of Alphago has never been shown before.
They mentioned in the press conference that all of the games that AlphaGo used to train were amateur matches pulled from online Go sites. No professional games, by Lee Sedol or otherwise, were included in its learning database.
can anyone confirm this? it would be absolutely ridiculous if something (human or bot) could get good enough to beat lee sedol without learning off any high level games. that idea is just ludicrous for any game
yes, it was said explicitly in the press conference after game 4 by hassabis. Just watch in on youtube, it is during the question block for Lee Sedol, but Hassabis interfers.
On March 15 2016 13:52 Draconicfire wrote: Yea, they mentioned in the post-game interview (game 4) that they only used amateur games from the Internet, Lee Sedol's games were not used.
Ok, cool.
Also I'm wondering if AlphaGo would play the exact same moves if a player used the same moves in two different games.
The policy and value networks should be the exact same results. But the monte carlo search tree is inherently random. There is always a chance that it randomly finds itself a better move at some point.
On March 14 2016 11:51 evilfatsh1t wrote: if lee sedol wins the next game i would think that if you turned this into a best of 7 lee sedol would comeback to win 3-4. successive wins from lee sedol could mean he noticed some patterns within the program
I think it's likely for Lee Sedol to win game 5 as well. The match is inherently imbalanced as the commentators once said. Alphago has been trained on existing matches, including those from Lee Sedol. But this version of Alphago has never been shown before.
They mentioned in the press conference that all of the games that AlphaGo used to train were amateur matches pulled from online Go sites. No professional games, by Lee Sedol or otherwise, were included in its learning database.
can anyone confirm this? it would be absolutely ridiculous if something (human or bot) could get good enough to beat lee sedol without learning off any high level games. that idea is just ludicrous for any game
"AlphaGo was not trained specifically to Lee Sedol's play. We train it in a general way, and in fact, the games that we used--the human games that we used to start the training were actually strong amateur [therefore, non-professional] games from Internet Go servers. So in fact, there are no games of Lee Sedol in our database, training database. And then as you know already, the way AlphaGo got stronger was to play itself. So in fact, I think it's quite equal in terms of the information, we didn't train it on Lee Sedol's games."
On March 14 2016 11:51 evilfatsh1t wrote: if lee sedol wins the next game i would think that if you turned this into a best of 7 lee sedol would comeback to win 3-4. successive wins from lee sedol could mean he noticed some patterns within the program
I think it's likely for Lee Sedol to win game 5 as well. The match is inherently imbalanced as the commentators once said. Alphago has been trained on existing matches, including those from Lee Sedol. But this version of Alphago has never been shown before.
They mentioned in the press conference that all of the games that AlphaGo used to train were amateur matches pulled from online Go sites. No professional games, by Lee Sedol or otherwise, were included in its learning database.
can anyone confirm this? it would be absolutely ridiculous if something (human or bot) could get good enough to beat lee sedol without learning off any high level games. that idea is just ludicrous for any game
Considering that AlphaGo has played millions of games against itself and is able to meaningfully learn from each one, I don't think it's too ridiculous at all. I would be surprised if the total number of 9 dan level professional go games played in human history has reached a million.
On March 14 2016 11:51 evilfatsh1t wrote: if lee sedol wins the next game i would think that if you turned this into a best of 7 lee sedol would comeback to win 3-4. successive wins from lee sedol could mean he noticed some patterns within the program
I think it's likely for Lee Sedol to win game 5 as well. The match is inherently imbalanced as the commentators once said. Alphago has been trained on existing matches, including those from Lee Sedol. But this version of Alphago has never been shown before.
They mentioned in the press conference that all of the games that AlphaGo used to train were amateur matches pulled from online Go sites. No professional games, by Lee Sedol or otherwise, were included in its learning database.
can anyone confirm this? it would be absolutely ridiculous if something (human or bot) could get good enough to beat lee sedol without learning off any high level games. that idea is just ludicrous for any game
Considering that AlphaGo has played millions of games against itself and is able to meaningfully learn from each one, I don't think it's too ridiculous at all. I would be surprised if the total number of 9 dan level professional go games played in human history has reached a million.
I think it's surprising, too. Almost ridiculous in what this program has achieved. The developers mentioned in the game 5 pregame talks about the next level of machine learning. This is about building an AI with no human training. This seems far beyond what could be achieved right now.
The fact that Alphago is beating 9-dans while being trained from high amateurs... this seems to be a step in that direction already. How soon before no training is needed?
I would think it was already able to get to that level by itself, but having a seed of initial games to work with just cut down on some computing time. Parsing a specific game format to get pro games wasn't a goal of the DeepMind project, as they mentioned they just pulled games from an internet cafe program.
But given that they created a framework for meaningful learning with each new game, I would give odds on the computer to be more advanced than humanity once the games it is able to play starts pushing past the millions. A professional may play a couple thousand in a lifetime.
I just wanna say that I've learned a ton about Go from these past 5 games through all the English commentary. Really fascinating game. Props to the AlphaGo team and Lee Sedol for getting me interested!
On March 14 2016 11:51 evilfatsh1t wrote: if lee sedol wins the next game i would think that if you turned this into a best of 7 lee sedol would comeback to win 3-4. successive wins from lee sedol could mean he noticed some patterns within the program
I think it's likely for Lee Sedol to win game 5 as well. The match is inherently imbalanced as the commentators once said. Alphago has been trained on existing matches, including those from Lee Sedol. But this version of Alphago has never been shown before.
They mentioned in the press conference that all of the games that AlphaGo used to train were amateur matches pulled from online Go sites. No professional games, by Lee Sedol or otherwise, were included in its learning database.
can anyone confirm this? it would be absolutely ridiculous if something (human or bot) could get good enough to beat lee sedol without learning off any high level games. that idea is just ludicrous for any game
Considering that AlphaGo has played millions of games against itself and is able to meaningfully learn from each one, I don't think it's too ridiculous at all. I would be surprised if the total number of 9 dan level professional go games played in human history has reached a million.
I think it's surprising, too. Almost ridiculous in what this program has achieved. The developers mentioned in the game 5 pregame talks about the next level of machine learning. This is about building an AI with no human training. This seems far beyond what could be achieved right now.
The fact that Alphago is beating 9-dans while being trained from high amateurs... this seems to be a step in that direction already. How soon before no training is needed?
When you got people like elon musk saying that this basically moved AI research up by about 10 years.. that pretty much indicates how insane a jump this development has been.
Well none of the AI techniques they used is new. Neuro-networks have been around for quite some years now, they just found a way to make it work well in this particular case.
On March 16 2016 22:40 Glacierz wrote: Well none of the AI techniques they used is new. Neuro-networks have been around for quite some years now, they just found a way to make it work well in this particular case.
They have stated numerous times that the framework is going to be much more flexible than just this case and can be used in tons of other ways than just Go
On March 16 2016 22:40 Glacierz wrote: Well none of the AI techniques they used is new. Neuro-networks have been around for quite some years now, they just found a way to make it work well in this particular case.
Alphago is based on a project deepmind made that, as a single unchanging program, could learn to play and master the entire library of atari games using only the raw pixels as inputs.
I have a feeling some of the AI techniques they're using are new...
What that have that's new is a way of mixing policy and value networks with a Monte Carlo search tree. So yeah, they "just found a way" but it's novel.
Other Go software have used neural networks before. But the value network for evaluating board positions is very difficult. Even human players will not be able to tell the value of a given board state, or which player is really ahead.
Also using neural networks doesn't mean anything unless they are trained. The training methods they used, as well as the deep layers of abstraction in the network, result in very low errors when predicting moves. These networks are similar to what was developed for deep dream and other image recognition ai's (see post above) and I think it's pretty cool.
Edit: should add that they threw a boatload of hardware at this problem, too. 1,202 CPUs and 176 GPUs with 40 search threads in the distributed version.
On March 17 2016 09:41 meatpudding wrote: should add that they threw a boatload of hardware at this problem, too. 1,202 CPUs and 176 GPUs with 40 search threads in the distributed version.
Makes me wonder what we could make with more hardware, that's pretty chump change as far as computing power is concerned.
On March 17 2016 09:41 meatpudding wrote: should add that they threw a boatload of hardware at this problem, too. 1,202 CPUs and 176 GPUs with 40 search threads in the distributed version.
Makes me wonder what we could make with more hardware, that's pretty chump change as far as computing power is concerned.
Deepind stated that the performance got worse when using more hardware. Which is not that uncommon.
There is a graph in the nature paper they released in january that shows how the strength scales with hardware. Basically it scales a lot from 1 to 2 GPUs, but then looks like it growth logarithmically at best. : ]
Looking forward to future announcements. They kind of hinted that they might work on a single-machine version, so we could actually get "our own" AlphaGo at some point. :D
Have people been playing go recently? I started getting back into it (having nothing to do with the ai defeating the top pro). What struck me is how, after several years off, I started off right back at the strength I was at when I stopped playing (~13-14 kyu). After playing some matches I seem to be appropriately ranked at 12 kyu on KGS. I'm still weak enough that there is a ton of variation in the results of my matches... I can win by 100 moku one game and lose by nearly that much in another, even though the handicap system should minimize that (assuming the difference in ranks isn't too big).
I think I'm starting to get good enough now where it is actually worthwhile for me to study some of the more common joseki, but I'm not 100% sure I will do it... we'll see. I have been doing some life or death problems which I know, in theory, is very helpful. For whatever reason I seem to be much better at solving the problems when they provided to me in isolation than I am at applying what I learned into real games. I suppose playing more games will help with that.
I'm also forcing myself to be a bit more flexible with time controls. I used to only play long games because I didn't like to feel rushed. This time around I'm doing somewhat shorter games, and while it's frustrating at times, I'm also seeing it help me against slower opponents, so I can't really complain. For perspective, I used to only play 30 minutes + 5 x 30 seconds, but now I'm also doing 20 minutes + 5 x 20 seconds (for many players I'm sure that's considered slow as well).
The funniest thing is how I'm slowly marching my way towards actually being a good player, and even when I reach low single-digit-kyu I still will not know how to play go on a real-life board or be able to score without the computer helping me
You don't really get weaker if you stop playing, just the reading gets worse over time and at kyu level you will rarely see people read more than 1 variation or 4 moves ahead.
For joseki there are 2 proverbs that go together: memorize joseki to get weaker and study joseki to get stronger. If you just memorize the moves, you will be confused when people stray from joseki. If you know why move x is played in the sequence, you can properly benefit from your opponent's mistake or decide to go out of joseki yourself because it is actually benefical for the whole board situation.
Time settings should just be what you're comfortable with, imo. I know people who play exclusively 10s blitz games, but their fundamentals are just awful, because they rely on their opponent not finding the proper refutation for their overplays in time. That's not true for everyone of course, but you see it a lot in blitz games. It's also a matter of practice, when my reading was best I could read several variations out in 10s, so a blitz player is not automatically a sloppy player.
I think learning how to count in a game is more important than knowing how to move stones on a real board to make scoring easier, but that probably depends on whether or not you play a lot of real board games. Counting definitely isn't very important until you reach dan level.
I don't play very often on KGS or Tygem anymore, but I always have about 5 simultanous games on DGS and watch a lot of professional games. I also spend a significant amount of my go time on creating and fixing problems for goproblems.com
Thanks for the insights. Actually, goproblems.com is what I use.
How do you study joseki? Obviously I can look up joseki and read the little tidbits associated, but I think what you are suggesting is more involved than that. Keep in mind I'm probably not at a point where I'm really going to go that far out of my way right now, but it would be good to know.
You just go from the simple to the complex. For example the common 3-4 joseki with high approach
What is the purpose of each move? 1 is a standard move to make territory in the corner, because if Black gets to play again, he can make an enclosure at 3, 4, 2 or 6. 2 is a standard approaching move with emphasis on influence (4th line). 3 attaches to a single stone, that means both sides will get stronger locally, it also starts the sequence that gives Black solid corner territory. 4 is touching back and as a general rule you should always respond to an attachment directly. 4 stops Black from further extending along the top and reduces the liberties of 3, leading to 5 draws back to give Black a solid shape, which also leaves a cutting point at 6, so White defends. After 6 the Black shape has a weakpoint, so he defends that. 8 makes a base for the White stones, the position is now about equal. Black gained territory and White got a stable group to limit Black's expansion from the corner.
If you are aware of the purpose of every move, you know why it is ok/not ok to omit it. For example if Black tenuki at 7, White can enclose Black and gain a lot of influence, because Black didn't defend the weakpoint in his shape.
This may actually be good for Black, if playing on top is super important and White won't be able to use the influence he gains here.
Is there a place where I can go to read more analysis like this? That is really helpful but I don't expect you to write a page or so on every common joseki lol.
Indeed, lol. I bit off a lot to chew there^^ However, I really don't mind helping out or reviewing games for TL users.
A good starting point are joseki dictionaries, they contain these explanations and some continuations. josekipedia.com has a lot of variations, you can also watch lectures by stronger players who explain these things. I learned a lot from gocommentary.com, a mistake in this particular joseki is explained in this video http://www.gocommentary.com/free-videos/how-to-punish-overplay-03.html My teacher reviewed a lot of Shusaku games (another go proverb is "study the shusaku fuseki") and explained the more basic concepts for kyu players https://www.youtube.com/playlist?list=PLS1Sn1wo0vB3da6Mlnpfl1fyDJUeypt09
I'm sure most people here know this already, but the "Master" account which terrorized Tygem and FoxGO was revealed to be a new version of AlphaGo. Considering the rise of other Go programms (Zen, various chinese programs and the mysterious "GodMoves"), the future of Go seems to take an interesting turn.
I really wish more pros would release some analysis on the Master games. I'd super want to get some insight on this stuff but it's so hard to come by english content. I know a couple of lower dan pros have done commentaries and that's some great insight but I was hoping for some top pro input.
The program used less than 1 minute per move in the beginning, and end game plays were just a few seconds. It feels so much quicker than the Lee Sedol match, pretty scary how much it's improved.
Ke Jie stuck with a game plan of taking early territory, but ended up losing quite a bit on the top corner from a white reduction move. He then tried to reduce the center, but it wasn't enough for him to win the game.
As expected, AlphaGo made passive plays that simplified the board state and played loose endgame moves that didn't really maximize the points it wins by. The game isn't as close as the final score suggests.
One of the deepmind team member said alphago gives a very slight advantage to white under the 7.5 komi rule, but did not elaborate on the exact percentages. Ke Jie is also known to be much stronger on white.
Also if you look at the 10 games that were just released of AlphaGo playing against itself (and 40 more to come in the next 4 days) you can see that 8 games were won by white. I suspect that trent will continue in all 50 games.
The claim of "3 stones stronger" doesn't seem like such a far off statement now (even though it's more like 1.5 stones maybe? who knows)
Every move it makes is ultra flexible and has multi purpose meaning waaaay beyond any human can understand. Funny we thought it would be a monster at reading, but in fact it's a monster at positional play and the ability to understand the tradeoffs and full potential of every move
I wouldn't be surprised if people start to play like it within a year or so. It's a shame that they are retiring the program. I hope someday it will be made open source. The computational requirement isn't super high, definitely possible to get an unbeatable version going on a home PC a few years down the road.
I have finally overcome my "i'm going to lose a lot" anxiety (after around 8 years of being interested in the game overall..) and started playing on ogs recently.
having a lot of fun so far, shame i didn't start earlier
I'm playing on OGS, https://online-go.com/player/1169028/, ranked 4kyu playing against bots. It might be fun to play against people or posters from TL. I don't see an easy button to challenge, might have to make a game and invite.