It 1 rax Fe without gaz.
It's seems better to double orbital around 3:00.
Perhaps the change of 6 SCV to 12 SCV change something.
I still work on my hypothesis.
Forum Index > StarCraft 2 Strategy |
Diamadoshikiller
France56 Posts
It 1 rax Fe without gaz. It's seems better to double orbital around 3:00. Perhaps the change of 6 SCV to 12 SCV change something. I still work on my hypothesis. | ||
Ryu3600
Canada468 Posts
| ||
Diamadoshikiller
France56 Posts
In reality i don't want to prove or show, i want to see and have discussion with the community about a crazy idea. My question is what is the perfect timing to orbital in LOTV? The modification of 6 SCV to 12 change something or not? (That can affect all build of terran) | ||
yubo56
658 Posts
And what does n mean, in the title of your graph? Your spike at 3 minutes from the late orbital is probably from when the first MULE drops. The SC2 "current mineral income" has some natural variability, and it will generally overshoot when there's a change in income, that's probably what the spike is coming from (P.S. As a fellow francophone, it's "gas" in English, not "gaz" ) | ||
Diamadoshikiller
France56 Posts
n= number of test Sorry for my english | ||
Ryu3600
Canada468 Posts
| ||
yubo56
658 Posts
If you transfer a worker, that means he loses N seconds of mining time due to travel time. The advantage of pulling more workers than 16 is so that later workers built don't have to lose those N seconds of mining time. But if both main and nat are regular mineral patches, by pulling workers such that the main has 16 - K workers, you're giving up K * N seconds of mining time so you save K * N seconds later. It's the same mineral count, you just get it later if you transfer too many This changes when you take a gold base natural, of course, so that's an interesting case to consider, but I think you should still be able to just do it mathematically rather than running trials (with trials, you have some noise introduced by how Blizzard measures mining rate + noise based on exact spread of workers. It's a closer imitation of a real game, sure, so you have an idea of "if I play a game, how likely am I to break even w/ this many workers vs that many" but the distribution should have the same average as the theoretical value.) | ||
Gladness
United States59 Posts
If there is any reason to delay the morph of the orbital, it is to spend its 150 mineral cost somewhere else immediately. The CC-first build from WoL/HotS did this because the earlier cc could quickly recover the cost of one delayed orbital -- mostly by accelerating the second. But delaying the 1st orbital in a gasless FE doesn't give you a cc or a 2nd orbital any faster. If you're fine tuning a specific proxy play that needs 150 minerals to drop an extra rax at a specific time, you should probably include that build order. If the question is whether Terran openings in general should do this, the answer is still no, as your own data show. | ||
Diamadoshikiller
France56 Posts
| ||
WeddingEpisode
United States355 Posts
I wondered this too - that is, how important Orbital is right away. | ||
Ryu3600
Canada468 Posts
| ||
Diamadoshikiller
France56 Posts
The only one build can use the delay of orbital it's a one rax Fe without gas or CC first. I try...but i fail | ||
yubo56
658 Posts
| ||
Diamadoshikiller
France56 Posts
You right cumul it's better for undertand. Without gaz the 2 build it's equivalent until 3:45. After it's just a tendancy orbital case it's better for the 30 last second. At the end (4:00) the mean difference it's just 264 mineral. For the 2 orbital you spend 300 mineral... That say clearly orbital at 4:00 not start to be effective!!! when you play without gas. With gas the build it's equivalent until 3:00. After 3:45 you have a real difference for orbital. At the end (4:00) the mean difference it's 342 mineral. When you take gas before second CC you need start with orbital!!!! But at the start MULE of orbital will not give you a real advance of mineral. It's more latter. But more later will you be alive. I will do the same with CC first just to see. For protoss and zerg what is the result? | ||
yubo56
658 Posts
The reason no orbital does within 300 minerals at 4 minutes is because you're not using the same end state for both build orders. Since even if you delay the orbital you still have to get it eventually, the correct end state would be to pick a state where both builds have completed orbitals (and maybe have dropped at least one set of mules). It's not fair to say "you only earn 264 minerals but spent 300 minerals" because in a normal game those 300 minerals are minerals you *have to spend* eventually anyways. So since both builds pay the 300 mineral penalty at some point, there's a strict 264 mineral profit for the orbital case. In your present analysis, "no orbital" should fall more than 300 minerals behind once you account for the lost SCV production time to change into an orbital, I imagine (not 100% sure) You should clearly describe the build orders you're using in the two cases, especially if you discover a particular build order that is more efficient to delay the orbitals, that way other players can use it! | ||
yubo56
658 Posts
Let T be the amount of time you delay your orbital compared to the earliest possible orbital. Call the time the first orbital finishes time "t = 0". During this time, you will build two extra SCVs then will mine with two more SCVS than the orbital case. During this T, there are three phases: before your extra SCV completes, when you have one extra SCV, and when you have two extra SCVs. The first two phases are each 12 seconds long (SCV production time). During the first phase, you mine the exact same as the orbital case, during the second phase you mine 12 extra minerals (12 seconds * 1 extra SCV) and in the last phase you mine (T - 24) * 2 minerals. During this same T, the orbital player will be mining with two fewer SCVs but will get a MULE at t=25s. From there forwards, he mines with *four* extra SCVs, so two more SCVs than you have. Thus, at the start of phase three, you're actually *losing* (T - 25) * 2 minerals, since you're mining at a 2 worker deficit compared to the MULE. So you earned 12 extra minerals during the first two phases but are losing minerals at 2 minerals/second during the third phase. Of course, there should be a fourth phase: when the player who delayed the orbital spends 25 seconds getting the orbital. During this time, the player who got the earlier orbital recuperates the 12 mineral difference, by the same logic as happens during phases one and two. So the minerals gained during phase 3 are a strict profit. The analysis terminates after phase four since the two build orders have converged: the same number of SCVs, two orbitals and two MULEs. So the only income difference between the two paths depends on what happens during these four phases. Summing our contributions we find a net income difference of 2 minerals/second * T. Saturation is irrelevant in our problem because we have two bases and fewer than 32 SCVs at all times (unless you pick a really big T, a really long delay), so every SCV mines optimally. For larger T, you run into saturation before the orbital player, who has fewer SCVs but has MULEs which can mine on the same mineral patch as SCVs., so your SCV advantage means even less than I've assumed above Notes: - The above analysis holds for multiple orbitals by just multiplying all of the numbers through by the number of orbitals. - You pay upfront for 150 minerals/orbital. I have ignored this cost assuming both builds eventually pay this cost and there's no advantage to having the extra minerals earlier. Of course, there are builds that leverage this extra mineral income, e.g. for additional barracks or maybe another CC. This is a case where your mineral value is time-dependent and should be analyzed on a case by case basis. - Your plot agrees with my model: the 2 orbitals seem to come online around 1:30, and by the end 3:45 you claim we've mined 264 extra minerals. My analysis suggested the orbital build has an advantage of 2 minerals / second - 12 minerals = 270 - 12 = 258 minerals. | ||
Diamadoshikiller
France56 Posts
For CC first it's possible to delay and orbital at 4:00. I think finaly i will try to play CC first with delay of orbital!!! | ||
Diamadoshikiller
France56 Posts
I will answer tomorrow. | ||
Diamadoshikiller
France56 Posts
My goal is not to have a maximum of mineral after 6 minutes . It's to have a maximum of minerals available from 0 to 4:00 minutes that will be used to produce a powerful attack between 4:15 and 4:45 (starting from my base) with stim timing between 5:00 and 5:30. With this mineral I will try to speed up my technology (if possible). But for that I did not have to loose too much of mineral at the beginning of the game. If my attack is powerful; I win or I have done enough damage that will compensate for my loss of mineral. This attack was not possible before. So the other races will have to adapt and certainly invest in equivalent cost. This will then create a new balance. Of course even if I have this push still need to run it well and do the damage (so it will depend on the level of the two players). In addition i will have to survive to all attacks before the push and adapt. | ||
Diamadoshikiller
France56 Posts
when we do the first orbital we spent 150 mineral but: - we not produce 2 SCV (that say we not spent 100 mineral) - we delay a supply i try ... i fail. | ||
| ||
Replay Cast
SW Korea Closed Qualifier Day1
Maru vs StatsLIVE!
Cure vs Classic
Solar vs GuMiho
ByuN vs herO
[ Submit Event ] |
StarCraft 2 StarCraft: Brood War Bisu 3894 Dota 2Mini 1395 ggaemo 619 Hyuk 305 ZerO 302 Soulkey 277 Snow 264 Rush 242 hero 223 Stork 209 [ Show more ] firebathero 155 Tyson 101 Hyun 89 Sharp 88 Mind 67 ToSsGirL 57 ZerO(Twitch) 55 Shinee 49 Backho 44 sSak 36 Mong 30 ajuk12(nOOB) 24 Dandy 14 Sexy 13 ivOry 6 Counter-Strike Other Games hiko1972 DeMusliM1119 crisheroes682 Mew2King521 Hui .518 KnowMe465 Lowko389 ArmadaUGS201 QueenE115 NuckleDu102 Liquid`VortiX98 Trikslyr83 Organizations Counter-Strike StarCraft 2 Other Games StarCraft: Brood War StarCraft 2 StarCraft: Brood War
StarCraft 2 • Adnapsc2 13 StarCraft: Brood War• Gussbus • LaughNgamez Trovo • Poblha • aXEnki • Migwel • intothetv • Laughngamez YouTube • Kozan • IndyKCrew Dota 2 League of Legends |
Big Brain Bouts
BSL
TerrOr vs XuanXuan
Dark vs JDConan
Korean StarCraft League
StarsWar
Maru vs Classic
GuMiho vs ByuN
Stats vs Cure
Solar vs herO
WardiTV Invitational
CSO Cup
ForJumy Cup
BSL
Zhanhun vs WolFix
Dienmax vs Cross
Sparkling Tuna Cup
StarsWar
[ Show More ] WardiTV Invitational
Online Event
ESL Open Cup
Afreeca Starleague
StarsWar
ESL Open Cup
ESL Open Cup
Afreeca Starleague
StarsWar
Club NV x Duckling Show…
GSL Code S
Stats vs SHIN
Cure vs GuMiho
ESL Pro Tour
|
|