|
On February 08 2013 16:34 SiskosGoatee wrote: It's not a point, it's a logical conclusion. If you think C does not follow from A and B you need a crashcourse in logic. I call it a point because there is not an acceptable transition between the three. The first point is practical, as I've said. The second point is an observation, whose correctness I concede. Point C disregards, completely, the mapmaker himself, and implies that his design choices are all rigidly governed by the rectangular border, which is wrong. See Daybreak, see Ohana, for 2 decent examples.
On February 08 2013 16:06 SiskosGoatee wrote: Like I said, if this was actually all his plan then it would be cosmological coincidence that it falls so neatly into a rectangle. I doubt the statistical probability of that. The map was designed to fit into the canvas. And you already admitted to the obviousness of that so you already admit that it wasn't his plan to make the shape like that but he had to to fit it into the canvas.
In fact, I wouldn't even be surprised if he never even considered the fact until the map was released that the area on top of that ramp leads to the difficulties in forming a concave. And if he did consider it it wasn't an intention but a concession because again, it would be a cosmological coincidence that his ramp design how he all wanted it from the start just happened to magically make the map fit tightly into a rectangle. Your abundant use of the phrase "cosmological coincidence" suggests that the probability of whatever you're describing sound remote or minute, when it isn't. Maps have always been rectangular. Continuing to design maps in a rectangular fashion, you develop a better skillset for creating well-crafted rectangular maps, as is only logical. It's not at all "cosmological coincidence", nor is it a concession. It is a development of a skill in reflex to a limitation. Any shape of map has a limit to it, each requiring a different sense of proportion and gestalt to succeed with. Rectangles just happen to be practical from a programming standpoint.
|
On February 08 2013 16:48 NewSunshine wrote:Show nested quote +On February 08 2013 16:34 SiskosGoatee wrote: It's not a point, it's a logical conclusion. If you think C does not follow from A and B you need a crashcourse in logic. I call it a point because there is not an acceptable transition between the three. The first point is practical, as I've said. The second point is an observation, whose correctness I concede. Point C disregards, completely, the mapmaker himself, and implies that his design choices are all rigidly governed by the rectangular border, which is wrong. See Daybreak, see Ohana, for 2 decent examples.
You honestly don't see the logic of.
A: x is influenced by y. B: y is has arbitrary elements. C: therefore, x is influenced by arbitrary elements
This isn't a point, this is a logical conclusion which follows from the premises. If you accept the premises the conclusion follows, there's nothing you can do about it.
Your abundant use of the phrase "cosmological coincidence" suggests that the probability of whatever you're describing sound remote or minute, when it isn't. Maps have always been rectangular. Continuing to design maps in a rectangular fashion, you develop a better skillset for creating well-crafted rectangular maps, as is only logical. It's not at all "cosmological coincidence", nor is it a concession. It is a development of a skill in reflex to a limitation. Any shape of map has a limit to it, each requiring a different sense of proportion and gestalt to succeed with. Rectangles just happen to be practical from a programming standpoint. So again, you admit that they have been designed specifically to be rectangular while you implicitly denied it the last time around.
Like I said (and to which you agree), the map is made to fit into a rectangular canvas, the canvas is not designed to fit around the map. The fact that we choose a rectangular canvas is arbitrary and based on tradition, something we once decided without actually thinking about it and continue to perpetuate up to this day simply because no one raised the issue while other shapes might be more optimal. Therefore, the map is designed around a self imposed limitation that it has to fit tightly into a rectangular canvas with no apparent reasoning behind it.
The only reason for it with respect to gameplay and not aesthetics is dead space management, and it has already been shown that NFZ's can solve this issue without bugging out if placed correctly. There is no reason to limit yourself to a rectangular canvas. People just do it because they've always done it and never bothered to ask themselves why they do it like with so many things. Like for instance, for years all ports on computers were always at the back, people just did it like that even though there was no reason, they just didn't question it until someone had the brilliant idea of putting USB ports and audio ports at the front of a computer case and it caught on massively. Just like that, we limit ourselves to a square canvas because no one ever bothered to consider that there are other shapes. NFZ's allow you to create a canvas in whatever weird shape you desire. Limiting yourself to a square canvas is like limiting yourself in a map to make all the mains perfect squares. I'm sure it can create some good maps but in the end it's still an arbitrary restriction you put on yourself that has no compelling reason for it and more can be achieved if you rid yourself of it.
|
Using space efficiently is a legacy of broodwar where the map sizes couldn't be changed and map-makers had to squeeze everything they wanted to do into the map size. Yes the rectangular map shape is arbitrary, what's wrong with it? And we can't use no-fly zones to cover large areas because they are extremely buggy, i already tried that.
The three points the op pointed are intentional. The main ramp is close to the map bounds so it's far away from the nat choke and prevent free forcefield lock. The highground path leading gives a better attacking position than running up the ramp. The side path leading to the corner expand is there to give a small counter attack path.
Concerning the pieces of terrain passing the map bounds, they intentionally do, you can see that i put doodads to show that this area is unpassable.
Concerning the map size (it's 126x132 btw), i started with 128x128 and then i reduced or enlarged to width and height to reduce unnecessary air space. You shouldn't adapt the layout to the map size but do the reverse, adapting the map size to your layout, sc2 gives this luxury which broodwar didn't have.
|
On February 08 2013 19:55 Superouman wrote: Using space efficiently is a legacy of broodwar where the map sizes couldn't be changed and map-makers had to squeeze everything they wanted to do into the map size. Yes the rectangular map shape is arbitrary, what's wrong with it? And we can't use no-fly zones to cover large areas because they are extremely buggy, i already tried that.
The three points the op pointed are intentional. The main ramp is close to the map bounds so it's far away from the nat choke and prevent free forcefield lock. The highground path leading gives a better attacking position than running up the ramp. The side path leading to the corner expand is there to give a small counter attack path.
Concerning the pieces of terrain passing the map bounds, they intentionally do, you can see that i put doodads to show that this area is unpassable.
Concerning the map size (it's 126x132 btw), i started with 128x128 and then i reduced or enlarged to width and height to reduce unnecessary air space. You shouldn't adapt the layout to the map size but do the reverse, adapting the map size to your layout, sc2 gives this luxury which broodwar didn't have. And Superouman appears to end the thread that should have ended when I first posted with similar statements. The OP, while possibly benign in intent, just seems to suffer from a lot of fundamental mapmaking knowledge.
To Siskos: I haven't bothered to read your subsequent responses. I apologize, as I've always followed through when debunking your arguments in the past (both on TL and Reddit), but in this case there isn't even anything to debate, unless I feel like wasting a lot of time for nothing over an entirely fabricated mapmaking issue. Furthermore, reading Superouman's post makes me feel even less inclined to waste my time, even if I did have time to kill.
EDIT: Oh shit, I misquoted the map bounds for Cloud Kingdom. I was thinking of Antiga's map bounds for some reason. Good eye, Superouman.
|
On February 08 2013 19:55 Superouman wrote:
Yes the rectangular map shape is arbitrary, what's wrong with it? And we can't use no-fly zones to cover large areas because they are extremely buggy, i already tried that. THere are ways to place them on the edges where they don't bug and get people stuck. I remember there was some Blizz post about it, the major trick is letting the hard edges not overlap and touch exactly.
What's wrong with it is that it's an arbitrary limitation which like any limitation is a limitation which doesn't serve a reason. Limitations in the end always limit varieties, but some limitations have a clear purpose, another limition people tend to use ismaking more than 2 bases on each map, that one has a clear purpose, this one doesn't.
The three points the op pointed are intentional. The main ramp is close to the map bounds so it's far away from the nat choke and prevent free forcefield lock. That doesn't speak about the lack of forming a concave above. If the map wasn't a square canvas it could both be far away and allow for a concave.
The highground path leading gives a better attacking position than running up the ramp. The side path leading to the corner expand is there to give a small counter attack path. Fair enough, but again, it's a major coincidence how tightly it fits into the canvas. As in, it might not be the most optimal form to achieve what you wanted to achieve, there might be a form which expresses this idea even better.
Concerning the pieces of terrain passing the map bounds, they intentionally do, you can see that i put doodads to show that this area is unpassable Of course, but it still fits into a rectangle so tightly that it's obvious that the map was designed to fit into the canvass rather than the canvass being designed to fit around the map.
Concerning the map size (it's 126x132 btw), i started with 128x128 and then i reduced or enlarged to width and height to reduce unnecessary air space. You shouldn't adapt the layout to the map size but do the reverse, adapting the map size to your layout, sc2 gives this luxury which broodwar didn't have. Absolutely, I personally just start at 256/256 and afterwards shrink. But I feel we should go even further and not only adapt the size of the canvass to the map but also the shape. Which is another nice luxury usage of NFZ's give you. There are ways to minimize and outright eliminate the pathing bugs if you place them right.
On February 08 2013 20:41 iamcaustic wrote:Show nested quote +On February 08 2013 19:55 Superouman wrote: Using space efficiently is a legacy of broodwar where the map sizes couldn't be changed and map-makers had to squeeze everything they wanted to do into the map size. Yes the rectangular map shape is arbitrary, what's wrong with it? And we can't use no-fly zones to cover large areas because they are extremely buggy, i already tried that.
The three points the op pointed are intentional. The main ramp is close to the map bounds so it's far away from the nat choke and prevent free forcefield lock. The highground path leading gives a better attacking position than running up the ramp. The side path leading to the corner expand is there to give a small counter attack path.
Concerning the pieces of terrain passing the map bounds, they intentionally do, you can see that i put doodads to show that this area is unpassable.
Concerning the map size (it's 126x132 btw), i started with 128x128 and then i reduced or enlarged to width and height to reduce unnecessary air space. You shouldn't adapt the layout to the map size but do the reverse, adapting the map size to your layout, sc2 gives this luxury which broodwar didn't have. And Superouman appears to end the thread that should have ended when I first posted with similar statements. The OP, while possibly benign in intent, just seems to suffer from a lot of fundamental mapmaking knowledge. To Siskos: I haven't bothered to read your subsequent responses. I apologize, as I've always followed through when debunking your arguments in the past (both on TL and Reddit), but in this case there isn't even anything to debate, unless I feel like wasting a lot of time for nothing over an entirely fabricated mapmaking issue. Furthermore, reading Superouman's post makes me feel even less inclined to waste my time, even if I did have time to kill. EDIT: Oh shit, I misquoted the map bounds for Cloud Kingdom. I was thinking of Antiga's map bounds for some reason. Good eye, Superouman. That you think superouman's points disprove my original point in any way shape or form proves you never understood it. It isn't about Cloud Kindom and like I said, this isn't about specific maps. Everything could've been completely intentional about CK and that still does not invalidate the point I was trying to make and if you think otherwise you never understood it. I could've used any map as an example, I just choose CK because it's very well known. That you also think that Daybreak applies less to my point than CK proves you don't understand the point I'm trying to make, it isn't about filling up the canvas as much as possible in case you wonder.
|
If the shape of the map wouldn't be a geometrical shape, how could you tell you can walk on this area and not this one? Pro players don't even remember if a map has all spawn allowed or not, i don't want to think about players who don't play a lot. I would be a mess for the players and developpers
|
On February 08 2013 21:35 Superouman wrote: If the shape of the map wouldn't be a geometrical shape, how could you tell you can walk on this area and not this one? Pro players don't even remember if a map has all spawn allowed or not, i don't want to think about players who don't play a lot. I would be a mess for the players and developpers Well, you can usually clearly indicate with doodats what the edge of a map is. And this only applies to airspace. I really don't think unusual airspace is a problem to novice players at all. Memorizing you can't fly over those huge trees on the edge of a map is like memorizing that the third geyser on Atlantis is rich.
|
On February 08 2013 21:35 Superouman wrote: If the shape of the map wouldn't be a geometrical shape, how could you tell you can walk on this area and not this one? Pro players don't even remember if a map has all spawn allowed or not, i don't want to think about players who don't play a lot. I would be a mess for the players and developpers This is what I realized overnight. This could get messy - the minimap needs to be very clear imo.
---
That said, I'm all in favor of more geometrical shapes for a map, for example: Trapezoid Parallelogram Diamond Triangle Hexagon Octagon
*OR*
They could give us a "CUT/SNIP" tool (similar to cropping) to cut the corners of maps... only 4 corners (1 cut per corner), but if they let us choose the x and y values with this tool you could make all the above shapes and more.
edit: Might as well throw us circles/ovals while they're at it tbh.
This can help arcade mapmakers too, not just melee.
edit2: and no on Ohana I wouldn't do anything with that dead space other than maybe cut/snip some of it
|
Arcade mappers can use triggers and custom minimap images, the jerks. Our problem is that the only way to adjust air pathing is buggy (unusable). Otherwise we could make any shape we want. (Obviously.)
In fact you could use these things anyway in a melee map, it wouldn't change the game at all.
|
I agree w/ barrin, some other shape boundaries should be implemented by blizzard (for melee maps).. it probably wouldn't even be that hard (you could leave the actual map as a rectangle, just the ability to affect the bounds is what's needed). Triangle would be especially nice so that 3p maps can finally be perfectly balanced (there's always been the problem of differing airspaces w/ 3p maps..).
|
Circle! Then we can make any number of players. cough igrok cough
|
btw, in my first comment on the Ohana map:
On December 17 2011 06:24 Barrin wrote: ... Also I think there's too little % of pathable area. More specifically, too much air space on the edges of the map for air units to maneuver around (too easily going undetected coming in and too hard to catch on their way out). ...
A better but more tedious way of fixing this would be to basically tilt the entire map clockwise by like 20 degrees while conforming the main/nat/third edge of map vulnerabilities to the way they were before (hugging) and simultaneously shaving off the excess on the right and left edges ...
[confusing, I know]
|
|
On February 09 2013 07:30 Fatam wrote: I agree w/ barrin, some other shape boundaries should be implemented by blizzard (for melee maps).. it probably wouldn't even be that hard (you could leave the actual map as a rectangle, just the ability to affect the bounds is what's needed). Triangle would be especially nice so that 3p maps can finally be perfectly balanced (there's always been the problem of differing airspaces w/ 3p maps..). Yeah, just being able to basically draw the canvas, maybe as a vector which of course automatically implies a no bugged NFZ would be great.
|
|
|
|