|
On February 14 2011 09:35 Schnullerbacke13 wrote:Show nested quote +On February 14 2011 08:43 mahnini wrote:On February 14 2011 08:28 Schnullerbacke13 wrote:On February 14 2011 07:39 mahnini wrote:On February 13 2011 20:53 Sfydjklm wrote:On February 13 2011 07:07 mahnini wrote:On February 12 2011 11:45 Space Invader wrote:On February 11 2011 05:12 mahnini wrote:On February 11 2011 04:58 GreatFall wrote:
I knew mules were strong but damn these figures make them look downright amazing. Also, I think that a macro based game like SC2 would be better off with a 300 supply cap. Such a nice summary and writeup man. You did a lot of work here. mules look great because they are being used constantly so you see a nonstop growth in mineral intake which brings your attention only to the huge jump that occurs in mining when the first mule lands. i don't think people realize what they are asking for when they want a 300 supply cap. that's a 50% increase in the number of units currently available, if anything it would have the opposite effect of late-game zvp the op wanted where in this situation protoss can turtle to 300/300 instead of 200/200 and move out and roll everything. what's the point of saturating additional bases if your opponent can have a nearly 100% larger army supply-wise than you? The effect of a 300 cap would more likely be that the current 3-4 base max army would be almost entirely irrelevant. Due to Zergs maxing out much faster than Terran or Protoss, a Zerg could attain a 300 supply army at approximately the same time a Protoss reached a 200 food army, if not sooner. And at that point a Zerg would just crush the Protoss. One major problem for Zergs at the moment is that they reach the 200 cap and they're forced to be aggressive because that point the Protoss is normally at around 150-160 supply and on 3 bases. This ~30 food advantage in army isn't nearly enough to crack a turtling player given how efficient their units are so the P is able to just defend until it reaches the 'invincible army'. right, but the entire point of that is we aren't talking about the supply being the problem. the supply isn't the problem. the theoretical problem is that zerg reaches max saturation too late to abuse his army advantage. but there is no data in the OP to back up that assumption. Theres teh data:
Zerg, SC2, with 54 workers equally distributed on 4, 5 or 6 bases: ~15384 minerals over 5 minutes.
Protoss, SC2, with 54 workers confined to 3 bases: 14586 minerals over 5 minutes.
And theres the analysis of the data:
Based on these data, the only way to secure a macro lead in SC2 seems to be by rushing to 3 fully saturated bases as quickly as humanly possible. The entire objective for zerg in SC2 seems to have been reduced to recklessly rushing to a macro lead as quickly, stupidly and foolishly as possible before the game caps the chance for any macro lead to develop.
A proper claim that we do not want the zerg gameplay to deteriorate into rushing to 3rd. 300 supply is just an offered solution. if i use 40 workers as the benchmark and compare 2 bases with 3 you'd probably see the same results, then clearly that third base isn't worth getting. what you see is a snapshot that does not take into account the production ability of zerg larva inject. when have you ever seen a zerg have the same amount of workers as a terran or protoss in the early or midgame? it almost never happens. his so called analysis is a giant assumption based off cherry picked data. zerg will always technically be "rushing" to secure the macro advantage as soon as possible, that's the entire point. all the adverbs dropped in there are signs of clear bias (recklessly, stupidly, foolishly). on top of that, all these conclusions are made without ever showing zerg worker production side by side with terran and protoss which is the one point his entire argument draws from. sorry you did not get it. droning more than a P or T always creates a weakness in army. the strength of Z was in BW having more or less equal amount of workers,but being on more bases (better mining efficiency). this is not rewarded that much in SC2. Also it is not possible to get a reasonable advantage beyond 3 bases (supply cap), so there is a small timing window where the Z can take a notable advantage of spawn larvae, because with > 80 workers there is not enough supply for army units. Conclusions for Z: better deny 3rd instead of trying to outmacro by taking a 4th. Attack when on 2 bases ~saturated (window of macro advantage lasts til 4th). ok but this isn't bw. if spawn larva were in bw or sc2 had bw mining mechanics zerg would be ridiculously strong. point being, the emphasis of a stronger economy has shifted from having more bases to having better saturation. this, in and of itself, doesn't really cause balance issues. the 200 supply cap doesn't affect anything, it's an external factor that people are blaming rather than the fundamental issue, which is time to saturation. if a zerg is on 3 base with a max army of 125 and a protoss is on 2 base with a max army of 150, if both max at the same time, then zerg is at a disadvantage. as others have pointed out, however, zergs do not saturate and produce at the same rate of other races, so the issue is not supply, the issue is whether zerg is able to take advantage of their window of opportunity where they have an economic and army size advantage. if your fundamental issue is with the way mining efficiency works, an increased supply cap to 300 does nothing for you. agree with shift to saturation instead of # expansions. But this means, the macro advantage of Z is for a limited time, because of supply cap. In mid to late game, Z will have no macro advantage anymore, so inject larva is not that useful, because income and supply are the limiting factors then, not production. I think SC2 is not designed for huge 4+ bases games, so i concentrate on deciding the game latest when having 3 bases saturated.
i can agree with most of this, but the problem with this, then, is that the OP does not demonstrate that zerg has a hard time securing an advantage on 3 base.
people are looking at this far too objectively, for lack of a better word. zerg doesn't NEED a 4th base to compete as long as the relative advantage is there with their 3rd base. even in bw at some point extra bases didn't really give that much return but it was considered that 4 base basically meant end game because that was the point where zerg had the largest advantage vs number of bases.
|
On February 14 2011 09:46 JustPlay wrote:Show nested quote +On February 14 2011 07:13 Novembermike wrote:
Basically, BW rewards you for expanding, SC2 rewards you for saturating bases. This is the most important part of the data, and it's also the only thing that the data shows clearly. It's actually depressing that this shift in reward exists, because it really trivializes additional bases more than you realize while playing.
yep. this is the only part that really matters.
this can cause huge problems in the future and is not good for the gameplay. look at the gsl maps. how can you do a macro map thats balanced when just taking your nat and 3rd is all you need. guess people and esp P will just sit on their bases,turtle and then do one big deathball push once they hit 200/200.
|
On February 12 2011 12:33 Bowdy wrote:
I just published a map on NA called "LaLush Test Map". It's Shakuras Plateau with some of the ideas from this thread implemented:
Mains and the top middle and bottom middle expansions have 6 mineral patches, with 2 far patches being gold.
All other expos have 5 patches, with 1 far patch being gold.
All bases have only 1 gas geyser with 8 gas per trip. Main base geysers have 3500 gas, all other geysers have 2500 gas.
It plays like a completely different game, and frankly I like it better. There would be tons of balance changes needed however, as the game just isn't balanced around this style of play. It definitely encourages expanding much more often, as the gold patches make it worth having a bunch of under-saturated bases as opposed to 2 or 3 fully saturated.
Play around on it if you guys want, but be warned: THE MULE IS IMBA UNTIL I FIGURE OUT HOW TO MAKE IT NOT GET BONUS MINERALS FROM GOLD PATCHES! Feel free to leave any feedback or suggestions other than that, gl hf!
edit: the gold patches have 2100 minerals instead of 1500 to mine out at the same rate.
This is a really interesting idea. Did anything ever come of this? Any more games played on it?
|
If they reduce supply costs of drone to 1/2 control, would that help out with Zerg's macro limitations?
|
On February 14 2011 08:43 mahnini wrote:Show nested quote +On February 14 2011 08:28 Schnullerbacke13 wrote:On February 14 2011 07:39 mahnini wrote:On February 13 2011 20:53 Sfydjklm wrote:On February 13 2011 07:07 mahnini wrote:On February 12 2011 11:45 Space Invader wrote:On February 11 2011 05:12 mahnini wrote:On February 11 2011 04:58 GreatFall wrote:
I knew mules were strong but damn these figures make them look downright amazing. Also, I think that a macro based game like SC2 would be better off with a 300 supply cap. Such a nice summary and writeup man. You did a lot of work here. mules look great because they are being used constantly so you see a nonstop growth in mineral intake which brings your attention only to the huge jump that occurs in mining when the first mule lands. i don't think people realize what they are asking for when they want a 300 supply cap. that's a 50% increase in the number of units currently available, if anything it would have the opposite effect of late-game zvp the op wanted where in this situation protoss can turtle to 300/300 instead of 200/200 and move out and roll everything. what's the point of saturating additional bases if your opponent can have a nearly 100% larger army supply-wise than you? The effect of a 300 cap would more likely be that the current 3-4 base max army would be almost entirely irrelevant. Due to Zergs maxing out much faster than Terran or Protoss, a Zerg could attain a 300 supply army at approximately the same time a Protoss reached a 200 food army, if not sooner. And at that point a Zerg would just crush the Protoss. One major problem for Zergs at the moment is that they reach the 200 cap and they're forced to be aggressive because that point the Protoss is normally at around 150-160 supply and on 3 bases. This ~30 food advantage in army isn't nearly enough to crack a turtling player given how efficient their units are so the P is able to just defend until it reaches the 'invincible army'. right, but the entire point of that is we aren't talking about the supply being the problem. the supply isn't the problem. the theoretical problem is that zerg reaches max saturation too late to abuse his army advantage. but there is no data in the OP to back up that assumption. Theres teh data:
Zerg, SC2, with 54 workers equally distributed on 4, 5 or 6 bases: ~15384 minerals over 5 minutes.
Protoss, SC2, with 54 workers confined to 3 bases: 14586 minerals over 5 minutes.
And theres the analysis of the data:
Based on these data, the only way to secure a macro lead in SC2 seems to be by rushing to 3 fully saturated bases as quickly as humanly possible. The entire objective for zerg in SC2 seems to have been reduced to recklessly rushing to a macro lead as quickly, stupidly and foolishly as possible before the game caps the chance for any macro lead to develop.
A proper claim that we do not want the zerg gameplay to deteriorate into rushing to 3rd. 300 supply is just an offered solution. if i use 40 workers as the benchmark and compare 2 bases with 3 you'd probably see the same results, then clearly that third base isn't worth getting. what you see is a snapshot that does not take into account the production ability of zerg larva inject. when have you ever seen a zerg have the same amount of workers as a terran or protoss in the early or midgame? it almost never happens. his so called analysis is a giant assumption based off cherry picked data. zerg will always technically be "rushing" to secure the macro advantage as soon as possible, that's the entire point. all the adverbs dropped in there are signs of clear bias (recklessly, stupidly, foolishly). on top of that, all these conclusions are made without ever showing zerg worker production side by side with terran and protoss which is the one point his entire argument draws from. sorry you did not get it. droning more than a P or T always creates a weakness in army. the strength of Z was in BW having more or less equal amount of workers,but being on more bases (better mining efficiency). this is not rewarded that much in SC2. Also it is not possible to get a reasonable advantage beyond 3 bases (supply cap), so there is a small timing window where the Z can take a notable advantage of spawn larvae, because with > 80 workers there is not enough supply for army units. Conclusions for Z: better deny 3rd instead of trying to outmacro by taking a 4th. Attack when on 2 bases ~saturated (window of macro advantage lasts til 4th). ok but this isn't bw. if spawn larva were in bw or sc2 had bw mining mechanics zerg would be ridiculously strong. point being, the emphasis of a stronger economy has shifted from having more bases to having better saturation. this, in and of itself, doesn't really cause balance issues. the 200 supply cap doesn't affect anything, it's an external factor that people are blaming rather than the fundamental issue, which is time to saturation. if a zerg is on 3 base with a max army of 125 and a protoss is on 2 base with a max army of 150, if both max at the same time, then zerg is at a disadvantage. as others have pointed out, however, zergs do not saturate and produce at the same rate of other races, so the issue is not supply, the issue is whether zerg is able to take advantage of their window of opportunity where they have an economic and army size advantage. if your fundamental issue is with the way mining efficiency works, an increased supply cap to 300 does nothing for you.
I agree that there is no objectiv explanation on why zerg is or is not weak because of the maccro mechanics. But I think you are focussing too much on "balance or not" and not on the game. What is clear in this situation is that SC2 has lost a dimension if you compare it to SC1. Expanding is not rewarding, so it's obvious that zerg's playstyle is the most frustrated in this situation.
It is fairly easy to secure 3 bases and sature them as zerg considering how the game is build at the moment. With 2 hatch and queens your can easily build 10 drone in a row saturation in this situation is easy. However, against an opponent who turtle and actually get his third (don't tell me it's hard), what can you do ? That's a pitty that you will instantly be in the lead (economically) or on even ground as a protoss for exemple, even if your opponent has 4+ expand.
Look at Morrow vs Socke on lost temple during assembly I think it's a perfect exemple on why something should be done. Socke making so many photon cannon because he knows he only need 3 bases (with gold), while Morrow has complete map control. Sure morrow won, the game is fairly balance in my opinion, but against such turtle play, zerg should not be forced to attack, expanding and taking map control should also be another way to counter this kind of play, especially considering how it is easy for zergs to do so.
With this said, it's pretty obvious that a 300 supply can be a good idea, even if it's not my opinion because it will fuck up so many things.
Having gold minerals in the middle of every mineral line is not a good idea either in my opinion because it will change so much the early game (just like having a 0,5 supply drone). The only "good" counter to this should be to reduce the number of mineral patches on expands and maybe bring back the 8 gaz trip.
|
On February 10 2011 19:39 mahnini wrote:if i'm not mistaken your issue is with the fact that extra bases do not pay off until you are supersaturated and therefore early expansions do not pay off right away. i believe this to be a valid concern, however, you extrapolate on these ideas and conclude that Show nested quote +Zerg’s play will be centered around saturating 3 bases as quickly as possible and launching suicide attacks at the opponents’ thirds. Protoss’ play will be centered around camping and delaying until they’ve reached their invincible end game composition on 3+ bases. Terran’s play will… no idea. this statement is made with the only backing being that at a certain supply all mining equalizes without regard for the time it takes to reach that supply. while it may be true that the most logical cutoff for obtaining extra bases stops at 3 due to supply constraints when fully saturated, this is done without the consideration of time to saturation. this is easily overlooked because of the omission of zerg mining data. the idea that zerg is required to rush to saturate 3 bases completely discounts zerg's additional production capability (aka the queen) which is obtained faster (build time-wise) and for less minerals (which also means faster game-time-wise) than extra hatcheries. a hatchery + queen spawns as much larva as 2.27 hatcheries (assuming you are constantly respawning larva from the hatchery + inject). the only comparison between bw and sc2 time to saturate is seen here in this graph we do not see the effect of chronoboost on saturation time, instead, we see data points of isolated mining per five minutes which, to a great extent, accentuates the midgame mining discrepancy. for example, if you're driving at 1MPH and i drive at 2MPH and we both do this for 10 hours, at the end i have a 10 mile lead; however, if we only drive for 0.1 hours i only end up with a 0.1 mile lead. this seems really stupid but it makes a big difference when the end quantity is more important than the rate because there is a threshold of usefulness. 1 mineral per second is completely negligible over 17 seconds compared to over 300 seconds; at the end of your so called midgame mineral discrepancy you probably end up with less of a difference in net minerals than you'd think. you also base your final conclusion that protoss will dominate on the 3 base limit, however, in a late-game situation, with the rate at which bases seem to mine out, i believe army production capability to be a much larger problem. due to new macro mechanics such as larva inject, chronoboost, and warp gates, much of the time-cost required to reinforce is moved to before the actual production (stockpiling larva, stockpiling chronoboost, and warpgates in general) and in late-game engagements it's usually a contest of who has the most durable army and who can reinforce the fastest, but when one race has both attributes then there is a (superficial) problem.
Although I didn't word it as clearly as you did, I think the paragraph you quoted implies what you imply about "time to saturation". There would be no point for zerg to rush to 3 bases if they saturated at the same rates as the other races -- or if their armies were as durable.
With that said, I actually agree with you that the effects of mineral surplus surges are somewhat exaggerated by my graphs, but I still believe they pose a problem and influence the way the game is played out on different sized maps though.
The reason Terran have been so stable and dominant on the Blizzard maps is exactly because of the fact that they have been able to negate the "time to saturation" issue by putting pressure and forcing the same amount of bases for as long as possible. They have been able to force "mining to be equalized" throughout games.
Furthermore, the issue you brought up about the races' different production rates can be used as an argument to explain why Terran will have a particularly hard time attacking across the new GSL maps. I believe that army durability will be the greatest deciding factor on the new big maps because races will have a much harder time killing eachother off. Mining will eventually equalize after which point there will be 1 attempt to break the more durable opponent through "quick" cross map reinforcements. After that you will no longer have a bulk of excess minerals to reinforce with a higher supply army due to better macro.
The issue of reinforcement rates and production capability that you bring up, I believe to play a bigger role on smaller Blizzard sized maps as opposed to large ones. On large maps from wat I've experienced and seen so far, production capability seems most pronounced in a defensive capacity as opposed to an offensive one (except possibly for Protoss). I think zergs in ZvP will be able to defend pushes galantly in a cost inefficient manner. But then attack across the map and kill P? No.
Next clash will probably be when protoss reaches 200/200 again, at which point there will be no bulk of excess minerals to abuse for a higher production rate.
That's why I think the 3base ceiling will force gameplay into the mould I described. Mining rate equalization will be a big problem on large maps.
If the game is going to be balanced for large maps and for a 300 supply cap, I think the right way to go is to make macro mechanics and especially the larva inject scale up through tiers and possibly through upgrades so a part of that "time to saturation" problem you mention gets addressed. Also the cap for amount of larva one single hatchery can hold lategame should probably be lowered.
But I think Blizzard will stick to the current design of the game, so no point in discussing any such changes. I guess we'll have to see how these large maps play out before making too many outrageous claims.
|
On February 14 2011 12:26 Rokk wrote:Show nested quote +On February 12 2011 12:33 Bowdy wrote:
I just published a map on NA called "LaLush Test Map". It's Shakuras Plateau with some of the ideas from this thread implemented:
Mains and the top middle and bottom middle expansions have 6 mineral patches, with 2 far patches being gold.
All other expos have 5 patches, with 1 far patch being gold.
All bases have only 1 gas geyser with 8 gas per trip. Main base geysers have 3500 gas, all other geysers have 2500 gas.
It plays like a completely different game, and frankly I like it better. There would be tons of balance changes needed however, as the game just isn't balanced around this style of play. It definitely encourages expanding much more often, as the gold patches make it worth having a bunch of under-saturated bases as opposed to 2 or 3 fully saturated.
Play around on it if you guys want, but be warned: THE MULE IS IMBA UNTIL I FIGURE OUT HOW TO MAKE IT NOT GET BONUS MINERALS FROM GOLD PATCHES! Feel free to leave any feedback or suggestions other than that, gl hf!
edit: the gold patches have 2100 minerals instead of 1500 to mine out at the same rate. This is a really interesting idea. Did anything ever come of this? Any more games played on it?
Yeah some teammates played a bunch of games on it. End result is basically the games are a ton of fun but theres some serious balance issues, notably in the early game (which therefore extend a large lead going to the mid/late game)
|
On February 15 2011 04:34 Bowdy wrote:Show nested quote +On February 14 2011 12:26 Rokk wrote:On February 12 2011 12:33 Bowdy wrote:
I just published a map on NA called "LaLush Test Map". It's Shakuras Plateau with some of the ideas from this thread implemented:
Mains and the top middle and bottom middle expansions have 6 mineral patches, with 2 far patches being gold.
All other expos have 5 patches, with 1 far patch being gold.
All bases have only 1 gas geyser with 8 gas per trip. Main base geysers have 3500 gas, all other geysers have 2500 gas.
It plays like a completely different game, and frankly I like it better. There would be tons of balance changes needed however, as the game just isn't balanced around this style of play. It definitely encourages expanding much more often, as the gold patches make it worth having a bunch of under-saturated bases as opposed to 2 or 3 fully saturated.
Play around on it if you guys want, but be warned: THE MULE IS IMBA UNTIL I FIGURE OUT HOW TO MAKE IT NOT GET BONUS MINERALS FROM GOLD PATCHES! Feel free to leave any feedback or suggestions other than that, gl hf!
edit: the gold patches have 2100 minerals instead of 1500 to mine out at the same rate. This is a really interesting idea. Did anything ever come of this? Any more games played on it? Yeah some teammates played a bunch of games on it. End result is basically the games are a ton of fun but theres some serious balance issues, notably in the early game (which therefore extend a large lead going to the mid/late game)
What kind of balance issues are we talking about? Would there be ways to modify maps to keep the same spirit while reducing the imbalance present in the early game, or is it something that would require Blizzard to patch?
|
On February 15 2011 04:34 Bowdy wrote:Show nested quote +On February 14 2011 12:26 Rokk wrote:On February 12 2011 12:33 Bowdy wrote:
I just published a map on NA called "LaLush Test Map". It's Shakuras Plateau with some of the ideas from this thread implemented:
Mains and the top middle and bottom middle expansions have 6 mineral patches, with 2 far patches being gold.
All other expos have 5 patches, with 1 far patch being gold.
All bases have only 1 gas geyser with 8 gas per trip. Main base geysers have 3500 gas, all other geysers have 2500 gas.
It plays like a completely different game, and frankly I like it better. There would be tons of balance changes needed however, as the game just isn't balanced around this style of play. It definitely encourages expanding much more often, as the gold patches make it worth having a bunch of under-saturated bases as opposed to 2 or 3 fully saturated.
Play around on it if you guys want, but be warned: THE MULE IS IMBA UNTIL I FIGURE OUT HOW TO MAKE IT NOT GET BONUS MINERALS FROM GOLD PATCHES! Feel free to leave any feedback or suggestions other than that, gl hf!
edit: the gold patches have 2100 minerals instead of 1500 to mine out at the same rate. This is a really interesting idea. Did anything ever come of this? Any more games played on it? Yeah some teammates played a bunch of games on it. End result is basically the games are a ton of fun but theres some serious balance issues, notably in the early game (which therefore extend a large lead going to the mid/late game)
Great to hear people trying experiments like this. I'd be interested to know how it would play out with say 7 mineral patches and 2 geysers on all bases/expos. That might not mess with balance too much, but still promote expansions. Feel free to PM me if you'd like to do some playtesting.
|
Forgive me for not fully reading all 20 pages. After skimming the first 10, I saw only one post that commented on the most effective tweak we could use. This is an outline of that solution, which I believe is the strongest change we can make from the standpoint of map design. Altering core game mechanics is off limits.
Perceived problem: Expanding is less important than saturation, so getting more than 3 bases brings little reward. Moderate worker count is functionally the same on one mineral line compared to split amongst two or more mineral lines.
Direct solution plan: Can you construct a variation of the mineral line such that saturation is less rewarding compared to expanding?
First, the number of mineral patches can be reduced, capping the bases' mineral income at a lower level, forcing players to expand. This only requires a faster expansion; it doesn't directly do anything about saturation>expo. Indeed, you can saturate faster, which might encourage more all-in play.
Second, you can decrease the capacity of some or all mineral patches. Generally, this would have a delayed adverse consequence to the same effect as the first option. The base will be mined out sooner, prompting a player to expand. Again, this does nothing to directly address saturation>expo, and again, this would likely encourage all-in play. It would take a 5+ minutes horizon of planning to incorporate saturation choices on a low capacity main base, and even natural. The third base seems like the first opportunity to use this without strongly promoting all-ins. (As seen on Tal'Darim Altar.)
Third, you can use mixed-in gold patches to incentivize expansions. This has a major pitfall in terran imbalance with the mule. The only way to offset this without changing the game mechanics is to put the gold patch farther away, giving less mining trips per mule. Is it possible to provide a gold patch close enough that it is more valuable than a normal distance blue patch, but which mitigates the mule? I will update with testing numbers and/or math.
Fourth, the best option. You can place mineral patches farther away. If you moved all patches further away, the effect would be to require more workers before full saturation, and an overall decreased rate of income. This is not what we want, exactly. In fact, this is an analogue of having more mineral patches (with less minerals returned per trip), which just favors saturation. What we want to do is place some patches at a standard distance and some far away, requiring 4 or 5 workers for full saturation. This would give you an income graph resembling the BW graphs. If you have N close patches, building workers past 2N yields lower income gains per worker, and you would require more workers total to be fully saturated. An expansion would provide you with more close patches to put your workers to better use, rebalancing expo vs saturation. This is a less drastic, finer-grained version of incentivizing with gold patches, and I think it supercedes the gold option.
This earlier post includes diagrams of these options, and was the only one I saw that discussed what I have listed as the fourth and best option.
This "solution" does affect the later stages of 1- and 2-base play quite a bit. I assume it would require significant readjustment in build orders and the general flow of the game. Regardless, it would certainly promote expansions beyond a 3rd base.
Comments on BW vs SC2: As a point of clarification, I want to discuss the underlying mechanics of this "solution" in relation to BW. Whatthefat's diagram shows a BW mineral line, which was generally straight and tangent, not concave. If you stacked up 8 patches in a roughly straight tangent line in SC2, the difference would be negligible. The actual increase in trip time is not much at all. In fact, lots of maps have slightly suboptimal minerals that illustrate this, like Desert Oasis. For one, SC2 CCs have an edge that's 5 squares long, so transverse distance doesn't increase the diagonal as much as in BW. However, the main factor is worker AI. Workers pair up relatively efficiently in SC2, and they bounce a lot more even in earlier stages of saturation in BW. This multiplied the travel times in BW, whereas the small trip increases are still small in SC2. Because of this, the staggered patch distances have to be exaggerated in SC2 to achieve an appreciable effect; you can't just use a straight tangent line. The best shape would be a highly exaggerated concave (with a deep pocket in the center) or a convex arc with distant edge patches.
I will try to do some testing for specific recommendations of mineral line arrangements. And I will get some pictures to help those who don't like concise prose exposition.
tl;dr can burn in hell, but: use some close patches and some far patches to simulate BW saturation.
@Bowdy, Whatthefat, other interested parties: I'm going to make a post in whatthefat's thread in Custom Maps to gather test maps and people to play them.
|
a lot of people here suggest maybe raising supply cap. i just wanted to ask what if they decreased the supply of the roach or hydra or whatever. i know its not well thought out but instead of just raising the cap i wonder what will decrease supply value of certain zerg units will do.
|
+ Show Spoiler +On February 15 2011 06:34 EatThePath wrote:Forgive me for not fully reading all 20 pages. After skimming the first 10, I saw only one post that commented on the most effective tweak we could use. This is an outline of that solution, which I believe is the strongest change we can make from the standpoint of map design. Altering core game mechanics is off limits. Perceived problem: Expanding is less important than saturation, so getting more than 3 bases brings little reward. Moderate worker count is functionally the same on one mineral line compared to split amongst two or more mineral lines. Direct solution plan: Can you construct a variation of the mineral line such that saturation is less rewarding compared to expanding? First, the number of mineral patches can be reduced, capping the bases' mineral income at a lower level, forcing players to expand. This only requires a faster expansion; it doesn't directly do anything about saturation>expo. Indeed, you can saturate faster, which might encourage more all-in play. Second, you can decrease the capacity of some or all mineral patches. Generally, this would have a delayed adverse consequence to the same effect as the first option. The base will be mined out sooner, prompting a player to expand. Again, this does nothing to directly address saturation>expo, and again, this would likely encourage all-in play. It would take a 5+ minutes horizon of planning to incorporate saturation choices on a low capacity main base, and even natural. The third base seems like the first opportunity to use this without strongly promoting all-ins. (As seen on Tal'Darim Altar.) Third, you can use mixed-in gold patches to incentivize expansions. This has a major pitfall in terran imbalance with the mule. The only way to offset this without changing the game mechanics is to put the gold patch farther away, giving less mining trips per mule. Is it possible to provide a gold patch close enough that it is more valuable than a normal distance blue patch, but which mitigates the mule? I will update with testing numbers and/or math. Fourth, the best option. You can place mineral patches farther away. If you moved all patches further away, the effect would be to require more workers before full saturation, and an overall decreased rate of income. This is not what we want, exactly. In fact, this is an analogue of having more mineral patches (with less minerals returned per trip), which just favors saturation. What we want to do is place some patches at a standard distance and some far away, requiring 4 or 5 workers for full saturation. This would give you an income graph resembling the BW graphs. If you have N close patches, building workers past 2N yields lower income gains per worker, and you would require more workers total to be fully saturated. An expansion would provide you with more close patches to put your workers to better use, rebalancing expo vs saturation. This is a less drastic, finer-grained version of incentivizing with gold patches, and I think it supercedes the gold option. This earlier post includes diagrams of these options, and was the only one I saw that discussed what I have listed as the fourth and best option. This "solution" does affect the later stages of 1- and 2-base play quite a bit. I assume it would require significant readjustment in build orders and the general flow of the game. Regardless, it would certainly promote expansions beyond a 3rd base. Comments on BW vs SC2: As a point of clarification, I want to discuss the underlying mechanics of this "solution" in relation to BW. Whatthefat's diagram shows a BW mineral line, which was generally straight and tangent, not concave. If you stacked up 8 patches in a roughly straight tangent line in SC2, the difference would be negligible. The actual increase in trip time is not much at all. In fact, lots of maps have slightly suboptimal minerals that illustrate this, like Desert Oasis. For one, SC2 CCs have an edge that's 5 squares long, so transverse distance doesn't increase the diagonal as much as in BW. However, the main factor is worker AI. Workers pair up relatively efficiently in SC2, and they bounce a lot more even in earlier stages of saturation in BW. This multiplied the travel times in BW, whereas the small trip increases are still small in SC2. Because of this, the staggered patch distances have to be exaggerated in SC2 to achieve an appreciable effect; you can't just use a straight tangent line. The best shape would be a highly exaggerated concave (with a deep pocket in the center) or a convex arc with distant edge patches. I will try to do some testing for specific recommendations of mineral line arrangements. And I will get some pictures to help those who don't like concise prose exposition. tl;dr can burn in hell, but: use some close patches and some far patches to simulate BW saturation. @Bowdy, Whatthefat, other interested parties: I'm going to make a post in whatthefat's thread in Custom Maps to gather test maps and people to play them.
I really support the idea of having a larger variance of distance in the mineral patches. I think doing things like making bases past the natural unattractive in some way compared to your main/nat it will simply promote 2-base all-ins instead of encouraging a macro style.
With the varying distances of patches, I would also be intrigued to see how it would affect gameplay if on top of that, it was possible to add another nexus/cc/hatch to get slightly more efficient mining. This could encourage the OC spam we saw out of boxer and a few others in the GSL, as well as zerg adding on additional hatches to their existing bases.
One of the big things I am worried about though is the mule, even if you arn't going so far as to add gold minerals, any time you reduce the efficiency of workers past a certain point, you are increasing the power of the mule, because it will simply mine from the close patches and have the same income as it does now.
I really hope something is done about this though, as my favorite style, both to watch and to play as with zerg is to take the whole map and sauron my way through, which right now is not nearly as feasible as it was in BW, due to this exact problem.
|
On February 15 2011 19:07 Phanekim wrote:
a lot of people here suggest maybe raising supply cap. i just wanted to ask what if they decreased the supply of the roach or hydra or whatever. i know its not well thought out but instead of just raising the cap i wonder what will decrease supply value of certain zerg units will do.
The problem is that supply can only increase/decrease in whole numbers. Roach/Hydra take up 2 supply right now. That may be a little too high, but 1 supply would be way, way too low. Imagine a 200/200 Zerg army with literally double the amount of units. Would be way to strong.
Only high tech tree units have some room to play around with. A 1 supply decrease to BroodLords? That might make much more sense.
The only way around this issue would be to have slightly different supply caps for different races. But I can only imagine how hard that would be to balance. It would require an entire redesign of the late game.
|
On February 16 2011 04:08 Shadrak wrote:Show nested quote +On February 15 2011 19:07 Phanekim wrote:
a lot of people here suggest maybe raising supply cap. i just wanted to ask what if they decreased the supply of the roach or hydra or whatever. i know its not well thought out but instead of just raising the cap i wonder what will decrease supply value of certain zerg units will do. The problem is that supply can only increase/decrease in whole numbers. Roach/Hydra take up 2 supply right now. That may be a little too high, but 1 supply would be way, way too low. Imagine a 200/200 Zerg army with literally double the amount of units. Would be way to strong.
I have had an issue with this for a very long time, especially when you note that both zerglings and banelings are 0.5 supply units.
Why can't hydras and roaches be 1.5 supply units?
|
On February 16 2011 04:29 Jermstuddog wrote:Show nested quote +On February 16 2011 04:08 Shadrak wrote:On February 15 2011 19:07 Phanekim wrote:
a lot of people here suggest maybe raising supply cap. i just wanted to ask what if they decreased the supply of the roach or hydra or whatever. i know its not well thought out but instead of just raising the cap i wonder what will decrease supply value of certain zerg units will do. The problem is that supply can only increase/decrease in whole numbers. Roach/Hydra take up 2 supply right now. That may be a little too high, but 1 supply would be way, way too low. Imagine a 200/200 Zerg army with literally double the amount of units. Would be way to strong. I have had an issue with this for a very long time, especially when you note that both zerglings and banelings are 0.5 supply units. Why can't hydras and roaches be 1.5 supply units?
That seems like a great idea...the only trouble is you can't really break up larvae appropriately. Perhaps instead of physically using the larva, making a roach/hydra could delay the production of an additional larva by a bit instead?
|
first of all, great post by lalush, and even though I didn't actually expected lots of 'smart' (non-ridiculous solutions tackling the main issue) responses, I've read a few.
I think the 300 supply cap is not the answer, it's just an easy way to look away the main issue, and will later generate more problems, if not the same ones but bigger
Even though comparing BW with SCII it's almost like apples and bannanas, some of the comparisons made here were really good, like whatthefat's approach on the short terms benefits of getting an early expo. In a time vs reward graph, this will look like a flat line down the axis with a big jump into a curve with not a great rate of increase, whereas the one we actually have will not have such a bg jump at the start but the increase rate will be higher at start.
EatThePath had some nice solutions too:
first solution - interesting, but we already have maps with 6-mineral expansions, and having an expansion with only 4 mineral patches sounds too risky, probably worth only in an extremely large game or in an extremely safe position, or only using it to get more mutas hehe
second solution - I actually liked this the most, why not having 3/8 of the mineral patches with only 1000 minerals?, this will affect mineral saturation a lot, will benefit getting expos early, and will make the need of expanding higher, bc not only your minerals will dry out quicker, but you will have oversaturation and well... u know, useless drones and stuff, contains will get interesting to break
third solution - not so sure about this, will actually increase the early benefit of expos, but this is kind of delicate bc zergs will benefit from this more, I'm not sure the time we need to wait on untill we actually get a benefit from an expo, needs to get so reduced
and fourth - needing 5 workers to fully saturate one mineral patch :S, i can't really see this..... mayb on some expansions, it this is made on the starting locations it will slow the early game quite a bit
I'm a Chem Eng, and I DO know how to read graphs (or plotts, I'm not really sure, english is not my first language hehe ), making some integrations of this graphs, changing the way the datta is plotted and plotting some 1st and 2nd derivatives of them will HUGELY help this topic, understand the issue better, and hence, giving more reliable solutions. Not everyone may be able to read and understand them though, and I won't do the graphs so... if Lalush or someone else gets really involved into this, here's THE way to go 
cheers!
|
On February 10 2011 02:31 FrostedMiniWeet wrote:
I've been hoping for a supply cap increase as well, at least to 250. Going back and playing Zerg in Broodwar was a fascinating experience, as the first thing I realized was just how friggin' huge my 200 supply army was. In SC2 I get to 200 supply and I'm like, what? That's it?
could it just be the fact that the zoom is more close in BW? :p
|
I just stumbled upon this thread, and am so glad it was written. I have had a similar idea that I was planning on writing about, but wouldn't have covered it as in-depth. good job lalush.
|
On February 16 2011 14:51 evanthebouncy! wrote:Show nested quote +On February 10 2011 02:31 FrostedMiniWeet wrote:
I've been hoping for a supply cap increase as well, at least to 250. Going back and playing Zerg in Broodwar was a fascinating experience, as the first thing I realized was just how friggin' huge my 200 supply army was. In SC2 I get to 200 supply and I'm like, what? That's it? could it just be the fact that the zoom is more close in BW? :p
Not sure, but the fact that units in SC2 take up more supply is what I'm thinking. Hydras, for example, are 2 supply instead of 1 as opposed to SC1...
BTW does anyone know where can I find the replay LaLuSh used in obtaining the 'real game statistics'?
|
On February 14 2011 21:40 LaLuSh wrote:Show nested quote +On February 10 2011 19:39 mahnini wrote:if i'm not mistaken your issue is with the fact that extra bases do not pay off until you are supersaturated and therefore early expansions do not pay off right away. i believe this to be a valid concern, however, you extrapolate on these ideas and conclude that Zerg’s play will be centered around saturating 3 bases as quickly as possible and launching suicide attacks at the opponents’ thirds. Protoss’ play will be centered around camping and delaying until they’ve reached their invincible end game composition on 3+ bases. Terran’s play will… no idea. this statement is made with the only backing being that at a certain supply all mining equalizes without regard for the time it takes to reach that supply. while it may be true that the most logical cutoff for obtaining extra bases stops at 3 due to supply constraints when fully saturated, this is done without the consideration of time to saturation. this is easily overlooked because of the omission of zerg mining data. the idea that zerg is required to rush to saturate 3 bases completely discounts zerg's additional production capability (aka the queen) which is obtained faster (build time-wise) and for less minerals (which also means faster game-time-wise) than extra hatcheries. a hatchery + queen spawns as much larva as 2.27 hatcheries (assuming you are constantly respawning larva from the hatchery + inject). the only comparison between bw and sc2 time to saturate is seen here in this graph we do not see the effect of chronoboost on saturation time, instead, we see data points of isolated mining per five minutes which, to a great extent, accentuates the midgame mining discrepancy. for example, if you're driving at 1MPH and i drive at 2MPH and we both do this for 10 hours, at the end i have a 10 mile lead; however, if we only drive for 0.1 hours i only end up with a 0.1 mile lead. this seems really stupid but it makes a big difference when the end quantity is more important than the rate because there is a threshold of usefulness. 1 mineral per second is completely negligible over 17 seconds compared to over 300 seconds; at the end of your so called midgame mineral discrepancy you probably end up with less of a difference in net minerals than you'd think. you also base your final conclusion that protoss will dominate on the 3 base limit, however, in a late-game situation, with the rate at which bases seem to mine out, i believe army production capability to be a much larger problem. due to new macro mechanics such as larva inject, chronoboost, and warp gates, much of the time-cost required to reinforce is moved to before the actual production (stockpiling larva, stockpiling chronoboost, and warpgates in general) and in late-game engagements it's usually a contest of who has the most durable army and who can reinforce the fastest, but when one race has both attributes then there is a (superficial) problem. Although I didn't word it as clearly as you did, I think the paragraph you quoted implies what you imply about "time to saturation". There would be no point for zerg to rush to 3 bases if they saturated at the same rates as the other races -- or if their armies were as durable.
i'm not sure i understand your point here, you're agreeing that zerg does in fact saturate faster than either terran or protoss, and is able to take a 3rd and saturate that faster and secure a relative advantage but still you think zerg can't compete economy wise?
With that said, I actually agree with you that the effects of mineral surplus surges are somewhat exaggerated by my graphs, but I still believe they pose a problem and influence the way the game is played out on different sized maps though.
they may but these opinions aren't backed by reliable data. the data that you've collected is exaggerated through the method of testing and even with a 5 minute duration there is only about a 500 mineral difference, 100 minerals a minute, under 2 minerals a second over about 8 workers at 17 seconds that's about a 300 mineral difference before reaching saturation. saturation point may come earlier but this isn't show by the data you have and isn't your point.
The reason Terran have been so stable and dominant on the Blizzard maps is exactly because of the fact that they have been able to negate the "time to saturation" issue by putting pressure and forcing the same amount of bases for as long as possible. They have been able to force "mining to be equalized" throughout games.
the problem with making a statement like this is that it is completely circumstantial and almost completely based on anecdotal evidence. i don't think many people would disagree that in XvZ to win you MUST keep the drone count down for the exact reason you mentioned (to keep zerg's economy at bay). so at this point you are blaming mostly the maps if i am not mistaken, but you've made the claim that mining is equalized throughout a majority of games without any evidence.
Furthermore, the issue you brought up about the races' different production rates can be used as an argument to explain why Terran will have a particularly hard time attacking across the new GSL maps. I believe that army durability will be the greatest deciding factor on the new big maps because races will have a much harder time killing eachother off. Mining will eventually equalize after which point there will be 1 attempt to break the more durable opponent through "quick" cross map reinforcements. After that you will no longer have a bulk of excess minerals to reinforce with a higher supply army due to better macro.
for the most part this is true but is taken out of context, and another wild conclusion makes its attempt at being driven home. there are a number of things that affect the effectiveness of armies: production mechanics, movement speed, cost, cost-efficiency, quantity, and so on. all of your scenarios are done assuming the midgame is nonexistent and different stages of map control are not considered.
The issue of reinforcement rates and production capability that you bring up, I believe to play a bigger role on smaller Blizzard sized maps as opposed to large ones. On large maps from wat I've experienced and seen so far, production capability seems most pronounced in a defensive capacity as opposed to an offensive one (except possibly for Protoss). I think zergs in ZvP will be able to defend pushes galantly in a cost inefficient manner. But then attack across the map and kill P? No.
you make a point to defend zerg production capability when in fact i was talking about protoss production capability. that isn't unreasonable, but as i mentioned it is quite a superficial analysis.
Next clash will probably be when protoss reaches 200/200 again, at which point there will be no bulk of excess minerals to abuse for a higher production rate.
again, you are making assumptions. understandably as a skilled player your word holds some weight, but this is clearly a case where you are simply theorycrafting.
That's why I think the 3base ceiling will force gameplay into the mould I described. Mining rate equalization will be a big problem on large maps.
your guess is as good as mine here as i have not seen that many games on larger maps. i would like to make the point, however, that you've yet to prove that mining equalization occurs in reasonable circumstances on small maps.
also, there's no question that a zerg gets their third base relatively quicker than either a terran or protoss, are you assuming the larger distances will enable a terran or protoss to expand more safely to their third? again, i believe there are a number of factors that you haven't considered: army speed, map control, time to saturation, relative army strengths at certain stages of the game, and so on.
If the game is going to be balanced for large maps and for a 300 supply cap, I think the right way to go is to make macro mechanics and especially the larva inject scale up through tiers and possibly through upgrades so a part of that "time to saturation" problem you mention gets addressed. Also the cap for amount of larva one single hatchery can hold lategame should probably be lowered.
i mention time to saturation not as an additional problem, but as a central point to the argument that you've not tested. again, you make the argument for a 300 supply cap when nothing that you've said really points to that being an efficient solution. every one of your scenarios are extremely late game with both armies being maxed on equal bases, of course this will cause a problem and of course raising the supply cap to 300 seems like a go-to fix given your fabricated scenario.
|
|
|
|
|
|
EPT![[image loading]](https://lh3.googleusercontent.com/_JjRV9kdI5FE/TVFHmIXBxkI/AAAAAAAABMY/9aA8r1_ijjw/ChronoANDmule.jpg)
