|
While I understand your graphs and your point, I feel like you're missing some points for practical gameplay.
Your argument is that quick expanding is useless if you don't have the workers to saturate your currently available bases, since mineral gain per worker is essentially linear until it almost hard caps at 24 workers.
I get that much, but the main advantage of getting an expansion is that you can start pumping out workers twice as fast, in comparison to before, which will speed up your mineral income by quite a lot. What your graphs will suggest is that you might just as well rally point your expansion workers to your main mineral field as long as it isn't fully saturated (I could actually see some terran players building an orbital in their base as part of fast expand walloff and pumping workers from both orbitals into their main mineral field until it's fully saturated, at which point they lift it up and move out to take their natural  ).
The real reason you'd expand past three base is of course the gas geysers as they are the main resource of importance in the endgame.
Tell me if I've missed something, but I feel like your point about mineral gain rates being linear is meaningless for the practical game since it doesn't take into effect that you can mass up on workers a lot faster with multiple orbitals, and of course the additional gas which the graphs don't have anything to do with.
The only relevance I can see with your macro analysis is for one base play and why it can be so strong.
Edit: I guess you can also TL;DR it by saying that if you want to go an extremely mineral heavy composition like m&m&m, you can pretty much stick to 2base, and might even be better off doing that.
|
Fabulous post. I feel like my understanding of the game significantly increased reading it. Which in sc2 directly translates to ladder rating!
I especially liked mineral surplus and its relation to imbalance.
|
On February 10 2011 08:26 Dalavita wrote:While I understand your graphs and your point, I feel like you're missing some points for practical gameplay. Your argument is that quick expanding is useless if you don't have the workers to saturate your currently available bases, since mineral gain per worker is essentially linear until it almost hard caps at 24 workers. I get that much, but the main advantage of getting an expansion is that you can start pumping out workers twice as fast, in comparison to before, which will speed up your mineral income by quite a lot. What your graphs will suggest is that you might just as well rally point your expansion workers to your main mineral field as long as it isn't fully saturated (I could actually see some terran players building an orbital in their base as part of fast expand walloff and pumping workers from both orbitals into their main mineral field until it's fully saturated, at which point they lift it up and move out to take their natural ). The real reason you'd expand past three base is of course the gas geysers as they are the main resource of importance in the endgame. Tell me if I've missed something, but I feel like your point about mineral gain rates being linear is meaningless for the practical game since it doesn't take into effect that you can mass up on workers a lot faster with multiple orbitals, and of course the additional gas which the graphs don't have anything to do with. The only relevance I can see with your macro analysis is for one base play and why it can be so strong. Edit: I guess you can also TL;DR it by saying that if you want to go an extremely mineral heavy composition like m&m&m, you can pretty much stick to 2base, and might even be better off doing that.
...not to be rude, but you missed a lot, apparently. Minerals mined per worker drop off sharply going from 2 workers per patch to 3 workers per patch, which begins to apply as you exceed 16 workers per base on minerals. This point was made quite clearly with just about every graph the OP posted, to say nothing of the actual content of his analysis. I understand that people miss things when there's a lot of text, but I thought they usually did better with the pretty pictures.
|
Btw, how is this not spotlighted yet?
|
Oh right now I'm so proud I made that 300 supply thread where Day[9] stated he liked my idea back then in beta 
Interesting read, I still am pro 300 Supply, should also maybe give more multitasking because splitting up is easier. 6 Supply units in 200 Supply situation can't be split up as well to fight on multiple fronts. But that's just speculation.
But of course maybe you could be fine with 4 base and close to 75 workers, which would be a whopping 20 more than you calculated with, basically another saturated base. Enemy will have 3 Colossi more that way though, or more accuratly you will be missing ~ 10 Roaches or 40 Banelings or something ^^
|
Always enjoy reading LaLush's posts and threads. They're always well thought out and written with plenty of pictures, even if I don't agree with the analysis.
The points made are quite interesting and I agree very much with his analysis on the mineral surplus leading to the imbalance or illusion thereof. The incredibly strong Terran presence of mass MM, along with the 4gate rushes are very likely the exact product of the macro mechanics, which mostly, if not completely affect mineral gain. It's interesting to note that the strongest 4gate and MM pressure/cheese builds all really only require one gas to work well. The large mineral surplus you get in the early game really blows the strength of those early game mineral-heavy units out of proportion.
However, I have to disagree with the 3base bottleneck proposed by OP. Although I have found (as Protoss) that I really only need 3 bases mining at any given point in the game to really be able to macro fully, sometimes I still get limited and capped by my gas needs (especially since I'm bad and tend to lose my HTs unnecessarily :[ ). As been stated before, I think any bases taken after the third will really mostly be for the gas. It's not uncommon for fourths and fifths bases to have nothing other than 6 workers mining gas and maybe a couple on minerals.
I think the best and perhaps easiest fix to the mineral surplus would be to reduce the number of mineral patches at the naturals. Changing income of the mains would screw up the metagame too much I feel, but lowering the number of patches at naturals might shift the play slowly towards a more macro-oriented style, if that is indeed what we want, as a community, for SC2 gameplay.
EDIT: On the 300 supply cap issue, I think Blizzard actually originally wanted to increase the supply cap. They said so in one of their interviews during beta I believe. However, they stated that they chose not to due primarily to technological limitations. I know a lot of computers already have problems with max supply armies with the current cap of 200/200. SC2 takes up a crapload of memory to run as is. I can only imagine how intensive it would be on your computer if the supply cap was at 300. It'd be awesome for the cap to be raised, but it's simply not practical from a tech standpoint.
|
OP reiterates many facts already known about mining incoming ... including the mule. Its good that a higher level player made this post so that no longer will we have to bear through the cesspools that were past MULE IMBA threads and so on. The replies here are pretty good as well.
Regarding the conclusion regarding SC2 itself and 1 base play, something I never really thought about - 5rax for terran, 4gate for protoss, and heck, you can sustain roach production with only 17 drones (14min+3gas). Actually, I'm not surprised. Its obvious blizzard built and balanced the game around 1 base play, especially when you look at the layout of most single player missions, and especially especially considering blizzard's small map mantra. It makes it more obvious that the mule, larva inject, and chronoboost were direct results of this and will only serve to break the game in the long run when bigger maps start to make their debuts.
For zerg though, its impossible to make any kind of comparison from BW to SC2. Income rates, production rates, and even unit cost/supply and health/survival rates are all factors relating to economy and they are so radically different from BW that you need to start from scratch. The only similarity is concept of the hatchery being the only production building, and the similarities end there thanks to larva inject.
|
Thinking about it:
couldn't the mapmakers just help this problem?
You could lower the mineralpatches per expansion by maybe one. That way distributing your workers on more bases will be more beneficial. Through this you could make expanding quicker and more often more beneficial than it is now!
What do you think?
|
Thank you... best OP I've seen on TL since I joined.
I agree with your thoughts and they add a wonderful new perspective.
The 3 base ceiling is real.
I think 300 pop cap would be awesome. Terran would need a DPS unit to replace marines in the super late game.
|
If I’ve learned anything from observing Blizzard the past year, it’s that it is largely pointless to suggest anything that would require alterations to the game engine itself. Whether it be about moving shot or built-in delay for firing projectiles (tank AI). If it can’t be achieved through the map editor, it likely won’t be “fixed” in the way you imagined. Thus I’m not even going to attempt to discuss changing worker AI.
i'd just like to say that all of those things are easily done in the map editor as i have done them myself in SC2BW mod.
|
On February 10 2011 09:04 Dominator1370 wrote:...not to be rude, but you missed a lot, apparently. Minerals mined per worker drop off sharply going from 2 workers per patch to 3 workers per patch, which begins to apply as you exceed 16 workers per base on minerals. This point was made quite clearly with just about every graph the OP posted, to say nothing of the actual content of his analysis. I understand that people miss things when there's a lot of text, but I thought they usually did better with the pretty pictures.
Note that I said essentially linear, I guess I could have phrased it better since looking back at the graphs, the mineral per additional worker after 16 workers is close to halved, but the main point of mine was the point where mineral income almost flattened at 24 workers whereas it's fairly steady before then.
I guess it even further shows that one base pushes can be strong because you're at very high mining efficiency with just 16 workers, but my points regarding expanding and gas usage still stand.
|
Hey how about they adjust the speed of workers. Terrans have a huge influx in minerals because of mules... make their workers mine slower. Since zerg is rushing for 3 bases to gaina macro advantage make their workers mine faster... This trying to stabalize the income that things like chrono boost, mules, and larva distrupt. To avoid descrepencies in gas they could only change the time to mine minerals not gas. Also does anyone else notice on the graphs that terran gets way more minerals in general because they can use mules on top of fully saturated bases....might be nice if that was not the case. This is probably why all marrine play like marineKing are so popular because of the influx in minerals for terrans.
|
Another point which I don't believe you have addressed and would be difficult to analyze is mules and gold bases.
Taking Xel'Naga Caverns as the prime example. Terran will already have 2 OCs up and running as they are planning on taking their third (at the gold). Once the terran begins to build his command center to be placed at his third the terran may decide to save up as much energy on his 2 OCs so that he can dump 4+ mules when the CC finishes (and then probably turns into a Planetary Fortress).
The mineral surge from 4 mules (and the continue mules drop from the 2 OCs) creates a huge burst of income for the terran, which I'm sure if someone does the math the income would equal around the same as 2 completely saturated "blue" mineral expansions.
Gold expansions are a nice idea and are usually located in 'harder' to defend locations, but with OCs dropping mules and by placing a planetary fortress at the location (making it nearly impossible to kill at all) is this something that is imbalanced?
I don't think blizzard really saw this massive economical advantage when they realeased the game because in isolation the idea of a gold base seems sound. The problem is when combining mules and a 'non-killable' Planetary fortress to the situation, is the combination of these things a little too powerful?
|
Lalush,
Thank you very much for this article! I enjoyed the abundance of graphs, analysis and objectivity! I have to agree with all of your points. I think that, overall, your analysis of problems in SC2 is incredibly accurate! The effect of worker/base counts reaching a ceiling is incredibly critical to the overall flow and dynamic of the army's supply count!
One idea I have to limit the effectiveness of 16 worker counts would perhaps increase the length of the mining time by a small fraction so that 2 probes can't stack perfectly on one patch? I believe this would make having 3 base 54 workers less effective than 4 base 54 workers.
Thoughts anyone?
p.s. thanks again for this awesome contribution!
|
Nice to see a high level player write something epic like this. :D Good read.
|
the 3 base vs 3 base thing blew my mind...... i've been panicking every time i get out expanded without reason, i suppose zerg benefit from extra expos in that they can build more units at once, but it seems they dont benefit economy wise very much.
|
Extremely interesting article, really shows the extra gas is really the important resource in expanding. I'm definitely interested to see what comes of this information and if there will be any significant changes to this.
|
I really liked this post.
However, I don't know if upping the supply to 300/300 would be that great.
Many units would have to be nerfed/buff since they are really strong in critical mass or really weak against critical mass units.
I'm mostly talking about the toss units, we already fear toss 200/200 more than anything else. I think it could be intressting in TvZ. But as it looks now, toss would gain A LOT.
Think about how many more void rays and colossus the toss death ball would have. NOT FUN!
|
You dont need 300 supply cap. You can just adjust the gas price of units to make more than 3 expansions beneficial. Or something in those lines.
|
Amazing writeup! Just have to give a man props for putting in this much research into it and drawing conclusions from there.
|
|
|
|
|
|
EPT
