|
First, I'd like to start by agreeing with the OP and a lot of other posts that there is a clear, simple problem here, in that expanding is not as important in SC2 as it was in BW, and that affects gameplay and makes for less exciting, less fun, less balanced games.
On February 15 2011 06:34 EatThePath wrote: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.)
I agree almost completely with EatThePath, and I particularly like the second idea.
If well thought out, a change to minerals contained per patch could be great. For example, it could work that at roughly the same time for all races they start to saturate (obviously not exact, but the differences aren't particularly great), the minerals in their base come close to running out. This would first of all mean that all-ins with fully saturated bases would be less effective, as they could not sustain any attack and there would be a very small timing window to make use of the advantage.
It might promote all-ins as an alternative to expanding, still, as the differences in mineral count before and after the proposed change would be small and players might be more tempted to expand as certain strategies (eg 1-base Colossus) would be less viable, so all-ins would be more effective against this.
However, the solution to this is already being put into practice. With bigger maps, this incentive to all-in would be greatly reduced - one of the main intentions of the change, I believe.
The second major change of this to gameplay would be the intended promotion of increased number of expansions in lategame - I find the thought of just 3base Protoss rolling over everything, and additional expansions being taken at a slow rate as minerals run out in the 1st and 2nd base quite depressing, and I'm primarily a Protoss player.
Unfortunately, there is one main flaw in this plan in the lategame: The exact same problem would remain, except more expansions would be taken over time. What I mean is that people will only expand when their previous mineral patches run out - that those mineral patches run out faster doesn't magically make the problem disappear, only makes it less noticeable. The effect would be that the Protoss player would reach 3 bases and make a deathball, as normal. Instead of staying tight on those and pushing, winning/losing etc, however, Protoss would get more bases as their previous ones ran out, and transfer workers more often. The players seem to be expanding more, but they would stay on the same number of mining bases and it still wouldn't be a great advantage to have 14 bases over 3.
|
I don't particularly prefer the 300 supply cap solution over simply changing worker AI mahnini. I spent a lot of time thinking about changing the number of mineral nodes, changing the return rate of workers, changing the layout of mineral fields. Pretyt much thinking about the ways Blizzard like to artificially "fix" problems.
Everything I could think of fixed a part of the problem (for example forcing people to expand earlier), but didn't address the whole problem (max saturation ended up being reached even quicker).
The 300 supply cap suggestion isn't really one i'm gonna stand behind and wholeheartedly promote. Tbh, I'm not very enthusiastic about it and it was mostly just a lightly made suggestion that to me seemed to address the most immediate issue of SC2 capping itself way too early.
But you're correct in that it really does nothing at all to address the quick max saturation issue and that I ignore the effects on the early and midgame when suggesting it in the OP. It seems like a good fix for lategame scenarios -- not so much beyond that (without more changes to the game).
I don't really have any good solution. Didn't have one when I wrote the thread or the "300 supply cap" paragraph title followed by a question mark.
As for large maps vs small maps: I haven't proven anything I claim. I'm basing the arguments on experience more than anything. ZvX has in my experience always been about striking and dealing damage before end game mining equalizes (especially vs P). With these new large maps, from experience, end game mining will frequently end up equalizing and games start to follow the pattern I describe. Basically no matter how the game progressed in its earlier stages, end game mining equalization in ZvP will always induce panic in zergs.
For ZvT, I can't predict what will happen as confidently. I think T are much more dependant on keeping zerg's drone count at bay throughout the game. On smaller maps, zerg's would definitely panic as terran would start to close the gap and equalize end game mining. But on large maps, I don't see the effects being as extreme, due to how terran reinforce their armies. (of course all speculation).
|
This thread was a great read, and I pretty much agree with all of it. I honestly do not have much to add.
I do however, have a small issue with your graphs.
Your graphs measure the mineral gain per minute (if I am reading them correctly), and not the total mineral gain. I think in your MULE related graphs this is disingenuous, and the graphs show a spike in minerals that is more severe than what actually exists.
Because Chronoboost is available earlier than Orbitals, and because Orbitals themselves cost minerals (a small thing, but still relevant), Protoss should be mining faster and have a lead on total minerals mined over Terran up until the MULE comes out. This lessens the impact of the MULE spike afterwords, since Terran first has to catch up with Protoss in minerals mined before it can get ahead.
I think at the end of the day, Terran would still be ahead of Protoss on mining efficiency per base, but I think an adjusted graph would not look quite as severe, and would be a more accurate depiction of what is actually happening.
|
Hey LaLuSh since you're still posting on this thread I'm gonna ask you: Which game did you get the statistics from?
|
To be released all over the starcraft universe ^^ Gave me a head ache btw that's not a thing to read before bed :p
|
On February 17 2011 13:18 Ketara wrote:
Because Chronoboost is available earlier than Orbitals, and because Orbitals themselves cost minerals (a small thing, but still relevant), Protoss should be mining faster and have a lead on total minerals mined over Terran up until the MULE comes out. This lessens the impact of the MULE spike afterwords, since Terran first has to catch up with Protoss in minerals mined before it can get ahead.
This is the beauty of making a statistical analysis ... you can leave out some factors to get the results you want to show. In this case the cost of building a base isnt figured into any of the graphs and Terrans have by far the biggest cost early on (which is the time period the OP is looking at). So it doesnt matter that the Terran gains more minerals because he needs it. Should be easy to see ...
|
On February 17 2011 23:13 Rabiator wrote:
This is the beauty of making a statistical analysis ... you can leave out some factors to get the results you want to show. In this case the cost of building a base isnt figured into any of the graphs and Terrans have by far the biggest cost early on (which is the time period the OP is looking at). So it doesnt matter that the Terran gains more minerals because he needs it. Should be easy to see ...
Ah... what?
I think you'll have a VERY hard time trying to prove that Terran has to spend more early game on base. That Terran spends more early game on buildings is only your opinion, and nothing more, until you give some evidence to back it up.
|
Having less mineral patches favours T too much because of the mule because it will still increase the income from a base by a fixed amount, but if there are less patches, and the overall income from the base is lower, then the extra income from the mule will be larger in relation to the overall income of the base and so the mule will increase the income with a much higher percentage, making it that much stronger.
|
Leaving the gas out of the picture pretty much nullify the interest of the analysis.
But still this shows why T is so good early game since their mineral-only unit is so cost efficient...
|
On February 20 2011 02:32 Fushin wrote:
Leaving the gas out of the picture pretty much nullify the interest of the analysis.
But still this shows why T is so good early game since their mineral-only unit is so cost efficient...
Gas is not interesting as it scales linear with number of gas geysirs mined, no one oversaturates gas. seems like you pretty much missed the insight this analysis provides =)
|
On February 18 2011 06:47 hugman wrote:
Having less mineral patches favours T too much because of the mule because it will still increase the income from a base by a fixed amount, but if there are less patches, and the overall income from the base is lower, then the extra income from the mule will be larger in relation to the overall income of the base and so the mule will increase the income with a much higher percentage, making it that much stronger.
No it won't. The MULE is just a greedy wall street broker, it gets you extra money NOW but at the expense of financial stability later.
There is a difference, but it is not a game changing one, MULES will make you mine out way way faster too. Theoretically terran one base is the only thing that might get stronger, but since one base also means one MULE the difference is negligable.
|
VanGarde, this is bullshit. You're more stable financially if you have 5K in your bank and 1 mining patch than when you have 50 recourses and 7 mining patches.
Money Now >>>>>>>> Money Later. Always.
|
On February 20 2011 03:40 Schnullerbacke13 wrote:Show nested quote +On February 20 2011 02:32 Fushin wrote:
Leaving the gas out of the picture pretty much nullify the interest of the analysis.
But still this shows why T is so good early game since their mineral-only unit is so cost efficient... Gas is not interesting as it scales linear with number of gas geysirs mined, no one oversaturates gas. seems like you pretty much missed the insight this analysis provides =)
Different races need different amounts of gas early on; even different builds require different amounts of gas. Some races are more reliant on gas than others. So gas does matter, because this "financial analysis" should never ever be done without looking at the actual need for the respective race & build. If you add these things into the analysis - which is pretty complicated due to the ton of factors added - you can then start comparing the races in their actual "fairness". This "dry analsis" is pretty useless unless you want to try long distance mining or any other saturation trick, but you would not need any reference to the MULE for that.
|
On February 20 2011 04:03 ToastieNL wrote:
VanGarde, this is bullshit. You're more stable financially if you have 5K in your bank and 1 mining patch than when you have 50 recourses and 7 mining patches.
Money Now >>>>>>>> Money Later. Always.
I am not going to justify the way you phrased that post and the superficiality of your answer by an actual reply. But no, above silver league it is a bit more complex than that.
|
Umm... Cliffs please.
Nice graphs.
|
On February 11 2011 06:53 Kazang wrote:
Chronoboost is 20 seconds of 50% increased production speed every 44.44 seconds.
20 seconds of time for a structure under the effect of chronoboost is equal to 30 seconds of production, an increase of 10 seconds for every chronoboost. So for every 45 seconds of game time, you get 55 seconds of production.
(10/44.44) x 100 = 22.5%
Two probes take 34 seconds to build, so almost fit into a single chrono (30 seconds of build time), as you can't have two thirds of a probe (you either have a probe or you don't) it's reasonable to say you get roughly 1 extra probe per chronoboost.
...
In terms of pure mathematical design SC2 is almost perfectly balanced, yet retains a very definitive difference in "feel" between the various mechanics, it's a work of art in my opinion. IMO you're overstating the value of the chrono-boost and understating the value of the mule. The first big thing to consider is supply...each probe cost 1 supply while a mule cost 0 supply. In the early game this means protoss has to build more pylons which are expensive. In the late game this means terrain can have less SCV's, freeing up more supply for military units. The other consideration is saturation... Terran can over saturate a mineral patch while toss can't. So there is very little worth in chrono-boosting probes if you don't have a home for them to mine. In fact this is why you don't see protoss use that many chrono-boosts on units once warp gate research is available. On that thought, stacking mules on top of SCV's already mining means terrans don't have to worry about diminishing harvesting rates that with mules that happen when go from 8 to 16 to 24. Then you have to factor flexibility...terran can do quicker, safer, and more flexible 'transfer' operations to expansions with mule heavy economy than toss can.
It's just not fair IMO to compare chrono-boosted probes to mules as one measly mule will pretty much pay for the OC and the rest of the mules will be pure profit while toss needs to wait a while for their additional probes to kick in but after that there remains very little options for chrono-boosted probes because the available mineral patches start to disappear. If I as protoss had the option to trade chrono-boost for mules, I would do so in a heart beat, and I doubt you would do the reverse.
|
We certainly need more data on how gameplay will change on larger maps. I'm very interested to see how the GSL plays on Tal'Darim Altar ( http://wiki.teamliquid.net/starcraft2/Tal'Darim_Altar ). Making an even bigger map to play on would really be best. As bad as Steps of War has been rated by some, having such an extremely close map shows very well the problems in terms of balance created by such a small map. Essentially we learn more from our mistakes then our successes. By making a HUGE map and playing it, we will be able to learn a lot in the other direction. In doing so hopefully we can find some balance in between.
In terms of match up balance, so many of the power compositions rely on slow but powerful units. Larger maps might make a big difference in terms of gameplay. If T and P feel too slow using tanks and colossus and starts to transition into more mobile units, zergs standing army suddenly becomes much more powerful. They may not need an economic advantage to win.
This still leaves us with a game where the game play choices between expansion advantage and defensive advantage end at 3 bases.
However I'm not sure we can get around this. If we altered the worker AI so that it was like BW and players wanted 5 bases, I'm not sure SC2 units could handle being that spread out. As Day[9] has often stated, controlling terrain in SC2 is much harder then BW. SC2 units might not be made to handle 5 spread out bases.
|
On February 20 2011 04:30 VanGarde wrote:Show nested quote +On February 20 2011 04:03 ToastieNL wrote:
VanGarde, this is bullshit. You're more stable financially if you have 5K in your bank and 1 mining patch than when you have 50 recourses and 7 mining patches.
Money Now >>>>>>>> Money Later. Always. I am not going to justify the way you phrased that post and the superficiality of your answer by an actual reply. But no, above silver league it is a bit more complex than that.
Not really, unless they put out so much pressure you can't secure expos, but Terran is fairly limited in that regard with expos being fairly easy to take, moreso with tanks, and thirds being defending via PF + tanks.
Quite frankly, it doesn't matter so much when you get your minerals so long as you get them. I'd gladly mine out my expos twice as fast, for twice the rate of income. The increased income allows me to build a better standing force, allowing me to both secure and deny expos more easily, thus not only offsetting the fact my bases get mined out, but allowing me to prevent easy expos from my opponent.
Saying you're getting the minerals now is rather moot, since you're more than welcome to not ever use the mules, or only use one or two to be on par with Protoss' income. The fact is, to do so is sheer stupidity, because the money is going to be mined eventually, and the sooner you mine it the sooner you can put up expos. You're more than welcome to use all your mules and sit on thousands of minerals until your main is mined out, then spend those minerals as though you were still mining via SCVs. It's pretty obvious why you don't, however, because getting minerals now is going to always be better. I'd gladly start with ten thousand minerals at the start of the game and my main mined out. I have a feeling I'd be winning quite easily over people who want the "long-term stability" of having a low income which...
...caps out at the same minerals mined regardless. The "stability" of mining slower isn't stability, it's inferiority. Would you like to also only put one SCV on each patch to make sure your main lasts even longer? The argument that mining faster is *hurtful* is so absurd it's insane.
|
Amazing thread Lalush. Very happy made this research.
But couldn't this be fixed by changing the amount of mineral patches per base, and make changes in the distance from the main base?
For every mineral patch added, with same distance as on normal Blizzard map, the linear growth would be increased by 2 workers. Thus bases with more mineral patches will favor making more workers early before expanding, and favor having a higher total worker count, that what is currently possible?
I understand that this cuts into your supply, hence your suggestion of a 300 supply cap. But a change in the count of mineral patches per base will change the income from expanding, compared to before.
6 mineral patches instead of 8?
If it is changed from 8 to 6, races will need to expand faster to maintain the same income as before in the same time. This will likely favour the race that expands first in the early game, and give more value and incentive to expanding faster, and more.
So I propose making the standard perhaps 6 instead of 8 patches per mineral line. As currently there are so many minerals patches per base in Blizzard maps, that having more than 3 bases (and thus 18 mineral patches) is not worthwhile.
If it is changed to 6 mineral patches per base, you will need 4 bases to reach 18 mining mineral patches. How this balance the ratio of mineral and gas, I don't know. But maybe these are things to consider and test and see how it ends working out. Perhaps changing the amount of gas per base to balance it out. Or maybe the increased gas will help SC2, as currently many feel there are too little gas in this game, and at least as a Zerg player, I only feel I expand to secure more gas when I go into lategame.
|
That was an excellent read. Thank you for the hard work and I'm looking forward to Chapter III.
Instead of looking to decrease mineral patches, how about decreasing the number of minerals in each patch.
|
|
|
|
|
|
EPT
