EPTA Treatise on the Economy of SCII - Page 22
| Forum Index > Legacy of the Void |
I have received requests on how to try the model out: Search "Double Harvesting (TeamLiquid)" by ZeromuS as an Extension Mod in HotS Custom Games to try it out. Email your replays of your games on DH to: LegacyEconomyTest@gmail.com might have partnership with a replay website soon as well  In Game Group: Double Harvest | ||
|
Arhaeus
Romania13 Posts
| ||
|
DinosaurPoop
687 Posts
On April 15 2015 13:49 Honeybadger wrote: as retarded and obnoxious as this idea seems, as the years go by, I think it's the real answer. That, or maybe there's some litigious reason they can't just take the ideas we present them for intellectual property reasons or some other bullshit Or maybe because we all also assume we are the correct and true in supporting this change. | ||
|
Munk200
United States52 Posts
| ||
|
ejozl
Denmark3533 Posts
 A way that the current LotV system can lean towards this model IMO, would be to increase the time a base is on only half the Mineral patches. This can be done by moving the 750 Minerals removed from the small patches into the bigger patches. So 4 patches have 750 Minerals and 4 patches have 2250 Minerals. This way there's the same amount of Minerals pr. base as in HotS, but very rapidly the base will only go to half the income. So that instead of changing the AI and everything, this way it makes more scenarios where it would essentially be 8 workers pr. base for optimal mining, just like in OP's Model. It would help the turtle player as opposed to the current LotV system and it would make it more possible to get into lategame higher economy, instead of 'Rushing into Midgame&stay Midgame forever' LotV model. The way it would differ from OP's model is that you don't get an explosion of income the second you take a new base with 8 workers, but I fear that is just so hard to achieve, without changing more than Blizzard is willing to. | ||
|
Daeracon
Sweden201 Posts
| ||
ZeromuS
Canada13410 Posts
On April 15 2015 23:13 Munk200 wrote: So... Couldn't you just increase the time it takes to harvest the mineral node from 2.7 seconds to 3 seconds? Keeping everything else the same, wouldn't that accomplish a similar outcome of making it less efficient after you have 1 harvester per mineral node? Unfortunately not due to travel time from nexus to the mineral line, the workers will wait 1 full second to mine while the other dude is finishing. We tried a small change and it made no noticeable difference On April 16 2015 01:24 Daeracon wrote: I wonder if a poll about which economy system people would prefer blizzard to use or perhaps rather test would be a good tool to show blizzard what the community wants. Perhaps even a new thread for that as this is already a bit far along. The problem with a poll is that there is always a silent side to the debate. Unfortunately, I also don't want to just get people on a hype train vote for it and prove it doesn't work. What I would rather do is continue seeing support from those who want to support this model, and enter into open and respectful debate with those who don't. I personally think the best solution is going to be somewhere in the middle of what we like and what blizz wants as design goals. The most we can do is continue to discuss the benefits and positives that come out of the double harvest model due to how it breaks the worker pair. There might be a better way to break the worker pair in blizz's hands and outside of ours and like i said, some middle ground gets found :D We just want what is best for SC2 just like blizzard. And all we can do is push our ideas and our reasoning and hope that an open dialogue can occur. And by open i mean back and forth, with the position blizz puts themselves whenever they do something publically, I feel it would be difficult to have a real conversation in public on this idea. | ||
|
ZeromuS
Canada13410 Posts
On April 15 2015 19:34 y0su wrote: I think it's easier to tweak DM than DH. DM probably requires a slightly longer mining time to more closely resemble HotS (if that's even needed) but that's an easy change, it doesn't fundamentally change something about the worker behavior. DH has a few "bugs": Trying to micro workers to keeping them from bouncing to far patches can result in sending the current worker to bounce between harvests isntead. Workers that mine out a patch on their first harvest don't look for a new patch. These aren't game breaking and simple things like "don't micro your workers" solves the first issue (but feels wrong). I'm sure Blizz can fix workers to look for new patches if the current patch mines out mid harvest but, if they're going to fundamentally change worker behavior is there a better solution? *This inability to find a new patch could be related to harvesting 2x4 in "Double Harvesting - FRB Edit" mod e: I was pleased to see that playing the mod did ease some of my concerns regarding worker behavior. Using queued (shift) commands in conjunction with harvest didn't behave any differently. Maybe it's just cause I'm one of those anal guys that wants to make sure when I send a worker to harvest gas no minerals are lost. Using "return cargo" won't help if a worker has already completed half a mining cycle. The fact that the worker doesnt look for another patch might just be related to the way that lalush implemented the triggers in the mod which may be breaking the "scan" moment after a patch ends. thats all And I would LOVE for an option to return the basketed 5 minerals prior to going to gas, but there are a lot of times in HotS that people (pros and non pros) send a worker with minerals to the gas geyser. 5 minerals is really inconsequential to lose at all but the absolute highest levels of play and professionals can easily work around that. Not to say that being able to return the basketted income isn't bad. | ||
Plexa
Aotearoa39261 Posts
On April 15 2015 23:18 ejozl wrote: Okay so lets assume Blizzard will never use this model A way that the current LotV system can lean towards this model IMO, would be to increase the time a base is on only half the Mineral patches. This can be done by moving the 750 Minerals removed from the small patches into the bigger patches. So 4 patches have 750 Minerals and 4 patches have 2250 Minerals. This way there's the same amount of Minerals pr. base as in HotS, but very rapidly the base will only go to half the income. So that instead of changing the AI and everything, this way it makes more scenarios where it would essentially be 8 workers pr. base for optimal mining, just like in OP's Model. It would help the turtle player as opposed to the current LotV system and it would make it more possible to get into lategame higher economy, instead of 'Rushing into Midgame&stay Midgame forever' LotV model. The way it would differ from OP's model is that you don't get an explosion of income the second you take a new base with 8 workers, but I fear that is just so hard to achieve, without changing more than Blizzard is willing to. Funnily enough, the best way to break the three base cap with minimal changes is to increase the supply cap  I don't think this is the best solution, but it achieves a similar purpose. It means you can keep the same worker:army supply ratio, but have a larger army. This means instead of 64 workers being "optimal" it could be 86 (with total supply ~300) which means that you're now capped at four bases (32 nodes) instead of three (before additional bases do not increase income). Since mains are likely to mine out by the time you're taking a fourth, that necessitates taking a fifth. And hello, we're seeing the desired expansion dynamic. Of course this has other implications, but those can be balanced as needed. | ||
|
ZeromuS
Canada13410 Posts
On April 16 2015 01:54 Plexa wrote: Funnily enough, the best way to break the three base cap with minimal changes is to increase the supply cap I don't think this is the best solution, but it achieves a similar purpose. It means you can keep the same worker:army supply ratio, but have a larger army. This means instead of 64 workers being "optimal" it could be 86 (with total supply ~300) which means that you're now capped at four bases (32 nodes) instead of three (before additional bases do not increase income). Since mains are likely to mine out by the time you're taking a fourth, that necessitates taking a fifth. And hello, we're seeing the desired expansion dynamic. Of course this has other implications, but those can be balanced as needed. Its the third out, but we still think that if we remove the theoretical cap on minerals through worker pairing you'll see the most diverse gameplay. | ||
|
Plexa
Aotearoa39261 Posts
| ||
|
Grumbels
Netherlands7032 Posts
I felt it was an interesting discussion, more instructive than I thought it would be. I'm so used to promoting these sort of BW economy models that it becomes difficult for me to think critically about them, so I think I've long since lost the ability to come up with meaningful counter arguments. But the people on the show at least had a fresh perspective, even if I would be more likely to agree with Zeromus than with them. + Show Spoiler + HuK said that by encouraging players to spread out workers you would lose out on some cool moments where a disruptor blows up a mineral line with lots of workers there. He also thought that too high economy would be hard to manage. Nathanias said that spreading out workers reduces some of your defender's advantage since the scv's are so useful defensively. Both of them liked the pressure put on you to constantly expand from the LotV economy and liked that the bases ran out more quickly. Nathanias also added that the resource scarcity with the LotV model prevented three base turtle builds, which he was happy about. Morrow said he wanted a high base count to translate into a very high economy game with mass production, like in many scenarios in BW and was more positive about the proposed change. Zeromus also said that he heard about LotV games where bases would run out so quickly you would lose the ability to use infrastructure you needed to secure your last base. Zeromus wanted to remind people that you could reduce minerals per patch with double harvesting as well, so that it was possible to find common ground between proponents of the DH & LotV model. A goal was to create a LotV extension mod with double harvesting in order to test the system. HuK felt conflicted about this idea, since he thought it would lead to ambiguous and confusing feedback. | ||
|
ZeromuS
Canada13410 Posts
On April 16 2015 02:41 Grumbels wrote: I was listening to Zeromus on Remax and I wanted to list the arguments that I heard there, (I haven't read this thread yet, please tell me if it's superfluous), since not all of them were mentioned in the OP. I felt it was an interesting discussion, more instructive than I thought it would be. I'm so used to promoting these sort of BW economy models that it becomes difficult for me to think critically about them, so I think I've long since lost the ability to come up with meaningful counter arguments. But the people on the show at least had a fresh perspective, even if I would be more likely to agree with Zeromus than with them. + Show Spoiler + HuK said that by encouraging players to spread out workers you would lose out on some cool moments where a disruptor blows up a mineral line with lots of workers there. He also thought that too high economy would be hard to manage. Nathanias said that spreading out workers reduces some of your defender's advantage since the scv's are so useful defensively. Both of them liked the pressure put on you to constantly expand from the LotV economy and liked that the bases ran out more quickly. Nathanias also added that the resource scarcity with the LotV model prevented three base turtle builds, which he was happy about. Morrow said he wanted a high base count to translate into a very high economy game with mass production, like in many scenarios in BW and was more positive about the proposed change. Zeromus also said that he heard about games where bases would run out so quickly you would lose the ability to use infrastructure you needed to secure a new base. Zeromus wanted to remind people that you could reduce minerals per patch with double harvesting as well, so that it was possible to find common ground between proponents of the DH & LotV model. A goal was to create a LotV extension mod with double harvesting in order to test the system. HuK felt conflicted about this idea, since he thought it would lead to ambiguous and confusing feedback. https://www.youtube.com/watch?v=I5GGDb1iwB8 thanks for the summary. We have taken huks comments, collated them for the most part and aim to address them in a new article next week. I need the weekend to write it | ||
|
People_0f_Color
177 Posts
Can you explain that chart? I'm confused--my understanding was the only difference between Double harvesting and Double Mining models had to do whether the worker "baskets" half the minerals it mines. From my understanding, this shouldn't impact the mineral curves--it simply solves the problem of pulling workers defensively not being the equivalent to pulling 2 HOTS workers. Can you clarify why the DH and DM curves look different on that graph? | ||
|
ZeromuS
Canada13410 Posts
On April 16 2015 04:20 People_0f_Color wrote: http://www.teamliquid.net/staff/ZeromuS/Economy/2harvvs2minevshots.png Can you explain that chart? I'm confused--my understanding was the only difference between Double harvesting and Double Mining models had to do whether the worker "baskets" half the minerals it mines. From my understanding, this shouldn't impact the mineral curves--it simply solves the problem of pulling workers defensively not being the equivalent to pulling 2 HOTS workers. Can you clarify why the DH and DM curves look different on that graph? Due to the way the AI works in double harvesting, the workers will bounce a bit more earlier than in double mining. The nice side effect is that double harvest is more linear and consistent an income curve and its also less harsh an overall income increase as you reach the 16 worker HotS softcap for mining | ||
|
purakushi
United States3302 Posts
On April 16 2015 03:53 ZeromuS wrote: thanks for the summary. We have taken huks comments, collated them for the most part and aim to address them in a new article next week. I need the weekend to write it Please do a good job (like I know you will) with your address to HuK's comments. I respect HuK's opinion, but I have a feeling Blizzard is just going to do something like take his comments as what pros say about the matter and not do anything in the end. Still watching the episode, though. Anyway, you (all) are doing an awesome job with getting this out there! I know you have been trying to dispel some misconceptions about the system and its similarities/differences from BW/HotS/LotV. Also, just how the economic system should come first before all of this balance talk. Comeback potential is improved, too. | ||
|
Hider
Denmark9447 Posts
Morrow said he wanted a high base count to translate into a very high economy game with mass production, like in many scenarios in BW and was more positive about the proposed change. Zeromus also said that he heard about LotV games where bases would run out so quickly you would lose the ability to use infrastructure you needed to secure your last base. I liked what Morrow said here. LOTV currently lacks the late-game income rate since your busy taking bases all the time, and thus find your self in a constant midgame. I wonder whether these 3 changes could help with that (1) Increase mineral income --> early game + higher income late game (would adress the "I am broke all the time"-feeling). (2) Change mineral pathes to 2250/750 from 1500/750 --> You obtain higher income as you get more bases Late game --> Much higher income rate. Snowball effect --> Reduced as you have more income you can fall back on if you lose a base. | ||
|
ZeromuS
Canada13410 Posts
On April 16 2015 05:35 Hider wrote: I liked what Morrow said here. LOTV currently lacks the late-game income rate since your busy taking bases all the time, and thus find your self in a constant midgame. I wonder whether these 3 changes could help with that (1) Increase mineral income --> early game + higher income late game (would adress the "I am broke all the time"-feeling). (2) Change mineral pathes to 2250/750 from 1500/750 --> You obtain higher income as you get more bases Late game --> Much higher income rate. Snowball effect --> Reduced as you have more income you can fall back on if you lose a base. No matter what the "forever mid game" is a result of the half patches. And if you just have the big patches be too large, you will end up with a 24 mineral node cap being reached easily and maintained for a new four base turtle in LotV once the meta settles. | ||
|
rigginssc2
18 Posts
If Blizzard was to change the economy they could obviously change the worker AI and not have to use an editor "hack" like you guys to get at either DM or DH. Given that, why wouldn't they instead adopt something that would lower worker efficiency with the number of workers added to a node? For example, the first worker gets 100% efficiency like now and like you propose. When a second worker is added, and they are paired, instead of them both getting 100% like now, maybe they both get 80% efficiency. No need to double harvest, just become less efficient. I throw that out as a single possibility and not as a proposal per se. The real point is, they can change the AI. Your real problem seems to be with the 2:1 efficiency. Do you really care that it has to be 1:1 or can it be some quadratic fall off as workers are added? Something else? What other ideas are there if you allowed yourself to change the AI? Remember, you aren't pitching a solution for a SC2 MOD, you are pitching a solution for Blizzard Devs to implement. | ||
Teoita
Italy12247 Posts
| ||
|
ejozl
Denmark3533 Posts
And if you just have the big patches be too large, you will end up with a 24 mineral node cap being reached easily and maintained for a new four base turtle in LotV once the meta settles. But you can never remove the cap. What's the difference between 8 patches with 1 on each vs 4 patches with 2 on each. If it's the same number of bases required and the same income? | ||
| ||
