EPTLotV Economy: Worker Pairing - Page 5
| Forum Index > Legacy of the Void |
|
tili
United States1332 Posts
| ||
|
OtherWorld
France17333 Posts
On March 07 2015 01:30 [Phantom] wrote: My question would be... Even though this would open up other possibilities that the current maps don't have, wouldn't it also restric the maps in other ways? Like the maps needs to have this design because zerg needs this number of bases and protosss this, and terran isn't good at defending muktilple positions and so on. I know this article is about the economy itself, but wouldn't it be possible that it has hum..unintended consecuencses? Also wouldn't harrass units would need to be rebalanced completely? I know you mentioned in your post other changes would need to be made, but those changes are exactly why this isn't happening.. There would need to be a lot of changes, and although it would be "better" ask yourself blizzard really benefit for doing all those changes? That's an important question. I hope when you work with blizz you will be able to give some ideas to them, and they give you some insight of why some things are doable and some aren't. To be honest it's the mapmaker's job to make maps that are not out of phase with the game design, not the opposite. | ||
|
Big J
Austria16289 Posts
On March 07 2015 01:30 [Phantom] wrote: My question would be... Even though this would open up other possibilities that the current maps don't have, wouldn't it also restric the maps in other ways? Like the maps needs to have this design because zerg needs this number of bases and protosss this, and terran isn't good at defending muktilple positions and so on. I know this article is about the economy itself, but wouldn't it be possible that it has hum..unintended consecuencses? Also wouldn't harrass units would need to be rebalanced completely? I know you mentioned in your post other changes would need to be made, but those changes are exactly why this isn't happening.. There would need to be a lot of changes, and although it would be "better" ask yourself blizzard really benefit for doing all those changes? That's an important question. I hope when you work with blizz you will be able to give some ideas to them, and they give you some insight of why some things are doable and some aren't. The only way to keep the balance and maps close to what we have is that they aren't changing the economy to begin with. But they do so greatly. And the start into the game which completely overthrows build orders to begin with. The question is not, which are the greater changes at this point. Both will have dramatic effects on the balance and the maps. The question is which end result do we want from economic changes. I'm afraid that blizzard's solution will neither lead to much improved gameplay (1year in when the game gets somewhat figured out; at the beginning every change will make the game look fresh) while already destroying the game that we know. | ||
|
eneyeseekay
239 Posts
Video demonstrations that show a timelapse of a standard vs non-pairing mining configuration would illustrate the concept much better, but aside from that, this was a great writeup! If you and the other mapmakers are serious about pushing this concept, show Blizzard the proof in action, and try to get some reputable players who believe in it as well on board. Show it of on Reddit, try to get it as much exposure as possible, and maybe, just maybe, something might click. Also, to respond specifically to your point about being spread out, I'd just like to say that it sounds like an overall improvement, for all races. At the center of each race, there are strengths that this spread-base play really enhances. Take Warp-ins for example. If your main target is almost always an opponent's third or main and either is taken out, you will most likely issue another Warp-in and move on the the natural, in which case the game ends. Sure, MsC recalls can salvage a botched Warp-in, but to be honest, the crowd reactions to that kind of play is usually met with a bad taste. If you are Warping-in at an opponent's 5th, you're slamming dents in them instead of scooping out their vitals. Suddenly, you're getting ahead and not outright killing an opponent with a successful Warp-in attack. This kind of thing can be applied to things like drops and air harass, run-bys, and also Nydus play. It sounds like it would create less "all or nothing" scenarios, and getting outplayed would be something that would be much more easily identified. A side effect of all this could very well be that game lengths become extended, but I don't know..Long, scrappy, "neck-and-neck" games are exciting and are typically regarded as some of the greats. | ||
|
Big J
Austria16289 Posts
On March 07 2015 01:45 OtherWorld wrote: To be honest it's the mapmaker's job to make maps that are not out of phase with the game design, not the opposite. I think the game should be designed with certain features in mind. Map features are just a design aspect of the game too. Telling mapmakers to make boring maps because the unit design doesn't allow for better ones is just as bad as letting all maps through and creating a game around it that plays the same regardless of map features. | ||
|
nath
United States1788 Posts
On March 06 2015 09:59 The_Red_Viper wrote: Yeah i still hope they consider this, but it's blizzard after all right ... I actually think highground advantage would be almost useless in sc2 due to the pathing, it would help a little bit in easier defending, but usually armies will be able to fire all at once after moving, so it won't change enough to be worth it probably. ever play terran? | ||
|
Insidioussc2
Germany96 Posts
It probably does some unexpected balance changes, but i think the game balance should be based on game design. And in that perspecitve "No worker pairing" is imo the better design as it benefits a more risky and stategic style of play. So I hope Blizzard gives it a try. | ||
|
Gwavajuice
France1810 Posts
On March 07 2015 00:37 Uvantak wrote: Hehe fixed Yes and no, bases are very valuable as they are now, the problem is that you can't take advantage of new bases without going over 80 workers because of worker pairing, with the changes I'm pretty much not forcing players to expand, I'm rewarding them for doing so, and by that I'm talking about highly increasing their income by allowing their limited workers to be more efficient than his opponent. Actually It would have helped immensely, like I clearly stated in the OP. soO had around 90 workers entering the late game, in the little snippets where the observer shows soO's bases you can see that even when he had 90 workers and around 8 total bases he wasn't getting an income advantage over his opponent because many of his mining bases weren't saturated up to 16, which is the point where you can start sending workers to other expansions and get a supply positive/efficient return from them, as I have said previously, it doesn't matter of you have 12 or 24 bases, if you have 48 workers and your opponent also has 48 workers but on 3 bases both players will mine have the same mineral income. Of course he did had an income advantage seriously you can't possibly say otherwise, oh yes it was gaz income soO wanted but you discarded gaz mining from the problem  . anyway soO had the biggest bank ever and it didn't save him.Removing worker pairing wouldn't have changed how Yoda slowly built his late game mech, if anything he could just have built faster. soO couldn't break him, he thought that having a max income to rebuilt his air army 3 times in a row was a cool idea but it was not, your could have tripled his income, with his strategic choices he would have lost nonetheless. Anyway, there is so much things to say. First your graphs are off. sorry by you need to rework them to include mineral depletion. And this will only be possible if your X axis somehow include time. It's not the same thing to build a 4th base at 12 minutes and a 4th base at 20 minutes. You can't just say "DH model increase 4th base benefit by 7.5% at 35 workers" This model needs to be dynamic. For instance stadard model lead to stay longer on 3 base if you only have 48 drones on mineral but 48 drones on 3 mineral lines will clear the patches faster than on 4 if I'm not mistaken (according to your graph it's pretty much the same for every model : 12 drones = 480 Min/min => base empty in 25 minutes to compare with 16min 45 with 16 drones) I'm not trying to be a smart ass here but as it now, the main reason why a terran need to get a 4th is not the mineral income which is good anyway thx to mules (I ll get back to this) but the fact that his main will be empty at 15 minutes or so. Second and most importantly, simply forgeting about gaz makes the whole thing much less relevant. Minerals are only important when you build your economy but quickly enough it's the gaz that is the most important thing. Why zergs want a 4th? for minerals or for gaz? What do Protoss want? flood even more chargelots or get templars? Third your graphs doesn't take mules into account, and that's a big deal actually, it would be cool to have the same graphs for P and Z and one for T with as many mules as bases in the model. But maybe I'm in the details here. That's why I won't talk about the fact a scv building a CC is not mining for 100 seconds and that a drone does disappear each time a hatchery is build, I won't talk about worker travelling time either All that said, my point is I see what you're trying to do there but I think there are plenty of shortcuts you take that make the whole thing unaccurate economically. I think a better method might be to think about a target economy (like 2000 minerals 720 gaz per minute for example) and see how each model gives a different optimal solution to reach that point (with a somewhat dynamic model that takes in game time into account) and how it would actually impact the game. And then try to see what real benefit it would create when compared to Blizz current solution of lowering ressources per base. | ||
|
Cyro
United Kingdom20278 Posts
| ||
|
ANLProbe
667 Posts
| ||
|
coolman123123
146 Posts
On March 06 2015 14:53 Uvantak wrote: Because I vespene geyser positioning asymmetries I left all vespene incomes out because I would need to write a complete annex talking about it, but for a complete cardinal or vertical vespene positions you will want to have 2 workers per geyser mining for a total of 84 gas/min per geyser, if you added a third harvester this worker would mine at a lesser efficiency than the first 2 (30 gas/min instead of 42) meaning that it would not be supply efficient, but as I said vespene geysers are complex and asymmetric regarding the income they give based on their position in relation to the town hall, so in general they are a pain to work with. + Show Spoiler + ![[image loading]](http://i.imgur.com/Bqohnru.png) For the most part I just tried to disregard vespene income as much as possible, and round the numbers of workers and incomes to make everything easier to understand sacrificing as less as possible the precision of the calculations, but as I said in the OP if anyone wants to get into the work of counting the exact vespene income for each of the 48 possible geyser spots is free to do so, I would love to have that math. Well at the time of writing I was thinking of a "normal" TvZ of Bio/mine/thor vs Ling/bling/muta/ultra comp, this would be only theory, I'm really not expecting players to actually reach said amount of bases, It could perfectly happen, I'm only not expecting it, as you said it would be a multitask nightmare for both players, but for spectators, it would be heaven. Maynarding, is FAR from being hard, I do not know if you are doing this on purpose, but if you have 14 workers on your main what you would do is to take 7 and send them to your natural, any income loss would be offset by the higher worker efficiency, then you would continue to make workers normally. Regarding the multi-base scenario, that's something that already happens when you are in that many bases, I don't understand your point, this is Starcraft not Farmville or a Moba, it IS supposed to be hard because you are doing it yourself, not an automated AI, if you don't agree with that there is not much that I can say to you. Here is the fundamental difference, I don't think the game should be any harder with regards to macro. Maynarding is not hard but it can be a chore to maintain 16 workers on each base, trying to keep each base as close to 8 as possible is that much harder because you still want to go over 8 workers (I assume), but you also want to make sure that each base has the same amount of workers roughly, whereas with 16 workers you simply move your rally point and stop populating that base once you hit that point, and you do not have to worry about maintaining equal numbers on each base. This change would increase the skill cap indeed, but in a way that does not make the game more interesting or fun. Luckily this change has 0% chance of actually going through. | ||
|
Uvantak
Uruguay1381 Posts
On March 07 2015 02:42 Gwavajuice wrote: Of course he did had an income advantage seriously you can't possibly say otherwise, oh yes it was gaz income soO wanted but you discarded gaz mining from the problem . anyway soO had the biggest bank ever and it didn't save him.Removing worker pairing wouldn't have changed how Yoda slowly built his late game mech, if anything he could just have built faster. soO couldn't break him, he thought that having a max income to rebuilt his air army 3 times in a row was a cool idea but it was not, your could have tripled his income, with his strategic choices he would have lost nonetheless. Anyway, there is so much things to say. First your graphs are off. sorry by you need to rework them to include mineral depletion. And this will only be possible if your X axis somehow include time. It's not the same thing to build a 4th base at 12 minutes and a 4th base at 20 minutes. You can't just say "DH model increase 4th base benefit by 7.5% at 35 workers" This model needs to be dynamic. For instance stadard model lead to stay longer on 3 base if you only have 48 drones on mineral but 48 drones on 3 mineral lines will clear the patches faster than on 4 if I'm not mistaken (according to your graph it's pretty much the same for every model : 12 drones = 480 Min/min => base empty in 25 minutes to compare with 16min 45 with 16 drones) I'm not trying to be a smart ass here but as it now, the main reason why a terran need to get a 4th is not the mineral income which is good anyway thx to mules (I ll get back to this) but the fact that his main will be empty at 15 minutes or so. Second and most importantly, simply forgeting about gaz makes the whole thing much less relevant. Minerals are only important when you build your economy but quickly enough it's the gaz that is the most important thing. Why zergs want a 4th? for minerals or for gaz? What do Protoss want? flood even more chargelots or get templars? Third your graphs doesn't take mules into account, and that's a big deal actually, it would be cool to have the same graphs for P and Z and one for T with as many mules as bases in the model. But maybe I'm in the details here. That's why I won't talk about the fact a scv building a CC is not mining for 100 seconds and that a drone does disappear each time a hatchery is build, I won't talk about worker travelling time either All that said, my point is I see what you're trying to do there but I think there are plenty of shortcuts you take that make the whole thing unaccurate economically. I think a better method might be to think about a target economy (like 2000 minerals 720 gaz per minute for example) and see how each model gives a different optimal solution to reach that point (with a somewhat dynamic model that takes in game time into account) and how it would actually impact the game. And then try to see what real benefit it would create when compared to Blizz current solution of lowering ressources per base. Oh for sure, you are correct that the post is incomplete in some regards, specially when including timeframes for expansion depletion, but as I stated in OP, this is because I'm not getting anything out of writing this down, I simply do not have the time to do all the measurements, check and re-check the data, make the graphs, make animations so the graphs can be easily understood and all. If you want the high resolution/high quality data for things like gas asymmetries or mules affecting income you will have or to trust Liquipedia or do it yourself, I left an open invitation to anyone that's willing to put the time into collecting said data, so anyone that has the time and will can add to the discussion. As I have said previously I would love to have that data, the problem is that I simply do not have the time to get to it. On March 07 2015 02:42 Gwavajuice wrote: Removing worker pairing wouldn't have changed how Yoda slowly built his late game mech, if anything he could just have built faster. soO couldn't break him, he thought that having a max income to rebuilt his air army 3 times in a row was a cool idea but it was not, your could have tripled his income, with his strategic choices he would have lost nonetheless. I fail to see how it would have been faster than it was done, the one that would have had the min income advantage would have been soO, because he would have been able to be more efficient with his workers than Yoda. You mention gas income and this is correct, this is the reason why players tend to take a fourth base, but this line of thinking is heavily based on StarCraft II constrains. As stated in the OP a higher mineral income advantage allows for players to be able to make more low tier units to trade inefficiently vs what in this case is a gas expensive Mech army, instead of needing to invest a high amount of vespene into higher tech units that spawn free units that achieve the same thing roach, zerglings and hidras would do in a economy that allows of asymmetric trades. This is why I'm saying that thinking of needing gas to be able to fight your opponent heavy gas units by making units that give a false sense of asymmetry is not a good way to go about it, because the same thing can be achieved by correcting the economy. I highly recommend you to mess around with the BW/Starbow economy mod in specific, so you can see how much of an effect it has, let me repeat myself what I said in the OP, I'm not saying that changing the economy is a end all and that we all will be happy, there are many things I disagree with respecting unit design which I think could be improved, but changing the economy to allow for asymmetric mineral vs vespene fights is a huge step forward. | ||
|
EatThePath
United States3943 Posts
What you're asking for is to solve the metagame in a new environment with different economic settings. Multiple times over, no less, for comparison. Of hypothetical changes. Surely we don't have to go that far, if it were even possible. I think the point your post is driving at is "why do I care about getting minerals more efficiently?" There are two simple answers: 1. You need less workers to build your army. 2. You need less workers to build your army. Let me explain. On the one hand, by spreading your workers to more bases (a privilege you've won through map control or pinning the opponent or superior multitask and positioning or... etc.) you get more money than you would otherwise which means you can ramp up production and churn out units faster. Taking more bases beyond your 3rd is an economic accelerator, not just an imperative on a timer. On the other hand, the economy you gain by using more base locations frees up supply in workers you don't have because your base count makes up for it. This lets you build a larger army which in all cases increases your efficiency in exchanging material, which is the whole purpose of a strategy based on superior economy. In short, it's not about getting more minerals to burn, it's about increasing efficiency. The point is not that the economy cap is higher (although functionally that's probably true, and your discussion of expansion timing has some bearing here); the point is that the economic system allows different styles of play that can reward skillful management of more assets, liabilities, threats and defenses, which incidentally is also more fun to see in action. That, and it makes for a deeper game. | ||
|
varsovie
Canada326 Posts
On March 07 2015 02:42 Gwavajuice wrote: I'm not trying to be a smart ass here but as it now, the main reason why a terran need to get a 4th is not the mineral income which is good anyway thx to mules (I ll get back to this) but the fact that his main will be empty at 15 minutes or so. Second and most importantly, simply forgeting about gaz makes the whole thing much less relevant. Minerals are only important when you build your economy but quickly enough it's the gaz that is the most important thing. Why zergs want a 4th? for minerals or for gaz? What do Protoss want? flood even more chargelots or get templars? Third your graphs doesn't take mules into account, and that's a big deal actually, it would be cool to have the same graphs for P and Z and one for T with as many mules as bases in the model. But maybe I'm in the details here. That's why I won't talk about the fact a scv building a CC is not mining for 100 seconds and that a drone does disappear each time a hatchery is build, I won't talk about worker travelling time either We're actually talking about base econ, not empty bases. A Terran taking a 4th or 5th because his main/nat mines out is simply taking another 3rd ECON base. The point of changing the mining paradygme is so Terran (or any other race) actually gets an incentive to take 4th or 5th while there is still mineral left in the main and nat. Right now it only makes is harder to defend with no benefit at all unless you're willing to make more worker and therefor have a weaker army. Of course they can get extra gas if needed, but that's another issue than mineral mining. MULES aren't into the graph because it doesn't matter with the number of base saturation (unless you manage to overstack 2 mules per patch), same for worker time investment in buildings since the graphs do not take account time but just worker/base. As for race differences it wouldn't matter since workers and base template are symetric in SC2 unlike BW. | ||
|
404AlphaSquad
839 Posts
On March 06 2015 11:43 coolman123123 wrote: I think the problem with a change like this, in Blizzard's eyes, is that it is confusing to new players. The current system, where 1 base 16 workers = 2 base 8 workers, is a lot more straightforward. It is the default mode of thinking. As a new player, it would make very little sense to me why it's better to have 2 bases 8 workers as opposed to one base 16. Additionally, this change would mean you are spending MORE time babysitting your workers and how they are distributed, something that obviously is not very fun. With the LotV change, Blizzzard is, in essence, trying to achieve a similar outcome with a simpler solution. I'm not saying the LotV econ will be better, or that these changes wouldn't be positive, but I believe this is why something like this will not happen. Who cares about new players? It is about the longlevity of the game and not some casuals who buy the game, cannon rush and then leave it again. Having a stable economy system should be the basic of any RTS game. First you design economy, then macromechanics, buildings, techs and then you build units on top of that. Having a clearly inferior economy system just for sake of "not confusing" new players who dont understand anything anyways and balance whine non-stop is really bad for the game. You cannot build an RTS to please every single person. Especially when it comes to these details, no newcomer will notice, and even if they do, GRATZ for them! They learned something. I am sick of dumbing down games to handfeed them to people who call themselves "gamers", but dont want to actually learn to play a game decently. | ||
|
albertojamed
United States1 Post
User was banned for this post. | ||
|
Deimos
Mexico134 Posts
| ||
|
Big J
Austria16289 Posts
On March 07 2015 02:42 Gwavajuice wrote: Of course he did had an income advantage seriously you can't possibly say otherwise, oh yes it was gaz income soO wanted but you discarded gaz mining from the problem . anyway soO had the biggest bank ever and it didn't save him.Yes, he had an income advantage because he went unorthodoxly high in drones. 90-100drones is not normal. And even with that amount of drones he had roughly like 10% more minerals mined. Not sure about the gas, but it should not be much more than 20% more gas mined either (and Yoda would have caught up in that regard if the game had gone longer). The whole point is that if you go for extra bases and saturate them (like soO) you have a small army that cannot attack, even though you have more income. If you don't saturate extra bases, you can max once and go, but afterwards you are dead. Both options don't make for good gameplay against a turtling opponent. | ||
|
Gwavajuice
France1810 Posts
On March 07 2015 04:05 EatThePath wrote: @gwavajuice: What you're asking for is to solve the metagame in a new environment with different economic settings. Multiple times over, no less, for comparison. Of hypothetical changes. Surely we don't have to go that far, if it were even possible. I think the point your post is driving at is "why do I care about getting minerals more efficiently?" There are two simple answers: 1. You need less workers to build your army. 2. You need less workers to build your army. Let me explain. On the one hand, by spreading your workers to more bases (a privilege you've won through map control or pinning the opponent or superior multitask and positioning or... etc.) you get more money than you would otherwise which means you can ramp up production and churn out units faster. Taking more bases beyond your 3rd is an economic accelerator, not just an imperative on a timer. On the other hand, the economy you gain by using more base locations frees up supply in workers you don't have because your base count makes up for it. This lets you build a larger army which in all cases increases your efficiency in exchanging material, which is the whole purpose of a strategy based on superior economy. In short, it's not about getting more minerals to burn, it's about increasing efficiency. The point is not that the economy cap is higher (although functionally that's probably true, and your discussion of expansion timing has some bearing here); the point is that the economic system allows different styles of play that can reward skillful management of more assets, liabilities, threats and defenses, which incidentally is also more fun to see in action. That, and it makes for a deeper game. Cool story except nothing in the OP prooves DH or starbow would do that (sorry btw I didn't realize it was just copy past from an existing post) because the data used are imho badly used if not incorrect. The way the "N base vs N+1 bases" graphs are built seems inaccurate to me and doesn't describe properly the incentive of making the decision to build an extra base. all it comes down to is : do I want to spend the money to build a base and if I do so how long will take to get my rmoney back? the graphs provided here don't answer this question. On March 07 2015 04:07 varsovie wrote: We're actually talking about base econ, not empty bases. A Terran taking a 4th or 5th because his main/nat mines out is simply taking another 3rd ECON base. The point of changing the mining paradygme is so Terran (or any other race) actually gets an incentive to take 4th or 5th while there is still mineral left in the main and nat. Right now it only makes is harder to defend with no benefit at all unless you're willing to make more worker and therefor have a weaker army. Of course they can get extra gas if needed, but that's another issue than mineral mining. MULES aren't into the graph because it doesn't matter with the number of base saturation (unless you manage to overstack 2 mules per patch), same for worker time investment in buildings since the graphs do not take account time but just worker/base. As for race differences it wouldn't matter since workers and base template are symetric in SC2 unlike BW. Mules do count cause the graphs are base on an income ratio (income on N+1 bases / income on N bases), mules add income and math laws tell you that if something is added on both part of a ratio, you can't just discard it (eg : 2/3 is not 3/4) As for the incentive I still believe the calculation displayed in the OP are very misleading and don't show the real stuff. I take a very simple example : 48 workers. 3 bases standard (16 workers each) : income is 1920 Min/min 3 base double harvest : 1860 (if I read correctly these graphs) 4 base standard (12 workers) : income is still 1920 4 base DH : 2000 (it seems) conclusion : incentive = 0% for standard and 7.5% for DH. yeah! I m gonna take a 4th base but even if I don't t think about mules, mineral depletion, workers travel time to the new base and everything, the 4th base is not free nor appearing instantly. so it's 400 minerals and 100 sec (assumption : I'm protoss and my probe can go back to mining instantly). for a gain of 140 Min/min. I ll need 2min51 to get my 400 mins back, add the 100 seconds and you have 4 min 31 before I simply get my investment back. So suddenly even if I'm ultra simplistic (unrealistic I dare say) the incentive is not so wonderfull... I don't want to throw stones at anyone, I just want to say the benefits brought by such a change seems at first very minor to : a - reducing the amount of mins per base b - (really my favorite but that's just me) better design of units and maps. Or I just totally missed the point which happens when my dumbness attacks... | ||
|
OtherWorld
France17333 Posts
| ||
| ||
