|
|
On April 22 2015 12:46 StarStruck wrote:Show nested quote +On April 22 2015 09:16 MiniFotToss wrote:
I understand the 96 worker comparison in 4 base and 2 base, but seriously, no one can realistically builds 96 workers in a game and have enough army to kill your opponent, defend his push or win. yes it's kind of stupid when it gets to that point and always has been. A player shouldn't have to get to that point. Heck even 60 in BW was kind of ~_~. We want more supply.
With TL model you would get that but people would also have more resources to tech. So I don't think Blizzard misunderstood TL proposal. I think it was the other way around.
|
Excellent write up!
On April 22 2015 09:16 MiniFotToss wrote:
I understand the 96 worker comparison in 4 base and 2 base, but seriously, no one can realistically builds 96 workers in a game and have enough army to kill your opponent, defend his push or win.
Maybe not straight up kill your opponent, but if you have map control and increased income, that means it's going to be harder for them to choke you out, meaning your goal shouldn't have to be to just kill your opponent. For example, you could throw smaller armies at him (or harass) while you keep expanding and reproducing with your higher production and income.
|
I still think you haven't really addressed how this resource curve will effect early game balance between the races. All of the 3 races are very balanced early game despite have significantly different economic models (mules, chronoboost and larva injects). Although they are now very balanced, this balance is tenuous and changing one factor, especially one as fundamental as the economy, would have large ripple effects on every aspect of the game.
The model you are supporting allows greater mining for more bases. Currentlty, Zerg is generally required to have one more mining base than Teran or Protoss to be even with them in army strength. As such, the model you are suggesting will benefit zerg more than the other races and result in them having more mineral income.
To continue the theory crafting, the match-up it would likely most effect is Z v P. Zerg taking 3 bases before pool is not an uncommon build in ZvP, while it does not happen often in T v Z due to the threat of heavy reaper openings and the need for earlier gas. In Z v P a zerg opening 3 base before pool under your model would now have significantly more income then before. If we can agree that the HOTS meta is currently very balanced, then your model would by definition unbalance this by favouring zerg. In addition, the zerg larva inject mechanic would mean that this advantage would snowball very quickly allowing even quicker 4th bases for zerg. The current balance for Z v P would change throughout the match-up.
Now we would have to solve this balance issue, to improve Protoss's chances and we need to keep in mind that every change to Z v P is not in isolation as it affects our other matchups, so what are our options? Do we give protoss an early game harass unit to punish early zerg 3rds similar to the reaper? What does this do to our P v T balance? Do we nerf zerg larva injects to slow down their economic advantage? How do we fix our Z v T balance after this?
A small change in the early economic model would change the entire balance throughout the game and in all matchups. While you can make the argument zerg is constrained by gas, there are very few zerg compositions that require significantly more than 3 bases and 6 geysers at the mid game and increasing the zerg's income at a faster rate than the other races would allow them access to 8 geysers faster. By definition your model encourages this as the marginal gain of each additional mineral mining worker is less and less, which would encourage zergs to take their gas earlier.
I think you have done great work on the theory, but balance design needs to take all of the many parts of the game into account. You are simply looking at changing one item in a vacuum and ignoring the ripples this fundamental change would have.
I believe Blizzard has chosen the current LOTV model, not out of laziness or spite, but because it is the most similar to the current stable HOTS model.
TLDR - changing the economic model for something as fundamental as income collection rates in SC2 is a significant change that would require balance work on almost all aspects of the game and is not a small adjustment.
|
ok, so you wrote a very long article that even the author would be happy if 5% would actually read it. Then it turns out that David Kim was not amongst the 5%. And you reply with yet more pages to correct one of his sentences.
How about you explain things briefly and to the point, instead of writing 5 paragraphs about every little thing? It is clear that you have good writers on the staff, but it also seems like there is some confusion regarding how many words is needed and appropriate to get a message across in this context.
In this case, the one single point you make seems to be "you don't get anywhere close to double income on 4 bases vs 2 bases in our model, unless you have unrealistic amounts of workers, as seen in the fig here. [FIG]" I understand that you want to add the "thanks for your reply" paragraph at the start, but after that I feel you are using a lot more words than you need. Did you feel that you needed to add more word to motivate a post? >_>
I know that you are not allowed to go "TL:DR" on teamliquid, and for good reasons, but you kindof have to realise that the longer the article is, the less likely it is that people will actually read it. David Kim and the LotV team is by no means immune to that effect.
|
This is really, really encouraging. Mad respect for Plexa for getting this out in such a timely manner. This discussion actually has me more amped than the fact that, if what Dayvie said is true, I'm getting a beta key soon ;p
Really looking forward to Zeromus' next article, the show matches, and Blizzard's response. Everyone involved in this kicks so much ass it's just silly.
PS on Thursday evening pacific time I'll be in the Double Harvest group looking for games. People should come out, it was a little lonely there today.
|
Very well written response! I don't think I would be so diplomatic if someone would twist my idea....
I think there is one more reason for double-harvesting that needs to get across better:
With normal mining algorithm there is a single point at which a mineral field is "fully saturated". If the number of workers is below that point - they mine at full efficiency. If the number of workers is higher - they start to wait one for another, reducing the efficiency. You can change mining speed, wait time or travel time, but all it will accomplish is to move the saturation point up or down.
In the standard mining, the saturation point is at around 2.5 workers per mineral patch. That means - 2 workers mine at full efficiency, while 3 workers is a bit too much (workers must wait). As a result 16-workers in a single base will be as efficient as 16 workers split into two bases.
By tweaking the numbers, you can put the saturation point lower, say at 1.9 worker/patch. When that happens, the 2nd worker will wait for the first one, reducing the efficiency right there. The problem is, that the workers occupy the mineral patch all the time. Adding 3rd worker will not benefit mining at all! Thus 16-workers on a single base already fully saturate it.
Double Harvest is different - it adds additional variable into the component (two harvests and two wait times at minerals per trip, but a single travelling time). This "breaks" the hard saturation point. It is possible for one harvester to wait for another, yet still having the mineral patch have an "unattended time", that 3rd worker can benefit from.
I think the above can be an important property of DH that remains unnoticed by David Kim. He thinks, "Efficiency drop at 16 workers implies that base is fully saturated". Which would be true for standard system that he is more familiar with but not DH.
|
Aotearoa39261 Posts
On April 22 2015 14:50 Cascade wrote: ok, so you wrote a very long article that even the author would be happy if 5% would actually read it. Then it turns out that David Kim was not amongst the 5%. And you reply with yet more pages to correct one of his sentences. How about you explain things briefly and to the point, instead of writing 5 paragraphs about every little thing? It is clear that you have good writers on the staff, but it also seems like there is some confusion regarding how many words is needed and appropriate to get a message across in this context. In this case, the one single point you make seems to be "you don't get anywhere close to double income on 4 bases vs 2 bases in our model, unless you have unrealistic amounts of workers, as seen in the fig here. [FIG]" I understand that you want to add the "thanks for your reply" paragraph at the start, but after that I feel you are using a lot more words than you need. Did you feel that you needed to add more word to motivate a post? >_> I know that you are not allowed to go "TL:DR" on teamliquid, and for good reasons, but you kindof have to realise that the longer the article is, the less likely it is that people will actually read it. David Kim and the LotV team is by no means immune to that effect. I actually think he read it, but got the wrong idea (probably from the 1:1 efficiency thing) and then analysed the remainder of the article with that faulty premise. This is directed towards him and his team to correct that. If you want the TLDR then it's these two graphs ![[image loading]](http://www.teamliquid.net/staff/Plexa/dh10/dh10_hots4base.png) ![[image loading]](http://www.teamliquid.net/staff/Plexa/dh10/dh10_4base.png) and the simple question; which curve would you prefer in your game?On April 22 2015 14:50 rdvark5000 wrote:
TLDR - changing the economic model for something as fundamental as income collection rates in SC2 is a significant change that would require balance work on almost all aspects of the game and is not a small adjustment.
Absolutely. No one is claiming otherwise! Any change to the economy is going to have extensive repercussions to the game; as we're seeing at the moment in LotV. Our point is that changing the economy has such sweeping effects that if you're going to tweak it, it NEEDS to be done early in the beta so that the game can be balanced before it comes out. You can't decide 2 weeks before beta ends that the economy isn't having the desired effect, hence why we're putting such a heavy emphasis on getting this tested 
We're also not in a position to make balance changes etc. as that oversteps the point of our article.
|
Nice post, although it's licking Blizzard's feet a bit too much for my tastes. And yeah DH 8 would have been better, but well DH 10 is better than LotV model I guess
|
Using the classic sc2 harvest model of 5 per trip, could the standard patch of 8 minerals be reshaped so that the closer ones are mined perfectly with 2 harvesters, then some at 2.5 then some at 3 and then some at 3.5 for example?
That would reward base expanding strats in the short run by putting workers on the closer patches across bases earlier, being a short term incentive to expansion.
Then the closer patches could have less minerals to start with (like the current LotV), which would provide the long term incentive to expand because the low minerals patches would vanish relatively ast.
(I am not remotely a good player or a balance specialist, so maybe this is terrible.)
|
United States4883 Posts
On April 22 2015 14:50 Cascade wrote:
ok, so you wrote a very long article that even the author would be happy if 5% would actually read it. Then it turns out that David Kim was not amongst the 5%. And you reply with yet more pages to correct one of his sentences.
How about you explain things briefly and to the point, instead of writing 5 paragraphs about every little thing? It is clear that you have good writers on the staff, but it also seems like there is some confusion regarding how many words is needed and appropriate to get a message across in this context.
In this case, the one single point you make seems to be "you don't get anywhere close to double income on 4 bases vs 2 bases in our model, unless you have unrealistic amounts of workers, as seen in the fig here. [FIG]" I understand that you want to add the "thanks for your reply" paragraph at the start, but after that I feel you are using a lot more words than you need. Did you feel that you needed to add more word to motivate a post? >_>
I know that you are not allowed to go "TL:DR" on teamliquid, and for good reasons, but you kindof have to realise that the longer the article is, the less likely it is that people will actually read it. David Kim and the LotV team is by no means immune to that effect.
Because the subject IS complicated. It's necessary to write a few paragraphs, especially if our goal is to clarify the facts of the original article.
If you want a TL;DR: http://www.teamliquid.net/forum/legacy-of-the-void/482775-a-treatise-on-the-economy-of-scii?page=29#580
That's as concise as I can put it, but it by no means addresses all the issues and concerns regarding the economy and how it actually differs from the HotS or LotV economies.
|
United States4883 Posts
On April 22 2015 15:05 Plasmid wrote:
Using the classic sc2 harvest model of 5 per trip, could the standard patch of 8 minerals be reshaped so that the closer ones are mined perfectly with 2 harvesters, then some at 2.5 then some at 3 and then some at 3.5 for example?
That would reward base expanding strats in the short run by putting workers on the closer patches across bases earlier, being a short term incentive to expansion.
Then the closer patches could have less minerals to start with (like the current LotV), which would provide the long term incentive to expand because the low minerals patches would vanish relatively ast.
(I am not remotely a good player or a balance specialist, so maybe this is terrible.)
I'm not sure I'm following correctly, so let me try to clarify first. So say that the closest minerals pair workers on a 2:1 ratio, and further away ones could be adjusted to 2.5:1 or 3:1. In this case, the 24 node cap is reached on even fewer bases (possibly even two) because you're not actually breaking the fact that at least 16 workers can mine efficiently on a single base (linear growth). So at first glance, I would say no.
However, second idea could be to find a way to change mining efficiency on close patches closer to 1:1 -- say 1.5:1 -- and pair patches that are far away (though this might be difficult to run through the AI unless the nodes are relatively further away). Mixed with a half patch approach, this could be an intriguing solution. However, a potential problem is that workers will be bouncing around from the 1.5:1 nodes and still interrupt perfect worker pairing on the 2:1 nodes, thus making it an entirely non-worker pairing model.
It's an interesting concept, but I think the difficulties of actually getting the behavior to line up properly while still maintaining global move speed and mining speed and the fact that it would probably work just as well to use a full non-worker pairing model make it not really a viable option. Of course, this is all theorycraft, but I'm not sure I like this solution as well as some others.
|
Hi guys, great communication going on here about all the economy stuff :-D
I haven't read through all the comments here and on Bnet but i just came up with an (hopefully original) idea concerning the whole accelerating the pace of the game stuff. And i just want to throw it out there or rather get your probably more informed guess about it.
Do you think it is viable to choose your economy model for LotV (which would reduce the timer put onto a player during the one and 2-base phase of the game compared to the 12 worker LotV low minerals start) and then from a mapmaking point of view make all other expansions on a map gold bases? And this for all maps in order to speed up the late game parts.
You would end up with a mappool that has normal mineral patches in the main (idk maybe 1500 minerals each or 1250... whatever number you want to settle on) or maybe for the sake of 2-base play also the natural is a normal mineral field but then starting form the next possible base to take all other mineral fields from 3-bases on are high yield (but with the same amount of patches and number of minerals on each patch as on the normal mineral patches). This would speed up the late mid- and late game. But how dangerous or complicated would you recon is the higher income/minute influence on production.
Or would you do it vice versa (first to bases are golds... lets say with the current HotS economy model) to speed up the game in the early stages. But this i guess wouldn't change the long streched games problematic.
So in short... Any thoughts on experimenting with gold bases (more than the occasional 1 or 2 for each player in the recent mappools)?
|
On April 22 2015 12:17 coolman123123 wrote:
Does anyone feel it's kind of inappropriate for the writers of a fansite to use it as a soapbox for their ideas? I'm okay with the initial post but now it's become a campaign to have the idea tested. We can sit around and pretend like we know for certain that this is the best course of action for the game, but we don't. It's just an idea, and unless something is positively the right thing to do, I don't think this proactive approach should be taken.
Team Liquid is more than a fan site. Look in to it's history. Go read the 15,000 word articles on Starcraft. The active community it still holds for Broodwar. These guys wake up and the first thing they think about isn't kissing their wife, but the new meta.
They know Starcraft. Felt this might be better, and are looking to set up a continued dialogue with the developers (both on this issue and others) To try and make the game a true Legacy.
Thanks to Mr Kim for reaching out. Blizzard always seems a little aloof to me, and I am very grateful to them and you for this, I hope it continues.
On April 22 2015 09:00 Plexa wrote:
We hope this greater level of communication is a sign of things to come.
We’re all invested in the success of StarCraft II and we hope that this dialogue can continue so that we end up with the best possible end product.
|
On April 22 2015 15:07 SC2John wrote:Show nested quote +On April 22 2015 14:50 Cascade wrote:
ok, so you wrote a very long article that even the author would be happy if 5% would actually read it. Then it turns out that David Kim was not amongst the 5%. And you reply with yet more pages to correct one of his sentences.
How about you explain things briefly and to the point, instead of writing 5 paragraphs about every little thing? It is clear that you have good writers on the staff, but it also seems like there is some confusion regarding how many words is needed and appropriate to get a message across in this context.
In this case, the one single point you make seems to be "you don't get anywhere close to double income on 4 bases vs 2 bases in our model, unless you have unrealistic amounts of workers, as seen in the fig here. [FIG]" I understand that you want to add the "thanks for your reply" paragraph at the start, but after that I feel you are using a lot more words than you need. Did you feel that you needed to add more word to motivate a post? >_>
I know that you are not allowed to go "TL:DR" on teamliquid, and for good reasons, but you kindof have to realise that the longer the article is, the less likely it is that people will actually read it. David Kim and the LotV team is by no means immune to that effect. Because the subject IS complicated. It's necessary to write a few paragraphs, especially if our goal is to clarify the facts of the original article. If you want a TL;DR: http://www.teamliquid.net/forum/legacy-of-the-void/482775-a-treatise-on-the-economy-of-scii?page=29#580That's as concise as I can put it, but it by no means addresses all the issues and concerns regarding the economy and how it actually differs from the HotS or LotV economies.
You think David Kim read the entire original article, but came out thinking 4 bases was twice the income of 2 because it wasn't clear enough? To me it's obvious that the misunderstanding is due to DK (or whoever at blizzard read it) didn't bother to do more than skim the article, not that they are too stupid to read a graph. Or possibly some internal Blizzard communication mistake, due to the article being too long for DK to read himself.
it's not super-straight forward, but in no way do you need the length of the original article or this reply to get the point across.
And yes, a TL:DR in the original article would have helped a lot I think. A lot. As you want to give a serious impression, you can call it summary or even abstract instead. At page 29 it doesn't help much I am afraid. :/
I mean, I really like the idea, and I think it has great potential. I'm just sad to see it get buried in a wall of text that 5% actually read.
On another note, do you (or Barrin maybe?) have a spreadsheet or something with the Hots and DH mining rates (income as function of harvesters on a single base for example) laying around? Would be fun to have a look, and I guess it'd be nice to have publicly available. (Found it, Barrin posted in the main thread.)
|
Disruptor looks gay. I'd say give us new units who looks badass/cool. Unit designs that pleases our eyes...like the dark zealots. Damn those r sexy.
User was warned for posting completely off topic
|
Aotearoa39261 Posts
On April 22 2015 15:45 Cascade wrote:Edit: not my quote plexa.  + Show Spoiler +On April 22 2015 15:02 Plexa wrote:Show nested quote +On April 22 2015 14:50 Cascade wrote:
TLDR - changing the economic model for something as fundamental as income collection rates in SC2 is a significant change that would require balance work on almost all aspects of the game and is not a small adjustment.
Absolutely. No one is claiming otherwise! Any change to the economy is going to have extensive repercussions to the game; as we're seeing at the moment in LotV. Our point is that changing the economy has such sweeping effects that if you're going to tweak it, it NEEDS to be done early in the beta so that the game can be balanced before it comes out. You can't decide 2 weeks before beta ends that the economy isn't having the desired effect, hence why we're putting such a heavy emphasis on getting this tested We're also not in a position to make balance changes etc. as that oversteps the point of our article. Ah I accidentally removed my reply to your post in an edit. lol.
|
On April 22 2015 15:49 skylinefan wrote:
Disruptor looks gay. I'd say give us new units who looks badass/cool. Unit designs that pleases our eyes...like the dark zealots. Damn those r sexy.
User was warned for this post
Post of the year thus far.
|
Great post, I hope that DK will read this soon instead of him only looking at this again when the current lotv-model has been tested thoroughly. I don't think it would be bad to have 2 different test maps at the same time, and that would be ideal to see how this pans out in practice.
|
On April 22 2015 16:29 Anacreor wrote:
Great post, I hope that DK will read this soon instead of him only looking at this again when the current lotv-model has been tested thoroughly. I don't think it would be bad to have 2 different test maps at the same time, and that would be ideal to see how this pans out in practice.
Hmm, that'd be intersting actually! Having two online test realms, as see which patch is played the most. There may be serious issues (such as everyone playing the patch where their race is buffed), but a fun idea!
|
On April 22 2015 14:49 rdvark5000 wrote:
I still think you haven't really addressed how this resource curve will effect early game balance between the races. All of the 3 races are very balanced early game despite have significantly different economic models (mules, chronoboost and larva injects). Although they are now very balanced, this balance is tenuous and changing one factor, especially one as fundamental as the economy, would have large ripple effects on every aspect of the game.
The model you are supporting allows greater mining for more bases. Currentlty, Zerg is generally required to have one more mining base than Teran or Protoss to be even with them in army strength. As such, the model you are suggesting will benefit zerg more than the other races and result in them having more mineral income.
To continue the theory crafting, the match-up it would likely most effect is Z v P. Zerg taking 3 bases before pool is not an uncommon build in ZvP, while it does not happen often in T v Z due to the threat of heavy reaper openings and the need for earlier gas. In Z v P a zerg opening 3 base before pool under your model would now have significantly more income then before. If we can agree that the HOTS meta is currently very balanced, then your model would by definition unbalance this by favouring zerg. In addition, the zerg larva inject mechanic would mean that this advantage would snowball very quickly allowing even quicker 4th bases for zerg. The current balance for Z v P would change throughout the match-up.
Now we would have to solve this balance issue, to improve Protoss's chances and we need to keep in mind that every change to Z v P is not in isolation as it affects our other matchups, so what are our options? Do we give protoss an early game harass unit to punish early zerg 3rds similar to the reaper? What does this do to our P v T balance? Do we nerf zerg larva injects to slow down their economic advantage? How do we fix our Z v T balance after this?
A small change in the early economic model would change the entire balance throughout the game and in all matchups. While you can make the argument zerg is constrained by gas, there are very few zerg compositions that require significantly more than 3 bases and 6 geysers at the mid game and increasing the zerg's income at a faster rate than the other races would allow them access to 8 geysers faster. By definition your model encourages this as the marginal gain of each additional mineral mining worker is less and less, which would encourage zergs to take their gas earlier.
I think you have done great work on the theory, but balance design needs to take all of the many parts of the game into account. You are simply looking at changing one item in a vacuum and ignoring the ripples this fundamental change would have.
I believe Blizzard has chosen the current LOTV model, not out of laziness or spite, but because it is the most similar to the current stable HOTS model.
TLDR - changing the economic model for something as fundamental as income collection rates in SC2 is a significant change that would require balance work on almost all aspects of the game and is not a small adjustment.
I agree. I posted something very similar to you on the last page. see below.
On April 22 2015 13:04 ImYourHuckleberry wrote:
Maybe it's for a lack of my understanding after reading the article and watching the VOD, but this income change seems to ignore innate strengths and weaknesses of each race. Perhaps, in a perfectly balanced unit world, this income equality would work, but we are not there. For instance, the number one argument in this scenario is that mass expansions without harassment essentially leads to large advantages for the player. I understand the basic concept and normally this is logical; however, with the current unit compositions and unit design that exists, this ideology seems HEAVILY Zerg favored. What is to stop a Zerg from massing queens and playing even more defensive to mass expand? They hold off early attacks and the game is over? Protoss and Terran (mech) are designed for better late game compositions. It is the job of Zergs, for instance, to stop them from reaching this point and/or mass expanding across the map. With the current units available, this income change suggested by TL seems to put the other two races on a "clock". It appears we are putting the cart before the horse with such a focus on income, when the crux is the strengths and weaknesses of each race. I think Zerg was given a lot stronger units in LOTV and now to mess with income changes seems like too many variables until we scrutinize the new units.
|
|
|
|
|
|
EPT
