Hi, I am back with another topic to analyze. This time it is about the possible imbalance in ZvZ due to the Hatcheries' Larvae always spawning on the south side. The legend says that if a mineral line is to the south of the base, rather than to the north, then the spawning Drones will have to travel much less to the minerals when they begin working, and this materializes in an advantage for that player. Is it true, though? Here is what I found:
As always, here is the pdf for convinience and better quality.
TL;DR: If one player has a base with a southward mineral line, the other with a northward one, the former does have an advantage, but it is minimal. If the mineral lines are diagonal, the difference is even smaller; if they face east and west, respectively, then there should be no imbalance.
I'm looking forward to your opinions, criticisms, etc.
I don't think it makes that much of a difference. However it would be interesting if someone could give us the win rate in ZvZ based on the players spawn.
The biggest difference in hatch-mineral positioning is the defense of early pools. If you are at a north hatch, your lings can easily be surrounded by enemy lings. If you are at the south hatch, your lings can never be surrounded without the risk of a drone surround.
There should have hatch orientation and an option to set it in melee initialisation as a minimum, that way mappers can use it if they wish.
More completely Blizzard could give players access to building orientation on hatches for other possible minuscule advantages, like spawning larvae away from potential bunker rushes.
Assuming you've done that in a way which works - isn't ugly as hell, is ignorable for newer players, doesn't require 6 fingers a hand to do, that sort of thing - the same for Terran addons at Barracks/Factory/Starport could be implemented as I'd guess that's decided more games than larva spawning.
For the Terran case though the can of worms(!) is larger: Addon only or whole building? Rotating or mirroring? Can you rotate flying buildings? Which adjacent building does the addon prioritise? The answers maybe straightforward but the implementation might not be.
What were we talking about again? Oh yea, 23 minerals @ first base saturation isn't that big a deal, but this is interesting stuff - would read more!
On August 15 2014 17:07 algue wrote: I don't think it makes that much of a difference. However it would be interesting if someone could give us the win rate in ZvZ based on the players spawn.
You could use a probitmodel and use all the factors that affect if you win or lose. Like, skill (which is unobservable so you might use a proxy), mapsize, apm, experience, race choice and so on... But I am quite sure that the coefficient on "spawn" is statistically insignificant. Although the idea is interesting.
On August 15 2014 17:14 lichter wrote: The biggest difference in hatch-mineral positioning is the defense of early pools. If you are at a north hatch, your lings can easily be surrounded by enemy lings. If you are at the south hatch, your lings can never be surrounded without the risk of a drone surround.
This is without a doubt a more important factor than losing mining time. I can't calculate this, though, but the mining time loss I could, so at least we now know that that is negligible.
Not wanting to take anything away from that analysis - but the tl;dr part was rather obvious . Though you seem to love those mathematics, so keep going!
I think there's a small imbalance with spawns in any match-up; for terrans, it's useful to get a spawn where your rax can build an add-on inside the wall instead of as part of it, for example. Without statistics, however, it's impossible to say if it matters over the course of an entire game.
If you take slightly less time to reach a certain amount of minerals, then you can make a new drone slightly faster, which is going to mean that you can mine with the new drones slightly sooner. I think OP didn't take this into account, but merely calculated the mineral disparity resulting from the distances that the drones would have to travel. It still shouldn't be a big difference even if you took the faster access to new drones into account, but it is something to consider if you want to be thorough with the analysis.
i tested it using the ai, set urself to ref in a custom game and get 2 ais to play against each other, i used zerg and the lingRoach timing, on all 8 goes the northward position had a 5 second lead of his warren timing which is something i didnt expect but their builds are literally identical each time . . .but that initial mineral assignment tho! worker acceleration play a factor in this? I would of expected the ai to mine super efficient in the south position, but was never the case. always seemed to be just under a split second behind on the mineral tick. this is a good example of how the game snowballs when one player gets a slight advantage . . . .hmm considering never drone scouting again!
On August 15 2014 18:38 GERMasta wrote: If you take slightly less time to reach a certain amount of minerals, then you can make a new drone slightly faster, which is going to mean that you can mine with the new drones slightly sooner. I think OP didn't take this into account, but merely calculated the mineral disparity resulting from the distances that the drones would have to travel. It still shouldn't be a big difference even if you took the faster access to new drones into account, but it is something to consider if you want to be thorough with the analysis.
Yes, only the Larva system of the Zerg screws up things, insomuch as being able to morph additional Drones does not only depend on whether you have enough minerals. I think, though, the effect is so marginal that even if it snowballs, it remains negligible.
On August 15 2014 19:31 Faust852 wrote: I don't think you can fix this tho. Putting larvas on the otherside would make them almost unclicable.
That can't be an argument. You would have to exclude northern high grounds then . Also, there's very few opportunities to actually click larva manually (eg. hellions sniping larva, or running banes into it). Most zerg players will use the hatchery to access larva.
On August 15 2014 18:38 GERMasta wrote: If you take slightly less time to reach a certain amount of minerals, then you can make a new drone slightly faster, which is going to mean that you can mine with the new drones slightly sooner. I think OP didn't take this into account, but merely calculated the mineral disparity resulting from the distances that the drones would have to travel. It still shouldn't be a big difference even if you took the faster access to new drones into account, but it is something to consider if you want to be thorough with the analysis.
Yes, only the Larva system of the Zerg screws up things, insomuch as being able to morph additional Drones does not only depend on whether you have enough minerals. I think, though, the effect is so marginal that even if it snowballs, it remains negligible.
That would be the real thing to find out I think. Sometimes, I have the very subjective feeling of "gosh I'm so slow", which is really more than a feeling then some proven thing. That might explain what happens, because though I'm only a diamond/masters player (rather diamond then masters), I'm quite capable of executing a perfect hatch first build until 28-ish supply. Sometimes, I just feel slow. Dunno exactly why.
On August 15 2014 19:31 Faust852 wrote: I don't think you can fix this tho. Putting larvas on the otherside would make them almost unclicable.
That can't be an argument. You would have to exclude northern high grounds then . Also, there's very few opportunities to actually click larva manually (eg. hellions sniping larva, or running banes into it). Most zerg players will use the hatchery to access larva.
On August 15 2014 18:38 GERMasta wrote: If you take slightly less time to reach a certain amount of minerals, then you can make a new drone slightly faster, which is going to mean that you can mine with the new drones slightly sooner. I think OP didn't take this into account, but merely calculated the mineral disparity resulting from the distances that the drones would have to travel. It still shouldn't be a big difference even if you took the faster access to new drones into account, but it is something to consider if you want to be thorough with the analysis.
Yes, only the Larva system of the Zerg screws up things, insomuch as being able to morph additional Drones does not only depend on whether you have enough minerals. I think, though, the effect is so marginal that even if it snowballs, it remains negligible.
That would be the real thing to find out I think. Sometimes, I have the very subjective feeling of "gosh I'm so slow", which is really more than a feeling then some proven thing. That might explain what happens, because though I'm only a diamond/masters player (rather diamond then masters), I'm quite capable of executing a perfect hatch first build until 28-ish supply. Sometimes, I just feel slow. Dunno exactly why.
The feeling of being "slow" has mostly to do with mineral stacking. On some maps, the natural minerals that workers mine from suck more than others, but if you are really paying attention to drone stacking for the first minute or two of the game, you can really minimize a lot of this inefficiency.
I wish you would take on a "bigger" aspect in future posts, of course it is not easy to know beforehand wether something will have a big or small influence.
On August 15 2014 19:31 Faust852 wrote: I don't think you can fix this tho. Putting larvas on the otherside would make them almost unclicable.
That can't be an argument. You would have to exclude northern high grounds then . Also, there's very few opportunities to actually click larva manually (eg. hellions sniping larva, or running banes into it). Most zerg players will use the hatchery to access larva.
On August 15 2014 19:09 Sholip wrote:
On August 15 2014 18:38 GERMasta wrote: If you take slightly less time to reach a certain amount of minerals, then you can make a new drone slightly faster, which is going to mean that you can mine with the new drones slightly sooner. I think OP didn't take this into account, but merely calculated the mineral disparity resulting from the distances that the drones would have to travel. It still shouldn't be a big difference even if you took the faster access to new drones into account, but it is something to consider if you want to be thorough with the analysis.
Yes, only the Larva system of the Zerg screws up things, insomuch as being able to morph additional Drones does not only depend on whether you have enough minerals. I think, though, the effect is so marginal that even if it snowballs, it remains negligible.
That would be the real thing to find out I think. Sometimes, I have the very subjective feeling of "gosh I'm so slow", which is really more than a feeling then some proven thing. That might explain what happens, because though I'm only a diamond/masters player (rather diamond then masters), I'm quite capable of executing a perfect hatch first build until 28-ish supply. Sometimes, I just feel slow. Dunno exactly why.
The feeling of being "slow" has mostly to do with mineral stacking. On some maps, the natural minerals that workers mine from suck more than others, but if you are really paying attention to drone stacking for the first minute or two of the game, you can really minimize a lot of this inefficiency.
Hm, I do pay attention to mineral stacking. Seriously. Thing is, I mostly play 3v3/4v4 these days, and on many of those maps, the mineral patches are truely poor placed, so that might be the reason? (ofc, I do not always stack perfectly, so that might be the reason, too, yep)
On August 15 2014 18:12 boxerfred wrote: Not wanting to take anything away from that analysis - but the tl;dr part was rather obvious . Though you seem to love those mathematics, so keep going!
Giving a mathematical proof and exact measure of what intuition can guess is part of a mathematician's work. I love these series too.
On August 15 2014 19:53 HaRuHi wrote: Great math as usual! Thank you.
I wish you would take on a "bigger" aspect in future posts, of course it is not easy to know beforehand wether something will have a big or small influence.
I try to cover problems that are relatively easy to solve. Alas, they often tend to be insignificant ones, as more practical applications can be really complex. Anyway, any idea for the topic of a future post is appriciated, although I still have some ideas, but I will run out of them eventually.
On August 15 2014 19:31 Faust852 wrote: I don't think you can fix this tho. Putting larvas on the otherside would make them almost unclicable.
That can't be an argument. You would have to exclude northern high grounds then . Also, there's very few opportunities to actually click larva manually (eg. hellions sniping larva, or running banes into it). Most zerg players will use the hatchery to access larva.
On August 15 2014 19:09 Sholip wrote:
On August 15 2014 18:38 GERMasta wrote: If you take slightly less time to reach a certain amount of minerals, then you can make a new drone slightly faster, which is going to mean that you can mine with the new drones slightly sooner. I think OP didn't take this into account, but merely calculated the mineral disparity resulting from the distances that the drones would have to travel. It still shouldn't be a big difference even if you took the faster access to new drones into account, but it is something to consider if you want to be thorough with the analysis.
Yes, only the Larva system of the Zerg screws up things, insomuch as being able to morph additional Drones does not only depend on whether you have enough minerals. I think, though, the effect is so marginal that even if it snowballs, it remains negligible.
That would be the real thing to find out I think. Sometimes, I have the very subjective feeling of "gosh I'm so slow", which is really more than a feeling then some proven thing. That might explain what happens, because though I'm only a diamond/masters player (rather diamond then masters), I'm quite capable of executing a perfect hatch first build until 28-ish supply. Sometimes, I just feel slow. Dunno exactly why.
The feeling of being "slow" has mostly to do with mineral stacking. On some maps, the natural minerals that workers mine from suck more than others, but if you are really paying attention to drone stacking for the first minute or two of the game, you can really minimize a lot of this inefficiency.
Hm, I do pay attention to mineral stacking. Seriously. Thing is, I mostly play 3v3/4v4 these days, and on many of those maps, the mineral patches are truely poor placed, so that might be the reason? (ofc, I do not always stack perfectly, so that might be the reason, too, yep)
Well, yeah, some maps are just bad. For instance, I noticed today that on Foxtrot Labs in the top left (?) position, my drones will rally to a far away patch by walking behind the mineral line. The idea is not necessarily that mineral stacking will completely prevent these imbalances, but it should mitigate it some and give you the extra 10-15 minerals that you're used to.
On August 15 2014 17:14 lichter wrote: The biggest difference in hatch-mineral positioning is the defense of early pools. If you are at a north hatch, your lings can easily be surrounded by enemy lings. If you are at the south hatch, your lings can never be surrounded without the risk of a drone surround.
This is one of the main problems in the early game of ZvZ. This is a huge difference. U can avoid this problem making the pool under the larvas so when the queen spawns it is 50% covered by the pool and the hatch, but it's not the same.
On August 15 2014 19:52 StatixEx wrote: only way it can be solved if this EVER poses some kind of issue is that the camera is relative to you so you always spawn south
On August 15 2014 17:54 Dapper_Cad wrote: Licheter makes a strong point.
There should have hatch orientation and an option to set it in melee initialisation as a minimum, that way mappers can use it if they wish.
More completely Blizzard could give players access to building orientation on hatches for other possible minuscule advantages, like spawning larvae away from potential bunker rushes.
Assuming you've done that in a way which works - isn't ugly as hell, is ignorable for newer players, doesn't require 6 fingers a hand to do, that sort of thing - the same for Terran addons at Barracks/Factory/Starport could be implemented as I'd guess that's decided more games than larva spawning.
What's wrong with just having larvae obey a rally point?
On August 15 2014 19:52 StatixEx wrote: only way it can be solved if this EVER poses some kind of issue is that the camera is relative to you so you always spawn south
Oh my, this was in NHL games(TAB key switched the view, so you can play all periods from down to up). I so much want this in SC2, I have a really big issues playing from Northern regions.
But is the advantage big enough to have any significant difference as the game goes on?
Is the difference 3.5 minerals per drone at the worst case scenario? Then its not worth looking into really, build order counters, and simple micro mistakes are far more influential than any kind of small drone imbalance like this.
On August 19 2014 01:22 .kv wrote: the day they fix this will be the day they give terrans the option to put their addons on the left if they want to
This could be together with the "view" rotation in LotV. But, yeah, I do not believe it will be there.
Is there any bnet forum or thread where we can write these ideas? I know, that some Blizzard employees read TL.net from time to time, but I fear it is not enough. (Hope dies last )
On August 15 2014 19:31 Faust852 wrote: I don't think you can fix this tho. Putting larvas on the otherside would make them almost unclicable.
That can't be an argument. You would have to exclude northern high grounds then . Also, there's very few opportunities to actually click larva manually (eg. hellions sniping larva, or running banes into it). Most zerg players will use the hatchery to access larva.
On August 15 2014 19:09 Sholip wrote:
On August 15 2014 18:38 GERMasta wrote: If you take slightly less time to reach a certain amount of minerals, then you can make a new drone slightly faster, which is going to mean that you can mine with the new drones slightly sooner. I think OP didn't take this into account, but merely calculated the mineral disparity resulting from the distances that the drones would have to travel. It still shouldn't be a big difference even if you took the faster access to new drones into account, but it is something to consider if you want to be thorough with the analysis.
Yes, only the Larva system of the Zerg screws up things, insomuch as being able to morph additional Drones does not only depend on whether you have enough minerals. I think, though, the effect is so marginal that even if it snowballs, it remains negligible.
That would be the real thing to find out I think. Sometimes, I have the very subjective feeling of "gosh I'm so slow", which is really more than a feeling then some proven thing. That might explain what happens, because though I'm only a diamond/masters player (rather diamond then masters), I'm quite capable of executing a perfect hatch first build until 28-ish supply. Sometimes, I just feel slow. Dunno exactly why.
The feeling of being "slow" has mostly to do with mineral stacking. On some maps, the natural minerals that workers mine from suck more than others, but if you are really paying attention to drone stacking for the first minute or two of the game, you can really minimize a lot of this inefficiency.
Hm, I do pay attention to mineral stacking. Seriously. Thing is, I mostly play 3v3/4v4 these days, and on many of those maps, the mineral patches are truely poor placed, so that might be the reason? (ofc, I do not always stack perfectly, so that might be the reason, too, yep)
Well, yeah, some maps are just bad. For instance, I noticed today that on Foxtrot Labs in the top left (?) position, my drones will rally to a far away patch by walking behind the mineral line. The idea is not necessarily that mineral stacking will completely prevent these imbalances, but it should mitigate it some and give you the extra 10-15 minerals that you're used to.
Can you upload a picture indicating which was the mineral patch you where targeting? In Foxtrot i'm using one of the standard mineral placements specially to avoid such issues, if this issue still persists i will have to talk to the other mapmakers so we all stop using this configuration.
PSA: If anyone finds something that looks clearly like a bug or an unintentional issue with a map, please post in the corresponding map thread here in TL, otherwise we will have no way to fix it or even to know about it.
♦ Here i leave you a picture so it is easier for you to highlight which is the mineral you were targeting.
EPT![[image loading]](http://i.imgur.com/8Ea9Gzd.jpg)
![[image loading]](http://i.imgur.com/stGWcL8.jpg)
![[image loading]](http://i.imgur.com/yJ4pg8B.jpg)
![[image loading]](http://i.imgur.com/vwK98M8.jpg)
.
. Though you seem to love those mathematics, so keep going! 
)
