|
On November 30 2010 09:30 CarbonTwelve wrote:Show nested quote +On November 30 2010 07:39 Cambam wrote:
Good to know that another BO optimizer agrees. Btw, I didn't know your app did zerg too, I'll have to check it out. Yeah, I need to find a mod to change the thread title. It'll support Terran too soon (hopefully soon, working on it now). Show nested quote +Does it have creep tumor available as a requirement? I'd really like to see what BO optimizer thinks if you require a creep tumor by 4:30 or something. After watching Ret's game this morning where he had 75% of metalopolis covered in creep, I really think getting that early tumor is significant and would cause the BO optimizer to maybe consider going hatch first to get a 2nd queen faster. Right now, it just spits as soon as each queen spawns, but if it had to use a creep tumor with the first 25 energy, it would probably either get the first queen sooner or get a second queen sooner. You can specify the number of creep tumors you want. I think there's a bug with it though as requesting say 5 creep tumors and 10 drones works fine, but if you do 5 creep tumors and 30 drones it starts going crazy and never satisfies the target. Will play with that one a bit more...
How about a relevant question that is being ignored?
Do all these "optimizers" and "build testers" value drone mining equally, or do they factor in diminishing marginal value relative to increasing marginal saturation?
If they do not do this, then all of these pool first results are wrong. In the testing I have done so far, hatch first is ALWAYS ahead in total minerals mined. Unless someone can produce a replay that can surpass 2710 minerals mined or 44.58 drones by the 6 minute mark, then all of these discussed builds are inferior to the builds leading on my thread.
http://www.teamliquid.net/forum/viewmessage.php?topic_id=172481
This discussion is also irrelevant because at no time do you provide sufficient data or even define the preferred variables in testing a builds "economy" in order to make any kind of meaningful comparison from one to another.
EDIT: Sorry, I double posted this one...
|
On November 30 2010 09:36 jdseemoreglass wrote:
How about a relevant question that is being ignored?
Do all these "optimizers" and "build testers" value drone mining equally, or do they factor in diminishing marginal value relative to increasing marginal saturation?
Yes, they do.
If they do not do this, then all of these pool first results are wrong. In the testing I have done so far, hatch first is ALWAYS ahead in total minerals mined. Unless someone can produce a replay that can surpass 2710 minerals mined or 44.58 drones by the 6 minute mark, then all of these discussed builds are inferior to the builds leading on my thread.
It's interesting, the exact number of drones you request seems to give quite different results. For instance, 44 drones does indeed go for hatch before pool (however it does 11 hatch 10 pool and completes at 5:49), but if you go for 40 drones it does a 10 pool 16 hatch build (completes 5:32), or 42 drones does 11 pool 17 hatch (5:39).
I suspect that the answer is that both builds give you good economies, just at slightly different points in time.
|
On November 30 2010 07:50 Skrag wrote:
I just tested 14pool/16hatch vs 15hatch/15pool vs 16pool/16hatch, on lost temple at 9oclock. (Yes, map and start position do make a difference, you can't directly compare numbers from different maps or even different starting spots)
In each build, I built an extractor at 18, although I think in the 16pool version I ordered it a little differently putting it slightly later, and therefore with more gas.
At the 7 minute mark, this is what they all had:
14pool 16 hatch
625/340 58/60
6 drones 3/17
1 OL 17/25
Spawn 36/40 & 8/40
1 larva
...
You did something wrong, i just tested this exactly how you said and i get:
14pool 16hatch
734/368 60/60
2 drones 12/17
2 OL 9/25
Spawn 3/40 $ just finished
6 larva
|
I like the idea of finely specifying waypoint requirements. I think it would be cool to do something like an earlier poster did and note the timings that a pro like fruitdealer gets his zerglings and creep tumors and drones and such and plug those timings in as waypoints and see if the optimizer can improve. CarbonTwelve's app allows you to require creep tumors.
This will give us a much more realistic BO instead of "fastest way to get 50 drones mining only minerals".
|
On November 30 2010 09:58 CarbonTwelve wrote:
...
It's interesting, the exact number of drones you request seems to give quite different results. For instance, 44 drones does indeed go for hatch before pool (however it does 11 hatch 10 pool and completes at 5:49), but if you go for 40 drones it does a 10 pool 16 hatch build (completes 5:32), or 42 drones does 11 pool 17 hatch (5:39).
I suspect that the answer is that both builds give you good economies, just at slightly different points in time.
But again that is just the fastest way to get x drones, i can guarantee you that 11hatch 10 pool is very behind in minerals mined even if at 5:49 it has more drones and will never catch on an normal build.
Anyway the econ builds should be tested without overdrone because then you tend to equalize things.
What i mean is you should not get more than 2 drones/patch 3 drones/gas.
So if we test 2base mineral only we should stop at 32 drones.
|
On November 30 2010 10:12 icezar wrote:
But again that is just the fastest way to get x drones, i can guarantee you that 11hatch 10 pool is very behind in minerals mined even if at 5:49 it has more drones and will never catch on an normal build.
Yeah, I was thinking it'd be a good idea to set minerals mined as a target. Will add that to the list...
|
On November 30 2010 10:06 Cambam wrote:
I like the idea of finely specifying waypoint requirements. I think it would be cool to do something like an earlier poster did and note the timings that a pro like fruitdealer gets his zerglings and creep tumors and drones and such and plug those timings in as waypoints and see if the optimizer can improve. CarbonTwelve's app allows you to require creep tumors.
This will give us a much more realistic BO instead of "fastest way to get 50 drones mining only minerals".
I do not think you could know what waypoints to take from a pro build.
If you take too many the program if will not find anything better, if you take less it might be that you are missing something. Just think that they have different builds for different maps, matches and even players.
And i am pretty sure the builds that they do are not perfect.
|
On November 30 2010 10:19 CarbonTwelve wrote:Show nested quote +On November 30 2010 10:12 icezar wrote:
But again that is just the fastest way to get x drones, i can guarantee you that 11hatch 10 pool is very behind in minerals mined even if at 5:49 it has more drones and will never catch on an normal build. Yeah, I was thinking it'd be a good idea to set minerals mined as a target. Will add that to the list...
That would help a lot with these testings, maybe also make minerals, gas, larva and supply as targets.
Or even make time a target and only score on minerals,gas etc.
|
On November 27 2010 09:12 IcyPringle wrote:
Instead of debating and rambling on why doesn't a zerg player or op just test out the 3 builds and give us some times/data so we can look at and actually KNOW which build is the best.
this has been tested and discussed on several threads a month back or so. steve bonnell II's stream (aka Destiny) did an hour of head to head with copro testing this and pool first was just as good economically as hatch first in several different scenarios. it would be nice if he would post those replays.
|
On November 30 2010 09:10 icezar wrote:
Yes, but that is exactly the problem with you apps, they give you the fastest way to something.
There's a much bigger problem with them than that. They don't actually mirror reality, giving results that aren't even correct all of the time. If the optimizers all want to come up with 15pool/16hatch, then they're just enough off from reality that they can't come up with the correct answer, because it's possible to show mathematically, after extracting real in-game times of pool and hatchery completion, that 15hatch/14pool is ahead of 15pool/16hatch as an economic build.
|
On November 30 2010 09:58 CarbonTwelve wrote:
It's interesting, the exact number of drones you request seems to give quite different results. For instance, 44 drones does indeed go for hatch before pool (however it does 11 hatch 10 pool and completes at 5:49), but if you go for 40 drones it does a 10 pool 16 hatch build (completes 5:32), or 42 drones does 11 pool 17 hatch (5:39).
I suspect that the answer is that both builds give you good economies, just at slightly different points in time.
And this is why I have no faith in the optimizers for this sort of thing. It's very easy to demonstrate that any 10pool is easily surpassed economically. The faster queen simply does not make up for the fact that you're wasting larvae spawn time, and slowing everything else down.
As my 11Overpool example showed, it's not nearly as bad as people thing, but those builds *will* be at an economic disadvantage, and the fact that the optimizer says different doesn't make it so.
|
On November 30 2010 10:01 icezar wrote:
You did something wrong, i just tested this exactly how you said and i get:
14pool 16hatch
734/368 60/60
2 drones 12/17
2 OL 9/25
Spawn 3/40 $ just finished
6 larva
Ok, beyond the obvious questions (did you do it on lost temple at the 9 oclock position? 12oclock will give different results), it just doesn't matter if I did something wrong compared to you. I did things the same way for *myself* in every game, down to making sure I got the same split at the beginning (and restarting yabot if I misclicked), which still leaves a valid comparison between *my* 14pool 16hatch and *my* 15hatch 15pool.
If you're slightly more optimal in doing everything than I am, that really just doesn't matter. Because whatever inefficiencies I have as a player will be consistent from game to game. Although to be perfectly honest, I'm really suspecting you played at 12-oclock or on a different map altogether, because I'm not 100% on the ball, but there's no way I'm *that* far off.
|
On November 30 2010 11:55 Skrag wrote:
And this is why I have no faith in the optimizers for this sort of thing. It's very easy to demonstrate that any 10pool is easily surpassed economically. The faster queen simply does not make up for the fact that you're wasting larvae spawn time, and slowing everything else down.
As my 11Overpool example showed, it's not nearly as bad as people thing, but those builds *will* be at an economic disadvantage, and the fact that the optimizer says different doesn't make it so.
Personally I'd like to see proof that it's 'easily surpassed economically' before dismissing the optimiser so readily. At the moment I don't have enough time to do any testing or calculations of my own, but I do have a lot of faith in the optimiser I've built, and if it finds that 10 pool 16 hatch allows it to achieve the result faster, I'm fairly confident that that is indeed the case, or at the very least it isn't a significant disadvantage.
|
On November 30 2010 12:15 CarbonTwelve wrote:Show nested quote +On November 30 2010 11:55 Skrag wrote:
And this is why I have no faith in the optimizers for this sort of thing. It's very easy to demonstrate that any 10pool is easily surpassed economically. The faster queen simply does not make up for the fact that you're wasting larvae spawn time, and slowing everything else down.
As my 11Overpool example showed, it's not nearly as bad as people thing, but those builds *will* be at an economic disadvantage, and the fact that the optimizer says different doesn't make it so. Personally I'd like to see proof that it's 'easily surpassed economically' before dismissing the optimiser so readily. At the moment I don't have enough time to do any testing or calculations of my own, but I do have a lot of faith in the optimiser I've built, and if it finds that 10 pool 16 hatch allows it to achieve the result faster, I'm fairly confident that that is indeed the case, or at the very least it isn't a significant disadvantage.
uh, hey guys... I've pointed you in the direction of tests and results several times now. We have significant data on 10 pool, overpool, 14 pool, 16 pool, 14 hatch, 16 hatch, and more... You can use this to compare to the results of the optimizers, unless you think there is a significant flaw in our testing methods.
http://www.teamliquid.net/forum/viewmessage.php?topic_id=172481
|
Skrag has done significant testing of his own and is very experienced at it; he's actually been one of the few testing these things out. Adding his own results to that pool (no pun intended) would be great. More data points are always good.
|
On November 30 2010 12:22 jdseemoreglass wrote:unless you think there is a significant flaw in our testing methods.
The biggest flaw is in your scoring, not the methods.
|
On November 30 2010 12:22 jdseemoreglass wrote:uh, hey guys... I've pointed you in the direction of tests and results several times now. We have significant data on 10 pool, overpool, 14 pool, 16 pool, 14 hatch, 16 hatch, and more... You can use this to compare to the results of the optimizers, unless you think there is a significant flaw in our testing methods. http://www.teamliquid.net/forum/viewmessage.php?topic_id=172481
Yeah, and this basically confirms that the economy of 11 pool 18 hatch is pretty close to 16 hatch 15 pool (based on there being slightly fewer minerals mined, but more drones meaning higher income, which should balanced out the minerals mined), which is what I was suggesting. That's why I say I'd like to see some proof from Skrag that 10 or 11 pool is 'easily surpassed economically'.
|
Whenever i go FE with Zerg i always find myself having extra minerals in mid to late game. I have so many minerals (around 800-900) that i don't feel like expanding again. I'm rank 25 silver so i'm still pretty noobish. I'm not sure what i'm doing wrong? i am always injecting and building units.
|
On November 27 2010 22:06 UncleOwnage wrote:
There's one factor nobody takes into consideration; minerals aren't yours until mined/secured. This of course, requires a bit of explanation.
If you can't secure an extra base, even if it's your natural, you have no way of knowing if you'll ever get those minerals. Usually, your main is secure by default, only in rare cases will you not be able to mine your main out.
This brings me to the main point: 20 drones spread on your main and natural is better than 20 drones on your main, because with 10 drones on your natural, you will mine minerals you haven't (completely) secured yet. This will leave your "secured" mineral patches in your main left with more minerals, giving you more time mining there.
It's a sort of meta-game theory, in that some would argue "minerals mined are minerals mined, no matter where from", but I feel (and this is very crucial, this thinking is VERY subjective) that mining from any other base than your main as early as possible is beneficial.
If you later have to give up your natural, you will then have more minerals in your main to fall back on. This theory might also be more important for races that can one/two-base efficiently (I'm protoss, for the record).
This is why if you are sneaking an expansion(any you can't easily defend) you should always take one of "their" expansions. Then you guarantee yourself more potential income throughout the course of the game. I guess this is a little off topic...
|
Since this is Team Liquid I would like to hear what Ret has to add to this discussion. I've never seen a zerg economy like Ret's.
Ret, we need you 
|
|
|
|
|
|
EPT
