This is exactly my point by the way, and exactly why I'm bashing YOU and not the build.
You have obviously made what seems like an honest and understandable mistake (the third time I watched the replay, I figured out exactly what you must have done by being momentarily confused myself). But when you make this mistake (YOU making the mistake, not me), instead of trying to figure out what's going on, you flame away and tell me to stop trolling.
Which is how you've treated every single criticism or question in this thread, which is a HUGE disservice to a build that is clearly viable and useful, but that many people will ignore simply because you've been such a gigantic moron.
On December 11 2010 11:56 Skrag wrote: you've been such a gigantic moron.
It is statements such as these that are the reason I am calling you a troll...
I'm sorry, in normal replays clicking on the player was enough to see their mineral count. I am analyzing the replay data now...
Yes, and a normal person would have tried to figure out what was going on (since I was clearly able to watch the replay), and *asked* instead of immediately opening up with the flamethrower.
So yes, you've been a gigantic moron, and frankly I'm pretty pissed off about it.
No, but seriously: these last few pages has contained a lot of bashing back and forth. We need to make this an constructive conversation again.
What are you really discussing? Who's bashing who? Which scientific method is best for evaluating build orders? If 11 pool 18 hatch is a viable build? If the OP build is the most economical Zerg build? Which thread to discuss this in?
On December 11 2010 11:56 Skrag wrote: you've been such a gigantic moron.
It is statements such as these that are the reason I am calling you a troll...
I'm sorry, in normal replays clicking on the player was enough to see their mineral count. I am analyzing the replay data now...
Yes, and a normal person would have tried to figure out what was going on (since I was clearly able to watch the replay), and *asked* instead of immediately opening up with the flamethrower.
So yes, you've been a gigantic moron, and frankly I'm pretty pissed off about it.
I didn't flame anyone. I just want you to stop bashing me on my thread. Be a little more courteous and calm down. That's all I want.
So I took the results. At 3:00 the player has 1380 compared to my 1324. At 6:00, the player has 3995 compared to my posted 3879.
There is a complication due to the fact that the replay provided paused for a couple seconds before beginning. If you would like me to concede that an AI beat my results, then fine, I concede that an AI beat my results. I'm not sure how you got the 150 minerals, and honestly I don't really care. I'm just eager to get this behind us already and focus on the actual build.
On December 11 2010 12:13 Blueblister wrote: What are you really discussing? Who's bashing who? Which scientific method is best for evaluating build orders? If 11 pool 18 hatch is a viable build? If the OP build is the most economical Zerg build?
1) I'm bashing JD, he's bashing me.
2) Clearly any testing method that removes the human error component, while still actually being a test of the actual game, rather than a simplified build order calculator, is superior.
3) It's absolutely a viable build. The psychological effects of the super early pool alone are worth adding it to your repertoire even if there is some economic sacrifice.
4) The OP's own data shows that hatch-first builds are superior economically, but he's made the claim that it is more economic than any other pool-first build, and that claim appears to be false, with 13p15h being the likely winner economically, with a nice compromise between a fast queen, a reasonably timed hatchery, and no early larva loss.
First of all, i've been following the "debate" (well, it was one at the beginning) from the first thread. I have been really curious about this build, so i tested it on about 100-120 games (ZvP and ZvT, no ZvZ since i never prepare BO in ZvZ), mostly training games with teammates (2000-2400 range).
So, yeah, it's pretty surprizing how efficient it is, for a first pool build. After all these games, my conclusion is that it is a pleasant alternative to 14h15p, yet, i prefer to hatch first. I've got the feeling that it isn't safer at all : bunker rush is the same, you are supply block at 18, which is very very annoying, and makes mistakes easier to be done, creep is very late, which is especially important in the current metagame, and finally it does not fare better against 2rax than 14h15p. Interesting things to note is that you have some chance the guy may overreact if he's not used to it, which makes in those cases the build very viable.
Anyway, i'm reverting to 14h15p, because it feels more solid play to me. But i may adopt it if there are high level play displaying it (by high level, i mean top progamers).
Now, on another hand, I have to tell that I was totally DISGUSTED by what the way the 2 thread turned. First of all i was enthousiastic, really pleased to see something constructive about early zerg come out. Then, all became revolving around one build, and the OP began being totally biaised about it. Basically, you have been ignoring or flaming any critics, accusing them most of the time of trolling. You only see what you wanna see, it feels like this build is your baby, and you have to defend it no matter what, because your baby is the most beautiful baby in the world. No matter what people say. For someone like me who really enjoyed your effort at the start, it leaves a really bad taste.
I had a hard time carrying my experiments with this build, just because of your attitude. Which is really sad (and counter-productive for your baby).
Since i'm not saying that this build is the best build ever, etc, etc, i'm not welcome here, but that were my 2cents anyway, for those who actually care about unbiaised POV.
PS : You should remove the question mark, since apparently you have now stated you would not accept further critics, constructive or not. Also same idea, the [D] is irrelevant now.
PPS : Excuse my english.
PPPS : Don't get me wrong, i appreciate the efforts and the build a lot. I just don't like the atmosphere of the threads and the alleged debate.
Guys, please, stop attacking me. There is no need for ten people to pile up on the OP. I haven't flamed anyone. I'm just tired of people spouting bullshit on my thread, so I have to defend myself until they provide actual evidence.
Skrag has finally provided a replay that I can analyze, so I have conceded this his testing method has yielded superior results to the 11Pool. Can we stop all the attacks now? I know this is the internet, but this is getting fucking ridiculous...
On December 11 2010 12:16 jdseemoreglass wrote: I didn't flame anyone. I just want you to stop bashing me on my thread. Be a little more courteous and calm down. That's all I want.
Really?
Guys, the replay is garbage. I can't even see the mineral count... Please just leave my thread alone and go bash my build in your own thread. This is just a waste of time.
That sounds a whole hell of a lot like a flame to me, calling the results "garbage" and "a waste of time".
So I took the results. At 3:00 the player has 1380 compared to my 1324. At 6:00, the player has 3995 compared to my posted 3879.
How are you measuring, and how are you ensuring you hit right at the 6:00 mark? I measured using the spending + banked resources method, and the 13p was 156 ahead. As for ensuring you hit the mark, if you let the timer go to anywhere between 6:11-6:19, pause, and then hit 'b' once, it will back up to exactly 6 minutes.
There is a complication due to the fact that the replay provided paused for a couple seconds before beginning. If you would like me to concede that an AI beat my results, then fine, I concede that an AI beat my results. I'm not sure how you got the 150 minerals, and honestly I don't really care. I'm just eager to get this behind us already and focus on the actual build.
An AI beat your results, but that same AI showed 11pool being economically *behind* the 13pool.
So can we finally put this to rest:
Please do not post arguments that rely on the assumption that this build is behind. It has been days since I posted the replays and I haven't seen anyone beat the posted results, so hopefully we are done with that.
Since it has been shown that human error is a big enough factor that your replays aren't truly valid comparisons, and that now that we do have a completely consistent and repeatable way of comparing builds, those comparisons show 11pool to be economically behind other pool-first builds?
As for you asking somebody else to be courteous and calm down, well that's pretty much laughable from every angle.
Oh yeah, and lets not forget your post in the other thread where we have actually put a ridiculous amount of effort trying to determine if YOUR claims were correct.
Lol this thread is funny...
I've already shown with empirical evidence that 13P/15H is inferior economically to 11Pool, and the debate continues...
I know you guys are using some program or something for all your results... When you can post results than can beat this, then let me know.
Yeah, laughing at other people's hard work, without even taking the time to attempt to understand what's going on, was totally courteous of you, and not bashing at all.
Without knowing that the replay didn't allow clicking players, then the replay was garbage to me, and a waste of my time. If you call such a statement flaming, then we clearly have different definitions. I never attacked or flamed or bashed anyone, in my opinion. You on the other hand repeatedly attacked me personally, even resorting to name-calling, and when I ask you to be courteous you say it is laughable...
To reach the 6:00 mark, I used the replay bar and moved it to exactly 6:00. At that time, the resources spent = 4650 + resources owned = 45 - initial resources spent 700 = 3995
If you provide a replay of the 11Pool build using the same methods, then I will be happy to update the OP graph to reflect the results.
On December 11 2010 12:28 Skrag wrote: Oh yeah, and lets not forget your post in the other thread where we have actually put a ridiculous amount of effort trying to determine if YOUR claims were correct.
I've already shown with empirical evidence that 13P/15H is inferior economically to 11Pool, and the debate continues...
I know you guys are using some program or something for all your results... When you can post results than can beat this, then let me know.
Yeah, laughing at other people's hard work, without even taking the time to attempt to understand what's going on, was totally courteous of you, and not bashing at all.
It was funny that so much data was being used without anyone being able to actually watch a replay and verify the data. I requested the data once it became available. Again, if you call this flaming, then clearly we have different definitions.
You know, the thing that pisses me off the MOST about all of this is that there are some *really* cool things about an 11 overpool build, given that it can turn into a full-blown economic build that can compete with just about anything, but just as easily turn into super early aggression, but the way you have dealt with criticism in this thread has pretty much chased off everybody who could actually provide useful input, leaving most of the thread full of flamewars.
Hell, you even chased me off into another thread to try to get real, consistent, reproducible measurements, and I was one of the ones most vehemently defending the build early on in the thread.
Your blind fanatic single-mindedness has done far more to damage any real discussion in this thread than any of the critics have.
On December 11 2010 12:40 Skrag wrote: You know, the thing that pisses me off the MOST about all of this is that there are some *really* cool things about an 11 overpool build, given that it can turn into a full-blown economic build that can compete with just about anything, but just as easily turn into super early aggression, but the way you have dealt with criticism in this thread has pretty much chased off everybody who could actually provide useful input, leaving most of the thread full of flamewars.
Hell, you even chased me off into another thread to try to get real, consistent, reproducible measurements, and I was one of the ones most vehemently defending the build early on in the thread.
Your blind fanatic single-mindedness has done far more to damage any real discussion in this thread than any of the critics have.
I am not fanatical and single-minded... I just want all claims to be supported by evidence. Any arguments or criticisms that fail to reach this criteria will be ignored or criticized. You have provided actual evidence, and I am not being fanatical. I have conceded 3 times now that your testing method is better. You have made about 10 posts attacking me in this thread today alone, and I would appreciate it if you would either start being civil and apologize for the personal attacks or leave the thread altogether.
I was using 3839 for the 18pool resources, which is what I measured from your replay a while back. If the replay has since then been updated with a better one, then that explains my 150 vs your 116.
Jacobman has posted replays of all the builds in the other thread. Looking at his 11pool replay, it shows 3914 at 6 minutes, indicating that your 11pool build was much closer to optimal than your 13pool build, which is only to be expected given how much practice you have on the 11pool, and is part of the reason an attempt to objectively measure was needed to begin with.
My version shows better results, with 4095 minerals for 13pool, and 3975 for 11pool, mainly because I did true total optimization for the first 19 drones, loading up the closer mineral patches with 2 workers as soon as possible, and sending all of the first 19 to a patch that would not interrupt any mining time by having bouncing workers (or minimizing the bouncing in cases where it was impossible to get rid of entirely) but the 11 pool is pretty consistently about 100 minerals behind at the 6 minute mark.
Having said all that, there are some very real benefits to pooling at 11 rather than 13 or 14 that very well could be worth a small sacrifice of minerals. But you haven't let the thread be about what sacrifice is worthwhile, claiming only that there was no sacrifice at all, even when the evidence pointed to the contrary.
On December 11 2010 10:57 jdseemoreglass wrote: Skrag, I believe that the first rule of the strategy forum is "everything you say must be supported by evidence."
If you would just provide any evidence at all that my data is flawed, then I would listen to it. You have no data, only claims that my testing is flawed. I have my evidence. Until anyone in this entire thread can post a single replay outperforming my 11Pool replay, then I would appreciate it if everyone would stop repeating things that cannot be supported by evidence.
It's not your data that is flawed. Your data is fine. It's your methodology. The problem is that your graphs are based off of single on the spot trials. There is a lot of variance in each test, so that data doesn't really represent the builds. You didn't even use the best replays available at the time for the graphs.
Really the only way to be able to compare things well with your method is to do each build 10-20 times. Trials where the build was messed up need to be thrown out. After all of that you can compare the averages which is much more indicative of each build.
AI allows you to not worry about being consistent with each trial. I would guess you would just need more trials if you were to do it by hand to make up for the inconsistency in human play.
On December 11 2010 12:56 Skrag wrote: But you haven't let the thread be about what sacrifice is worthwhile, claiming only that there was no sacrifice at all, even when the evidence pointed to the contrary.
I denied that there was no sacrifice, because my tests showed there was no sacrifice, and no one was able to provide data to the contrary until about 5 days later lol...
Just for reference, I didn't supply the replays to compare to other replays economically. I did not spend time picking the best replay for each build. There are likely replays that yield better final results. The replays are just one of my data points that I used to get the average numbers I posted. I simply put one of the trials up in order for the AIs execution to be able to be looked at in case I made a mistake with the AI in one of them.
Once the opportunity for flaming disappears, Skrag leaves as quickly as he came...
No apology for turning my thread into a childish mudslinging battle for hours... He attacks me and calls me a jackass twice, a moron twice, a blind-single-minded fanatic, and so on, and he won't even bother to end with a civil comment after I give him what he wants.
Trying to provide for the TL community is really not worth the trouble it entails... which explains a lot.
EPT
