No, that's exactly why we want to compare them - to see what negative effects a subset has compared to using all flops. But you can't compare them without having the same pre and post configs.

How can you compare them even if you have the same gametree (config as you call it) but different number of flops ? I can't think of a reliable algorithm for that right now, although I wouldn't exclude the possibility of one existing.

One way I used in the past is looking at both solutions and seeing if some things don't jump in the eye as outright wrong. Or if one solution "makes sense" more than the other but this is tricky as the brain can think a pattern is right when in fact it is slightly wrong.

The solution with all flops is the "correct solution", so you just see how much the solution with fewer flops deviates from it. If there isn't much deviation, it's a great approximation. You hold everything besides # of flops constant so that you know any differences are due to difference in flop subsets.

Yes but this assuming you use the same solver to generate both. As far as I know there might be many equilibrium solutions for the same problem. Actually might be interesting to compute, generate a solution for the same tree with same solver but small set of flops, compare preflop see the difference.

Yes, multiple equilibria are possible, although it's unlikely they would be significantly different.

Anyhow, I still fail to see how your software is in any way superior.

Where do you get there being 1833 unique flops? That isn't correct, fwiw.

You say you're using bucketing, and then you say that your results are better because you compared your own sims ran on smaller subsets to full subsets. I'm way more suspect of your bucketing techniques (I've only heard of people doing this wrong) than small flop subsets.

FWIW there have been plenty of comparisons of small subsets (~80) to big subsets like half or even all flops. Obviously there will be some bias in the smaller subsets, but they compare surprisingly well. So, seriously, if I had to put my money on anything it would be bad bucketing.

Oh, and the best-response calculating. There are 3 major competing programs in this field and they are all in agreement about their MES calculating. I'd say it's safe to trust it's bug free.

Wrote my own python script.
Order hands by card value.
Then fix first card suite to heart.
If in original flop first card suite was equal to second card suite then set second hand suite to heart else diamond.
something similar to 3rd, set it to heart,diamond or cross depending on original flop.

And so I get 1833 flops. I know theory says 1755, did a compare and I have 78 flops like XhXdYd that are perfectly equivalent to XhXdYh. So there is the difference (9h9d4d could be encoded to 9h9d4h). In my code I have the right frequencies for those flops so there should be no problem there


Indeed two problems with bucketing, one is doing it right and the other is using the buckets in the game tree as hands in the same bucket on flop might land on different buckets on turn so the game tree becomes a digraph. So far I checked out my clustering and I think I got an ok-ish bucketing. As for the second part this is what I am mostly proud of, I devised an algorithm that goes in 4 passes (one for each street) that so far seems to work. I understand the suspicions, it is something that is hard to get right.

What if the bucketing is done right ?
Would not this get better results with all flops sets than using a subset and no bucketing ? Just asking!

What is MES ?
Wasn't talking about those programs, as a programmer I was saying that if a solver uses the same code for solving it and computing best response (with small changes) than a bug in the code would lead to buggy best response.


On a side note, I sense a lot of negativity here and can't explain. I am not here to badmouth piosolver or GTORB, I respect those programs and their creators because as a programmer I have a good idea what goes on in working on such a beast. What I have is a research project of mine that I worked for past two years fulltime and a bit more before that part time and for the first time I found an algorithm that allows computing an equilibrium using all flops, not a subset. At the moment it reached a very stable phase and can be developed in further directions and I love working on it. You guys should be happy that in addition to existing solvers there is another project that tries something different. Lighten up! If anyone would be willing to get involved/be a backer for this project then it's good, if not it's also good! I appreciate the talk about GTO in this thread and I love responding, but in some posts I feel like I'm being put to wall in front of the execution brigade! Maybe it's my fault for phrasing some things not in the right way, but hopefully I explained things better now!

i sense the same hostile undertone limax, i don't think it is your fault. thank you for bringing up this interesting discussion.

I don't think anyone's trying to be hostile (and sorry if I came off that way). Some of us just think that limax hasn't given any substantive reason to think his software is any better. When there are already great alternatives out there, something new has to go above and beyond them. But so far there's been no indication that limax's is superior in any way. There's nothing new about its functionality. It could be more efficient, but at this point there's no way to tell unless he runs some calculations. Basically the burden of proof is on limax, so I think I'm justified in being doubtful at this point in time. And I don't think that's negativity, really, just a healthy skepticism. If, like limax says, there really is something "new" about his solver, great! All I'm saying is, please convince me!

people fear change and dont like others being successfull hence grilling the **** out you, gl with the project look forward to seeing it develop glgl


Sent from my iPhone using Tapatalk

Yes, scared indeed. What will I do with myself if a new solver is invented!

These are questions that will be asked of you eventually if you continue to move forward with this. I don't understand why people think it's a bad thing and label it "hostility" or say "he's just afraid of change." If anything it is healthy discussion and is giving limax an opportunity to clarify things. He wasn't exactly comprehensive in his OP.

Oh, stop it with the talk of hostility. You know that's how these discussions always need to go. You make statements, people test you on them, you get the chance to prove how right you are.
It's all love over here. I would be super excited to see advancements here. And no matter what I think it's impressive that you've built a project up this big. It's more than anything I can say.

I'll be back to argue with you when I wake up

Advancements in solving will really take off when we all share code and work together on an open source solver.

Do all the CRUSH players ABARONE etc use these PIO GTO solvers? I dont understand half words talked!!!!

any 6 easily beats you, it's hard for some one to have 6/9 here but it's possible and the least he could be calling here with is trips. the board paired so the boat is possible you could have bet 1/3 the pot or maybe less than half and then access the hand from there

Are there any good US friendly sites for HUSNGs?

Black Chip/ACR and Ignition are two with good games and reliable cashouts.

Hey guys, is there any script to hide bankroll playing spins? I tried that few years old ahk-script, but it covered on the wrong place. PS has its own script at HU-sngs, but not on spins. It would be very much +EV for me if I didnt see bankroll everytime I register..

Do you mean on the buy-in popup dialog? If so, you can just set the option to not show that dialog when you register (also makes registering a one-click process).
In settings > gameplay > multitabling > 'allow registration without showing buyin dialog..'

does rakeback change next year or is it ~30% for supernova still

It's ~30% for a little but it moves to monthly i.e. we'll need to earn x vpp per month to stay supernova. At some point in the new year this is going to end for the new rewards program. We don't know what that involves but it would be shocking if it didn't strongly favour degenerates spewing money at the pokerstars' casino.

Any 2016 supernova will retain the status with no monthly requirements until the new program is implemented. Non-supernovas can become SN in a single month instead of a year. Those are both significant improvements.

Not saying the as-yet unknown 2017 program will be amazing for huge grinders, but why not keep an open mind.

Can you explain why degenerates spewing huge money in the casino should not get top tier rewards, which presumably should instead be given to cartel founders?

There is no competition between products. A generous VIP program for casino players does not therefore mean less rakeback for poker players. Remember that most players would prefer Spins and HUSNGs removed from the site entirely, in order that their unpopular game of choice may get a few extra fish and make it more profitable / viable once again (and they would be wrong, of course).

Rakeback is also going down a couple % right?

IIRC to 28% but I couldn't find the source on a quick search. Gut feeling is "maintaining Supernova" bonus is removed.

I don't like to see uneducated and/or addicted people actively manipulated into putting money into "no-skill" games. Call me old fashioned.