Your "Road Map" above has heads up as your #5 enhancement. Will you enter after adding heads up?

Paul,

I'd welcome your participation in this thread: http://forumserver.twoplustwo.com/29...botter-904198/

Same goes for everyone else of course.

So it does not use a Kalman filter?

Does the program treat variables as probability distributions (ie ranges are expressed in terms of probability not a discrete figure)?

Does the program use correlation inference. Ie does it use other statistics to refine probabilty. IE if there was in general a postive correlation between UTG PFR and 3bet does the program account for this?

link?

not sure that I'm allowed to post one but it's pretty easy to find.

No Kalman filter.

Yes, the program thinks about things in terms of: what we have vs. what we think our opponent has (i.e., probability of each possible hand) and what we think our opponent thinks we have (i.e., probability of each hand).

The opponent model thinks about hand strength on a relative basis (e.g., sometimes a straight is a crappy hand and sometimes top pair top kicker is a great hand).

The program has a point of view about how people will play poker in each position preflop and also knows things like average opponent's preflop raise %. It then will infer how the opponent will play in each position preflop (despite the fact that it only has limited information).

Not sure if that answers your question..

We'd love to! Maybe not version 1.0, but 2.0!

Also, we'd love for them to add NL multiplayer (i.e., what everyone plays).

The author estimated that it could probably be okay up to NL100. AFAIK Ivey is a proven winner at slightly higher stakes.

And even though people are trying to stir up a scandal in this thread and in the other, for some reason I'm not feeling as much infuriated as I probably should be(?) and I hate botters and what they might even further do to the games in the future as much as the next guy. Maybe it's because if this software had the capability of improving my game, I would use it even if it was written by Bernard Madoff. I only tried it for maybe 5 minutes, and it couldn't import the hands from the site I play on, so whether it can be of any help or not is another story altogether.

Where do you play?

Party pokah!

Good! We should have you working in a week or so.

People should just be aware that they can't trust this guy. If you're aware of that and still have no problem installing software he wrote than that is fine.

Clam AV says that your setup file contains "PUA.Packed.Armadillo".

False positive?

My understanding is that Armadillo is an exe compression tool (and possibly a license manager).

So it could be that they used armadillo to protect their code.

Neko beat me to it

Yes, Armadillo isn't anything to worry about and is quite common (it was used by PokerAceHUD IIRC and possibly PT3 uses it as well [same author]).

Juk

Any chance this works on Microgaming? I'd like to give it a try.

create a mailing list to be notified of updates. I for example will consider your product when it supports heads-up play, but don't feel like checking this thread or your site once a week until that happens.

Not yet. But, we'll add it. I'm not sure when we'll get to it at this time.

Yep. That's what handles the registration. When you set up our software, there is a single visit to the Armadillo server to confirm your license.

That's the only time any remote server is contacted.

I think Armadillo is owned by Digital River.

Okay. It'll probably be 5 months or more before we have Heads Up.

Are there plans to add fixed limit to the program?

We aren't sure. Eventually we probably will. We'll add tournaments next.

Then likely heads up.

Then limit is in competition w/ pot limit omaha.

Unfortunately, we probably won't get to fixed limit and/or PLO for a year or so.

Here's an interesting question:

I read in the NVG thread that the AI could beat up to about NL100 - what is is about the players at NL200 and above that makes them harder to beat?

Juk

You want to know why there are not many requests for tournaments? I do not play tournaments often and so I have limited knowldege in this area, but perhaps tournament players are less interested in this product because there are so many good ICM training programs already out. Sure the early stages of tournaments do need something like this, but I have read that tournament players generally agree that later-stage, high-blind play is what separates the winners from the losers. Thus, the demand for a program has already been met very well by those ICM programs. And if tournament players want to practice early-stage strategy, they can just use the cash-game part of the program. Thus, much of tournament play is covered. Personally, I would prefer it if tournaments was at the bottom of the list.

I also do not care if Party and iPoker are added, but that is just based on what I want, and I have no real argument to help push what I want here. I can understand why it would be high priority for you, so I guess there are just some things we must learn to accept. However, keep in mind that if you start adding one site after another, then each time a site modifies the format of its hand-history, it will take up some chunk of your development time.

I am very interested in the AI enhancement. But from what I understand, you already have better AI, and you did not roll it out. Will the enhancement be to just roll out more of what you already have, or will it be something else?

Cliff Notes to what I am most interested in:
(0) Fix bugs
(1) Major AI enhancement (what will this be?)
(2) Add heads up
(?) Add support for Hold'em Manager and Poker Tracker

I put a "?" next to HM and PT3 support because I am not sure about how important that is to me, but it is important.

At $200 and above, the pros play. There are some really excellent $200 buyin pros, and people can make nice livings at that level if they are very good (multi-tabling). There are few pros who make a living at $100 and below. I see a few playing Rush at $100, and there are some in general.

However, if someone has the intelligence and discipline to be a winning pro at $100 and below (but can't win as much at $200), then they will likely find better work than making $30K/year as a low stakes grinder.

At $200 and above, our AI's weakness at the 3 and 4 bet preflop game would get exploited enough by the pros that it would probably only make a little profit or breakeven.

While I say there aren't many pros at $100 NL there are a lot good players; I suspect 100NL online plays about like $1000 in a live card room.