We're planning to release v2.0 of Online Hold'em Inspector in early
December, 2003. One of the new features we're thinking about adding is Auto
Play option. We never wanted to add this feature but there's so much demand
for it that we might actually do it.. So we're wondering what the poker
community think about this possibility.
The Auto Play feature will allow Inspector to play Hold'em based on the
profile you've created. We'll provide all the user interface for you to
create a profile; so the Inspector will play as good as you programmed it.
It will be a very tedious process but some may find it worthwhile. We'll
include one sample profile -- average player in a typical full game. We'll
also include a simulator so you can test your profiles.
How do you think such program will affect the online game? Last thing we
want to do is kill the online poker. Please tell us why we should or not
include Auto Play option. At this time we're about 75% for including the
feature and 25% not.
Thanks in advance for your thoughts,
Robert
www.pokerinspector.com
"Instant Odds for Online Poker"
I seriously think you could make money at low limit Party tables by
saying "only play group I hands, always raise preflop and bet/call postflop"
_________________________________________________________________
Posted using RecPoker.com - http://www.recpoker.com
As for me personally I don't want to play with completely automated
grinders. I predict that their amount will momentally grow up to
unreasonable number making low-limit games steady and unprofitable. I will
ask online poker rooms I play at to ban your software. Many people will do
it together and publicity will be huge and awful. Your software
compatibility will force new players redistribution between rooms - those
who want to use it against those who avoid it.
Hopefully that you chances are small - players and poker rooms will be
interested to stop your business. Remember "Scoopmonster", project
advertised here about 6-8 months ago. They gone out of service after about
one week. They hurted just one of smallest and low profitable rooms -
Truepoker. May be they got some money from Truepoker, their first and last
market for auto-play feature, for closing the service and may be not - it
is too big money for one person to feel secure against. I don't know what
are moral standards of Carribbean offshore owners but retaining 100-300M
per year may allow different defensive measures for some of them. I
suggest you to find author/owner of Scoopmonster to discuss what happened
before going same way.
Soleo
Nah they could just scan the screen, and read all the info they need.
Then move the mouse to the appropriate place to click
fold/check/bet/raise. The gameroom window would simply have to be in the
right place. This screen scanning would be a difficult but certainly
manageable programming project. I'm certain some people have already done
it. As for commercializing it, I'd think the online poker rooms would be
very upset. Then again, I doubt there is much they could do to stop you.
I also doubt there is much anyone could do to stop you, since online poker
is in such a legal grey area.
O-PG
------------------------------------------
Start winning hundreds at Texas Holdem with Power Holdem+
http://www.online-pokerguide.com
>Are we ready for a Bot program?
ready or not, its inevitable. eventually, online poker will be about who can
afford the best best bots.
Hello,
It is only a matter of time before somebody does it.
If the bot is anygood it will kill online poker untill players can be sure
they are not playing against bots.who wants to play a bot?
patrick
"Poker Inspector" <ja...@pokerinspector.com> wrote in message
news:vnlu7ra...@corp.supernews.com...
>>It seems to me you can't make autoplay work
>>without some kind of development access to their interfaces.
>
>
> Nah they could just scan the screen, and read all the info they need.
> Then move the mouse to the appropriate place to click
> fold/check/bet/raise. The gameroom window would simply have to be in the
> right place. This screen scanning would be a difficult but certainly
> manageable programming project. I'm certain some people have already done
> it. As for commercializing it, I'd think the online poker rooms would be
> very upset. Then again, I doubt there is much they could do to stop you.
> I also doubt there is much anyone could do to stop you, since online poker
> is in such a legal grey area.
>
> O-PG
>
No. You can scan the screen regardless of screen position. Kinda
technical, but a windows application (like, say, Party Poker) can be
"read" through the Windows API. I'm sure that's what their program
already does (it "reads" the boards/your cards of several poker sites
and gives you statistical information about the hand as you play). It's
interesting that some people seem to think this is the "difficult" part
of making a bot. Reading the screen display of a windows application and
automatically clicking the "check/fold/raise button" is easy.Heck, you
can be using the same computer using another application at the same
time (it's doesn't "take over" the mouse).
> > It seems to me you can't make autoplay work
> > without some kind of development access to their interfaces.
>
> Nah they could just scan the screen, and read all the info they need.
> Then move the mouse to the appropriate place to click
> fold/check/bet/raise. The gameroom window would simply have to be in the
> right place. This screen scanning would be a difficult but certainly
> manageable programming project. I'm certain some people have already done
> it. As for commercializing it, I'd think the online poker rooms would be
> very upset. Then again, I doubt there is much they could do to stop you.
You wouldn't need access to the development interfaces, nor would you need
to scan the screen. You would just need to reverse-engineer the protocol
that the software uses to talk to the server. This has been done with
many instant messenger programs, and it isn't difficult if you know how to
do it. Once you know how the server tells the client what's going on in
the game, and how the client tells the server what action to take, any bot
writer can send the same data that a "real" client would, except it would
determine what to send automatically instead of prompting the user.
Of course, if anyone did this the online sites would probably find ways to
block unofficial client software. But it would be neat to try :)
"Poker Inspector" <ja...@pokerinspector.com> wrote in message
news:vnlu7ra...@corp.supernews.com...
_________________________________________________________________
_________________________________________________________________
Online sites could easily use encryption or simply change the format of the order
in which data is delivered. All poker software updates easily, it's no big deal
as a user to wait 60 seconds once a week while the software is updating.
In reality most sites will ignore the Bot program since if it really could make
money, the people who built it would keep it to themselves.
> Hello,
>
> It is only a matter of time before somebody does it.
> If the bot is anygood it will kill online poker untill players can be sure
> they are not playing against bots.who wants to play a bot?
But once you have a bot's strategy "figured out", it wouldn't be any
different than playing a human player. Except that the bot may be more
predictable.
_________________________________________________________________
Heh. That's funny.
>It will be a very tedious process but some may find it worthwhile. We'll
>include one sample profile -- average player in a typical full game. We'll
>also include a simulator so you can test your profiles.
>
>How do you think such program will affect the online game? Last thing we
>want to do is kill the online poker. Please tell us why we should or not
>include Auto Play option. At this time we're about 75% for including the
>feature and 25% not.
I think it's a great idea. After checking out your software,
I'm almost 100% certain that any bot you write will be utter
rubbish. It'll just add one more kind of fish to the ocean.
:)
- Andrew
Oh, you should totally do it. I don't think you're supplying nearly
as much of the material work as you think, so all such bots will probably
lose except in play money games. But if you don't do it, someone else
will -- and there's no reason not to do it anyway.
Have a ball! It'll be fun.
--
Paul Phillips | Where there's smoke, there's mirrors!
Caged Spirit |
Empiricist |
up hill, pi pals! |----------* http://www.improving.org/paulp/ *----------
Something along the lines of "Captchas" (named by Udi Manber) or client
puzzles which would make it much more difficult to have a bot involved.
Yahoo and Ticketmaster are a just a few companies already using this type of
technology to prevent bots from purchasing services or registering user
names.
Obviously there are plenty of technical issues involved with this type of
approach that would be outside the scope of this email. However, it would
seem to me, to be relatively easy to implement something along these lines
to prevent or thwart the use of bots in the online gaming world.
I'd imagine most of the major poker sites are already thinking along these
lines in preparation for what would seem to be the inevitable.
Once bots become prevalent in the online gaming world, having such a
technology would prove to be a big competitve advantage.
Just my $0.02.
Sean
"James smyth" <starl...@yahoo.co.uk> wrote in message
news:DEEeb.6031$Nu6.50...@news-text.cableinet.net...
"Poker Inspector" <ja...@pokerinspector.com> wrote in message
news:vnlu7ra...@corp.supernews.com...
Soleo
It is easy for poker client application to detect specific software
running at player's computer at current moment and then notify house
security department. I think that poker room must add user's permission to
do this into agreement. Yes, so far you can't prevent functioning of
private bots, but publically known and advertised at least somewhere at
least once are dead horses being reported to security and included in such
monitoring. Keeping list of such applications updated is our common task
and we must notify room's security about any new known software at the
market of this kind.
Detected use of such program say at PartyPoker, according to their terms
is good reason to close such account and take the money: "if it is
determined by Company that you have employed or made use of a system
(including machines, computers, software or other automated systems)
designed specifically to defeat the system".
Hey, bot users do you like to lose your bankrolls at PartyPoker and be
banned there? Excuse my malevolence, I like the fact that you will :-)
Also I would like to ask you to pay attention that current functions of
Poker Inspector are already in gray area according to PartyPoker's Terms.
Just nobody cares and asks to stop it working. But for fully automated bot
I think situation will be very different. So better not awake this monster
and stop there where you already stay.
Good luck all and be prepared to force online sites to fight some specific
applications running during the play!
Soleo
Remember scoopmonster? Do a google search, anyways, they created a bot
for TruePoker, and TruePoker changed their interface so that
scoopmonster didn't work anymore. Scoopmonster calculated odds much
like poker inspector. It was gaining market share. Then it added the
bot functionality and it went out of business. Something to think
about!
- Tony
The industry needs to address this and address this fast. There are
two ways they can go, as far as I see it.
1. Ignore it, and let it happen -- to the damage of online poker.
2. Embrace it, make a common user interface and scripting language so
that everybody can write their own bots. Then, have tables and
tournaments explicitly set aside for bot play. Where bots and humans
can play, or just bots, and tables and tournaments set aside for
strictly human play. Everybody that signs up will get two accounts.
Their human account and their bot account.
Counteracting bots will be trivial on the human tables. With strong
encryption and random card design between deals and random placement
of cards and position.
If you give them a way to do it legally, and then make it difficult to
do illegally, there is little or no motivation and incentive to do it
illegally. Like on the human tables and tournaments.
"Poker Inspector" <ja...@pokerinspector.com> wrote in message news:<vnlu7ra...@corp.supernews.com>...
Rubbish, TTH is the best poker software out there but very beatable. IBM spent
hundreds of millions on a half-decent chess computer called Big Blue, it's
actually a mainframe costing hundreds of thousands if anyone wants to buy one, it
wouldn't win the world championship. IBM had the resources of hundreds of the
best scientist in the world, a single programmer sitting at home simply wouldn't
have the intelligence or knowledge to build a 100% winning BOT and if they did
they would keep it to themselves, you or I would never hear about it except on the
front cover of the Times.
Not at all rubbish. Maybe, it was hard for Big Blue to build Deep
Blue, but I personally would definitely reject chess game for money
even against a decent kid bot from Chessmaster 9000. Probably even
against some cheaper computer chess program. Though I am not a very
bad chess player, I don't need Deep Blue's intellect to be well
beaten.
Yes, probably it's very hard to build 100% winning bot. It can be much
easier though to build, say, 60% winning bot. And if you were, say,
100% losing player you could easily go even for 25% winning bot.
Maybe, those underdeveloped bots do not expose direct personal danger
to you, a superb player. But when they achive just a little above
average level, they can simply damage this online entertainment. Most
bad and mediocre low-limit players will probably leave. What will you
do then? Taking money from the bots? With all their mediocre gameplay,
they won't cap the pot with some kind of a garbage hand, though...
Anyway, it's inevitable. It doesn't matter who eventually makes it.
This time PokerInspector, probably, is just too prompt about asking
for some money share, or for the "offer they cannot reject" ((c)
Godfather). But that's a serious danger and it shouldn't be
overlooked.
Tough standard.
But I guess that would be necessary when it's up against 100% winning
players.
Oh, btw, could you direct me to where one can aquire the intelligence and
knowledge to become a 100% winning player?
"James smyth" <starl...@yahoo.co.uk> wrote in message
news:W8Leb.6446$2Y1.53...@news-text.cableinet.net...
>
10 years ago I could beat any chess program in the world, now I can
buy a program for $30 that not only beats me easily, but can choose to
play deliberately badly in a number of different styles to give me a
chance, and teach me about my mistakes at the same time. It beats
masters (but not grand masters) at speed chess.
As a recent poker convert one of the attractions of poker vs chess is
precisely the sort of problems it poses that computers find difficult
to solve. Nonetheless I am convinced it can't be that long before
application of AI techniques lead to computers being able to give even
top players a good game
Use the dealer text as cues for action and reporting.
Personally I honestly don't care if you make it or not. They're already out
there anyway. Just make sure that there's a back door that if I'm sat on the
same table as one of your bots, make it dump everything to me :)
Actually that does pose a good question - what's stopping an author from
doing something like this? The code looks to see who's sat down and if one
nickname matches a list of people to "dump" to, it just plays all hands
against that person hoping to loose?
Get enough copies of the bot out there in use and you could make a serious
amount of money from this.
There has been many millions of pounds invested in the race to build a computer
that could beat a man in chess. That same investment isn't taking place today to
build the equivalent for poker. Poker is also going to be a much harder challenge
because of the luck involved and lack of a consistent logic compared to chess.
If you know your playing against a BOT, you can easily adapt, raising with little
forcing the BOT to go down the defensive code path because it thinks it's up
against a good hand allowing you to maybe make a cheap flush. If after a number
of months/years someone manages to gather a profile on you and feeds it to the
BOT, you simply play more straightforward and the BOT will be incapable of knowing
you have varied your play to maximise your results against it. It took some of
the greatest scientist in the world to make today's chess computers, a few little
shareware authors aren't going to create anything worth worrying about.
I have to disagree. Comparing chess and poker is just wrong. There
seem to be a few people in the thread making the argument that it's
harder to program a computer to play poker. For starters, that's the
wrong way of looking at it.
They are fundamentally different games. For all intents, chess is
working with an infinitely large data space to which you have full
knowledge of what where you are. The trouble is knowing what to do
based on that knowledge. If you lose a game, it's because you were
outplayed. Poker, you're playing with a basically finite data set, and
you can only make a best guess as to where you are. If you knew where
you were(i.e you could see your oponents cards and knew the order of
the deck), your decision is much easier.
As much as you are playing against your opponents in poker, there is
the unknown of the cards involved. The relationship between winning
and making the best move in each game is different. You can play great
poker and still lose horribly to a bunch of weak players. That tends
not to happen in chess. They are fundamentally different. Even if they
weren't, Deep Blue was designed to go up against the world's best
players and win. I suspect low limit online poker players are not the
world's best. You just don't need the same level of sophistication. So
comparing IBM's effort at making a chess playing computer to a couple
shareware programmers really isn't valid. There's the issue of scaling
up the level of play but you can't do that while ignoring the nature
of the game.
Could you beat a brain dead poker program simply by changing your
stategy? Of course. But trying to force it "go down the defensive code
path", isn't going to work against an program with any kind of
sophistication, these kind of programs simply don't have
offensive/defensive paths. They are adaptive beasts designed to
understand what is going on, not just look you up in a database. You
could have full knowledge of the program works (i.e you have the
source code) and still not know for sure what it will do in a given
situation. You're not going to suddenly crush it by playing a little
more straightforward. More likely, it'll beat you even more, because
you're now playing in a form you're not used to and one that it can
understand far better than you.
That said, making such a program to play a specific optimum poker
strategy (what ever you want to consider optimum) is practically
trivial. If fact if you wanted, you could make the computer generate
it's own strategy that could well be better than your 'optimum' one.
All you'd need is to teach it the rules, and let it learn by playing
itself. While you could still do that in chess, it's not really going
to work due to the nature of the problem.
I don't know anything about the programmers proposing this, but it's
cetainly possible do. And doing it well enough to impact online poker
isn't even that hard. I have to say - as a poker player, chess player,
scientist, and programmer - i think the difficulty in making a poker
playing program doesn't even come close that of a chess bot. But the
poker bot isn't going to win every hand - it's impossible.
So, why hasn't someone done so? It'd be naive to assume that someone
has not. But thinking practically you have to consider the scientific
merit of a poker playing computer. It's just not as (scientifically)
spectacular to make a machine beat a human poker player. It's only
going to get less so as computational power increases. At some point
in the past, say perhaps 20 years ago, that would have been a huge
feat, but certainly not now. 20 years ago there simply wasn't the
technology for Deep Blue either.
The point (as always) is that it comes down to money. Because of it's
shady scientific validity, it's not a project that's going to get
monetary support from any organization with any kind of scientific
reputation. It's not a scientific problem, many scientists would scoff
at the idea that they waste their time doing it (though they might
still do it for fun). No one could tell you with a straight face that
they built this state-of-the-art poker machine (with convient
automated interface to online gaming), that can wipe the floor with
any player, "simply for the good of science". They did it for money.
And it doesn't make them money to broadcast that they have it . . .
unless they want to sell it.
"Checkraiser" <a2...@yahoo.com> wrote in message
news:41ead93.03080...@posting.google.com...
> Perhaps you will be marketing a bot in a year or two.
We are not in business of creating bots.
.
.
.
Robert,
www.pokerinspector.com
"Instant Odds for Online Poker"
NOW
"Poker Inspector" <ja...@pokerinspector.com> wrote in message news:<vnlu7ra...@corp.supernews.com>...
> Hello RGPers,
>
> We're planning to release v2.0 of Online Hold'em Inspector in early
> December, 2003. One of the new features we're thinking about adding is Auto
> Play option. We never wanted to add this feature but there's so much demand
> for it that we might actually do it.. So we're wondering what the poker
> community think about this possibility.
>
> The Auto Play feature will allow Inspector to play Hold'em based on the
> profile you've created. We'll provide all the user interface for you to
> create a profile; so the Inspector will play as good as you programmed it.
> It will be a very tedious process but some may find it worthwhile. We'll
> include one sample profile -- average player in a typical full game. We'll
> also include a simulator so you can test your profiles.
>
> How do you think such program will affect the online game? Last thing we
> want to do is kill the online poker. Please tell us why we should or not
> include Auto Play option. At this time we're about 75% for including the
> feature and 25% not.
>
> Thanks in advance for your thoughts,
> Robert
> www.pokerinspector.com
> "Instant Odds for Online Poker"
COMMENT
First off all, who is demanding this feature to be included? Your
users?
You guys are just plain amazing. 2 months ago you wrote in the
newsgroup that you are not in business of creating a bot, and now you
are thinking of marketing it?
How should anyone trust you, anyway?
I don't really understand, if you are so bright, why don't you just
profit with botmaking? Not enough money? What I think is, you aren't
even half decent programmers as Andy already pointed out. You are
unable to do a winning bot so you will sell some crap do idiotic
customers. If you think creating bot is easy - it is not. Instead you
will be offering some profiles. You really think fish will make their
profiles? Profiles will win by theirselves, regardless of opponents?
Why don't you rather sell them some winning profiles? Oh yes, it was
an rhetorical question.
BTW: If anybody thinks that beating TTH is easy: I would just like to
add that profiles are meant to fight between themselves (and in that
they are extremely useful). Indeed, even in Bret Maverick there are so
many holes, that their "bot" is pretty crappy, even if it beats other
profiles. Both preflop and postflop are jokes, good players think MUCH
MUCH MUCH more in depth than Bret Maverick does. His tight/agressive
mechanical strategy just isn't suitable to real situations vs. real
players.
One thing more..it is not a great thing to beat a profile, whose
strategy you can look at? One should be really dumb to lose against
it.
And people, please stop joking about problems of how to read screen
and such bull shit. That's the easiest part of all. The hard part is
the poker logic
behind it.
BTW: I would really like to see that you start commercializing your
software. In that way, Party will be forced to change their software
throughly and then shut down your and all other private bots that are
running there. They could client software every day and then what? You
will return the money?
I bet you will.
regards, CR
Are you sure?
Game theoretic modelling of a 10 player game with incomplete
information sounds a lot harder than tree searching.
Recognising patterns of play in your opponents and adapting your play
so that you profit from their mistakes also sounds difficult for a
computer.
Coping with luck, on the other hand, is trivial. Computers can
calculate probabilities better than people, and don't go on tilt when
they are unlucky (at least not until they get really clever and start
thinking for themselves and having their own bank accounts, and that's
a few years away yet)
PS - since my last post I have been attacked by lots of viruses. Does
this always happen or is it some sinister poker bots trying to keep
the lid on this debate?
Bots don't go on tilt, bots don't lose discipline because they're
getting bored, bots don't make silly human errors. They may make other
kind of errors, but this isn't poker. At least not the kind I want to
play.
--
2 hi
------------------------------------------------------------------------
Access RGP from Pokeritis.com, create a poker journal, and more.
http://www.pokeritis.com/forums
View this thread @ Pokeritis.com forums: http://www.pokeritis.com/forums/showthread.php?threadid=22336
>Bots don't go on tilt, bots don't lose discipline because they're
>getting bored, bots don't make silly human errors. They may make other
>kind of errors, but this isn't poker. At least not the kind I want to
>play.
Plus, I would think they could learn from experience. Much the same
way Ask.com learns with every new question posed to it.
It all comes down to luck. In fact, I used POKERINSPECTOR to look at
the cards, if the traffic light was RED/GREEN/YELLOW -- it would make
a corresponding mouse click on the screen.
It goes up and down... mostly down in cash... but I've been thinking
of taking a Turbo Texas Player and making it a bot... It's not that
hard.
-elektr0n
to...@thsoftware.com (Tony H.) wrote in message news:<d4bcb28.03100...@posting.google.com>...