Extreme Gammon 2

Post by GRondo
How strong is this sofware ?

Strongest on Market

Post by GRondo
Has it bugs ?

Yes, like any software nothing is perfect.

Post by GRondo
What about features ?

XG = It is fast, and the strongest bot to date, with a pretty nice User
Interface. However, the interface is often considered to be "busy" and
overwhelming to some newer users (Same complaint as GNUBG). The 3D mode
looks very nice, and the Game Transcription (recording live games)
feature is one of the easier to use ones among all the bots.

Post by GRondo
Is it worth the money ?

Yes

Post by GRondo
Would you prefer Snowie or GnuBB ?

Snowie isn't used much anymore (and is costly).

GNUBG is still free, and runs on a variety of platforms. GNUBG is not
quite as strong as XG but is still very strong.

XG run on Windows only.

There is another alternative BGBLITZ. It runs across many platforms and
there are versions available for many cell phones (BgBlitz2go).

I actively use GNUBG, BgBlitz, and XG2.

Michael Petch

2012-04-09 22:55:53 UTC

One other thing - Extreme Gammon and BgBlitz both have trial periods so
you can try before you buy.

muratk

2012-04-10 07:45:23 UTC

Post by Michael Petch
One other thing - Extreme Gammon and BgBlitz both have trial periods so
you can try before you buy.

Bgblitz doesn't have an unrestricted trial period.

It says one in six games will be randomly aborted but to me it looked
much more often than one in six.

And if you try to play matches longer than 1-3 points, you probably
will never get to finish a match during the trial period.

Unusual and very irritating approach to deter continuous free use of
the software but probably also a very effective way to piss off and
turn off prospective paying customers...

With other options available, including the free gnubg, why would
anybody bother with this bgblitz shit...??

MK

Stick

2012-04-10 00:58:12 UTC

Definitely worth it. You can see my own random quip about XG
scrolling across the top: http://www.extremegammon.com/default.aspx

This was a comment I made on my forums, http://www.bgonline.org/forums/,
not requested by Xavier, the creator of XG.

Stick

muratk

2012-04-10 07:59:50 UTC

Post by Stick
Definitely worth it.

Backgammon bots are a textbook case of circular fallacy and peer
pressure.

Ignore these sick gambling assholes. Extreme gammon could entertain
even a
ten year old child with its dice rolls. Okay, well, make it twelve
years
old... ;)

Snowie and gnubg may be slightly better but never any developer of any
bg
bot was able to stand up for their products and bet against people who
claimed they could predicts the built in dice rolls of those bots...

The technology to conduct an experiment over the internet is here, and
would
be relatively very simple compared to what those slimy assholes try in
order
to guild the arbitrary bullshit that they dream up to make money off
of others...

But, as of yet, after many years, they can't defend their concocted
piece of
shit products even against odds that supposedly could never cause them
to lose... :))

MK

Tim Chow

2012-04-10 19:08:23 UTC

Post by muratk
But, as of yet, after many years, they can't defend their
concocted piece of shit products even against odds that
supposedly could never cause them to lose... :))

Sure they can, Fraidy-Rat. You're the only one who's

1. afraid to wager a decent amount of money;

2. afraid to state in advance exactly what kinds of predictions you
would make and how many you would make in a given period of time; and

3. afraid to accept odds that, according to you, are overwhelmingly in
your favor.

---
Tim Chow

muratk

2012-04-11 09:52:22 UTC

Post by muratk
But, as of yet, after many years, they can't defend their
concocted piece of shit products even against odds that
supposedly could never cause them to lose... :))

Sure they can, Fraidy-Rat. You're the only one who's
1. afraid to wager a decent amount of money;

Did any of you cocksuckers determine what would be a "decent
amount of money" and propose to me...?

And why would the amount would matter, since in the long run
I couldn't predict any more and any less than 50%...?

Whether I would bet a dollar or a million dollars, in the end
we would all break even, no?

You are such a teasing cocksucking faggot... ;)

2. afraid to state in advance exactly what kinds of predictions
you would make and how many you would make in a given period of
time; and

I have done so many many times, as recently as a few days ago with
a sample position I posted in this group. If you don't follow what
is being posted, it's not my problem.

3. afraid to accept odds that, according to you, are overwhelmingly
in your favor.

To the contrary. You are the con artists who claim that your bots are
much stronger than players like myself, who are statistically nobody.

I tried to capitalize on that in trying to get you arrogant assholes
to spot me some points against the "extraterrestrial bots"...

You couldn't stand behind all that of your rating bullshit and refused
to do so.

Then, I agreed to bet on even odds.

Now, you motherfucking slimy asshole faggots are asking me to spot you
points...??

Haaa haaaa haaa....

Keep coming back and I will keep pissing on you...!! ;)

MK

Tim Chow

2012-04-13 00:16:12 UTC

Post by muratk
Did any of you cocksuckers determine what would be a "decent
amount of money" and propose to me...?

Of course I did. $1 million. Should I add poor memory to cowardice
in your list of character faults?

http://groups.google.com/group/rec.games.backgammon/msg/11ca95a86135f4c7?dmode=source

Post by muratk
And why would the amount would matter, since in the long run
I couldn't predict any more and any less than 50%...?
Whether I would bet a dollar or a million dollars, in the end
we would all break even, no?

Why would I, or anyone, waste my time playing a game with you where I
expect to break even?

Anyone with half a brain (I know I'm excluding you) recognizes that if
they want to prove that they can predict something with better than
even odds, then they should offer odds that are *in their favor
according to their claims* but *in the other person's favor according
to the other person's claims*.

Post by Tim Chow
2. afraid to state in advance exactly what kinds of predictions
you would make and how many you would make in a given period of
time; and

I have done so many many times, as recently as a few days ago with
a sample position I posted in this group. If you don't follow what
is being posted, it's not my problem.

No you haven't. You did not state how many such predictions you would
make in a given period of time (say, three hours of play).

Post by Tim Chow
3. afraid to accept odds that, according to you, are overwhelmingly
in your favor.

To the contrary.

I have offered odds that, according to you, are overwhelmingly in your
favor, but you have rejected them, insisting only on "even odds."

Let me spell it out for your feeble intellect's benefit. Suppose you
claimed to be able to predict coin flips with 90% accuracy. I offer
to give you $1 for each correct prediction and receive $2 for every
wrong prediction you make. According to you, these odds are
overwhelmingly in your favor. You stand to earn 70 cents per flip.
So why would you refuse such an offer and insist only on "even odds"?
Either because you're an idiot or a coward. Take your pick.

---
Tim Chow

muratk

2012-04-13 11:39:59 UTC

Post by muratk
Did any of you cocksuckers determine what would be a "decent
amount of money" and propose to me...?

Of course I did. $1 million. Should I add poor memory to
cowardice in your list of character faults?

And what bank account was that $1 million escrowed at...?

As soon as you can offer evidence you $1 million deposited
for this purpose, I will do my part...

Post by muratk
Whether I would bet a dollar or a million dollars, in the
end we would all break even, no?

Why would I, or anyone, waste my time playing a game with
you where I expect to break even?

Maybe the cocksucking scum who develop and market cheating
bots...?

Maybe even cocoksuckers like you who have a $1 million to
blow away...?

One thins that you won't have to do is waste your time. In
fact, I will be the one who risks to waste more time than
any other of you slimy assholes.

You just set the seed and start the server. Then you can sit
back and suck on your cigar (lit with a $100 bill :) while
I struggle and waste my time actually doing the playing
against the bot (of your choice).

Of course, you don't have to watch, unless it turns you on.
Your secretary can report you the results... ;)

Post by muratk
I have done so many many times, as recently as a few
days ago with a sample position I posted in this group.
If you don't follow what is being posted, it's not my problem.

No you haven't.

Yes indeed, I did and it's still there for you to read.

You did not state how many such predictions you would
make in a given period of time (say, three hours of play).

Why does it matter to you?

And what if I said I will predict "as often as your cheating
bot cheats"? Would that be fair enough?

If you want a more precise answer, you tell me how often your
rigged bot will cheat and that will be the answer to your own
question... :))

Let me spell it out for your feeble intellect's benefit.
Suppose you claimed to be able to predict coin flips with
90% accuracy.

Blah blah... Irrelevant bullshit as usual... :(

So why would you refuse such an offer and insist only on
"even odds"? Either because you're an idiot or a coward.
Take your pick.

Obviously we have a language/communication problem here but
it all can be solved.

What I mean is that we will bet even money on the actual odds.

If you had read my article about how the bots cheat, you would
have understood this already.

If I bet on a one in 36 possibility, you cocksuckers lay down
35 units of betting (such as $35 millions of dollars) against
my betting only one unit (such as $1 million dollars)...

I am pretty sure you are of a capacity to understand this but
as sick of a faggot as you are, you will keep coming back for
me to piss on you again and again... :)) Enjoy then... ;)

MK

Michael Petch

2012-04-14 02:30:00 UTC

Post by muratk
And what bank account was that $1 million escrowed at...?
As soon as you can offer evidence you $1 million deposited
for this purpose, I will do my part...

What do you have to lose even if he had 10k in his account? If you can
predict the dice your going to come out a head pretty quick. If you can
predict the dice you should be jumping on this really fast because you
have almost nothing to lose and a whole lot to gain! No need for Escrow
on this.

muratk

2012-04-15 08:50:01 UTC

On Apr 13, 8:30 pm, Michael Petch <***@capp-sysware.com> wrote:
.
.
.
.

Post by muratk
And what bank account was that $1 million escrowed at...?
As soon as you can offer evidence you $1 million deposited
for this purpose, I will do my part...

.
.

Post by Michael Petch
What do you have to lose even if he had 10k in his account?
If you can predict the dice your going to come out a head
pretty quick. If you can predict the dice you should be
jumping on this really fast because you have almost nothing
to lose and a whole lot to gain! No need for Escrow on this.

.
.
Right, I am ready to jump on this offer without requiring that
cocksucking Chow proves he has a million dollar escrowed, if
you guys don't require the same from me.

Let's get started already, then...!!

MK

Tim Chow

2012-04-14 21:10:16 UTC

Post by muratk
And what bank account was that $1 million escrowed at...?

Clearly you don't have much experience handling this kind of money.

I don't have the $1 million escrowed yet because we weren't able to
reach even a verbal agreement that we were both happy with. I'm not
going to go to the trouble of setting up an escrow account and
liquidating enough assets until there's some promise of a tangible
payoff. There's a cost to moving this kind of money around, you
know. (Or I guess you don't know.)

If you don't believe I have access to such assets, then we can
approach the James Randi Educational Foundation. They have $1 million
already escrowed. You're not explicitly claiming paranormal ability,
but it's close enough that I think the JREF could be persuaded to
conduct a test. However, you're still going to have to reach an
agreement about the exact conditions of the test. If you can't agree
to the kinds of conditions I've proposed, I don't think you'll have
much luck with JREF either.

Post by muratk
One thins that you won't have to do is waste your time. In
fact, I will be the one who risks to waste more time than
any other of you slimy assholes.

That's not true. The setup you propose is insecure. I'm 90%
convinced now that you genuinely believe you have the ability that you
claim you have, but there's a 10% chance in my mind that you're a con
artist. I have enough experience to see how a clever con artist could
exploit the protocol you suggest to make the kinds of predictions
you're talking about. In order not to be swindled, I have to set up
rigorous controls and monitor the test. This requires considerable
time and effort.

Post by muratk
I have done so many many times, as recently as a few
days ago with a sample position I posted in this group.
If you don't follow what is being posted, it's not my problem.

No you haven't.

Yes indeed, I did and it's still there for you to read.

No. I read your post. All you did was to propose one specific type
of prediction that you would make. You didn't say how many such
predictions you would make in a given period of time.

Post by Tim Chow
You did not state how many such predictions you would
make in a given period of time (say, three hours of play).

Why does it matter to you?
And what if I said I will predict "as often as your cheating
bot cheats"? Would that be fair enough?

No. The reason it matters is that I'm not the president of the Fraidy-
Rat Charity Foundation. Why should I go to the trouble of running a
test where I stand to profit $0 and the only beneficiary is yourself?
There has to be something in it for me. I am not a bot programmer. I
have no financial interest in eXtreme Gammon. I don't know if eXtreme
Gammon cheats. I don't think it does, but I have no insight into the
workings of the code. I can't possibly know "how often the cheating
bot cheats." You're the one who has that insight.

If I can't estimate how much money I will make per hour for running
this test, I run the risk of your sitting there and complaining that
the bot has been rigged so that it no longer cheats. Three hours
later, you've made no predictions and I've wasted my time setting up
the escrow account and the test. You've wasted your time too. Nobody
wins. What's the point?

Post by muratk
If I bet on a one in 36 possibility, you cocksuckers lay down
35 units of betting (such as $35 millions of dollars) against
my betting only one unit (such as $1 million dollars)...
I am pretty sure you are of a capacity to understand this but
as sick of a faggot as you are, you will keep coming back for
me to piss on you again and again... :)) Enjoy then... ;)

I understand what you're proposing, but what *you* don't understand is
that you're offering no incentive for someone to go to the trouble of
running a test. If you want someone to go to this trouble then you
have to offer them what looks like a good deal.

It should not be a problem if you're actually able to make the
predictions you claim to be able to do, because you should be able to
offer a deal to me that looks favorable to me, but that is actually
favorable to *you* if you can do what you claim to be able to do.

---
Tim Chow

muratk

2012-04-15 09:12:38 UTC

On Apr 14, 3:10 pm, Tim Chow <***@yahoo.com> wrote:
.
.
.
.

Post by muratk
And what bank account was that $1 million escrowed at...?

.
.

Post by Tim Chow
Clearly you don't have much experience handling this kind of money.

.
.
You are right. I don't. And I am impressed that you could detect that.

So, then, I assume that you have such experience? Would you care to
tell us whan was the last time you handled a million dollars, for
any purpose, if not specifically for gambling purposes...?
.
.

Post by Tim Chow
I don't have the $1 million escrowed yet because we weren't able
to reach even a verbal agreement that we were both happy with.

Having proof of the funds if part of the process. If you don't have
a million dollars available for this bet, shut up and fuck off...

Or, just just bet what you actually have in your pocket...
.
.

Post by muratk
One thins that you won't have to do is waste your time. In
fact, I will be the one who risks to waste more time than
any other of you slimy assholes.

.
.

Post by Tim Chow
That's not true. The setup you propose is insecure. I'm 90%
convinced now that you genuinely believe you have the ability
that you claim you have, but there's a 10% chance in my mind
that you're a con artist. I have enough experience to see how
a clever con artist could exploit the protocol you suggest to
make the kinds of predictions you're talking about.

If you being the only one to know the seed for the dice roller
is not secure enough, what would you feel more secure...? Tell
us so that I can try to accommodate you mother mucking slime...

And please do tell us about your "experience" so that I or others
can learn from your experience and avoid being taken by con artists
like me... Go ahead, we are all listening cocksuking math faggot...

Post by Tim Chow
In order not to be swindled, I have to set up rigorous controls
and monitor the test.

What "rigorous controls"? What monitoring? Please, do us all a
public service and expand on those, could you...?

Once you do so and prove how much time and effort those would take,
I would be willing to make adjustments to the bet in order to be
fair...

Post by muratk
Yes indeed, I did and it's still there for you to read.

No. I read your post. All you did was to propose one specific type
of prediction that you would make.

What? You want me to give you 100 more examples...? Hah hah haaaa...

I didn't even have to give you any examples faggot. I want to nail
your ass for money!

Alternatively, I could publish 100 such examples in a book for
$29.95 :)

Post by muratk
And what if I said I will predict "as often as your cheating
bot cheats"? Would that be fair enough?

My challenge was for the bot developers and/or their abetters/ass
kissers.

If you don't have any charitable or financial interest in extereme
gammon,
just fuck off then... Let the puto himslelf speak...

Post by Tim Chow
I understand what you're proposing, but what *you* don't understand is
that you're offering no incentive for someone to go to the trouble of
running a test. If you want someone to go to this trouble then you
have to offer them what looks like a good deal.

That "someone" is not necessarily *only you* cocksucking asshole.
Think
of the others like the puto himself and show them the respect that
they
may find my offer worthwile...

Why do you think you have to be the one to defend all the slimes who
are
developing, peddling, promoting rigged garbage bg bots...

Post by Tim Chow
It should not be a problem if you're actually able to make the
predictions you claim to be able to do, because you should be able to
offer a deal to me that looks favorable to me, but that is actually
favorable to *you* if you can do what you claim to be able to do.

If the deal is not appealing to you, then don't take it, shut up and
fuck off mother fucking faggot...

Let the puto speak for himself... ;)

MK

Tim Chow

2012-04-20 00:52:33 UTC

Post by muratk
And please do tell us about your "experience" so that I or others
can learn from your experience and avoid being taken by con artists
like me... Go ahead, we are all listening cocksuking math faggot...

Post by Tim Chow
In order not to be swindled, I have to set up rigorous controls
and monitor the test.

What "rigorous controls"? What monitoring? Please, do us all a
public service and expand on those, could you...?

Your memory is poor. Back in 2009, I explained several different ways
that some apparently secure protocols could be defeated. In one case
you even admitted that the method I described had not occurred to
you. If you are really interested, you can search the archives of
this group for more details.
---
Tim Chow

Michael Petch

2012-04-20 01:50:58 UTC

Post by Tim Chow
Back in 2009, I explained several different ways
that some apparently secure protocols could be defeated

I believe that was around the same time you and I discussed
"Cryptographic Protocols"

muratk

2012-04-20 10:04:03 UTC

On Apr 19, 7:50 pm, Michael Petch <***@capp-sysware.com> wrote:
.
.

Back in 2009, I explained several different ways
that some apparently secure protocols could be defeated

I believe that was around the same time you and I discussed
"Cryptographic Protocols"

Even it was around the same time you faggots discussed a different
subject (i.e. "Cryptographic Protocols"), what did that have anything
to do with what we were discussing...??

Fine, Petch, you wanna go ahead and suck on faggot chow, knock
yourself off... We will just know that you are another one of them
motherfucking sick faggots... ;)

MK

muratk

2012-04-20 10:00:42 UTC

On Apr 19, 6:52 pm, Tim Chow <***@yahoo.com> wrote:
.
.

Post by Tim Chow
In order not to be swindled, I have to set up rigorous controls
and monitor the test.

What "rigorous controls"? What monitoring? Please, do us all a
public service and expand on those, could you...?

Your memory is poor. Back in 2009, I explained several different
ways that some apparently secure protocols could be defeated. In
one case you even admitted that the method I described had not
occurred to you. If you are really interested, you can search the
archives of this group for more details.

I believe that was about the dice rolls while playing on backgammon
servers, not about me playing against gnugb under specific conditions.

For me, you are the slimiest cocsucking math phd in this group... :0)

In this case, I will be playing against gnubg one-on-one, using the
predetermined dice rolls that *you cocksucker* will have the choice
about and the control over...

Very simple to understand but not for sick slimes like you who try to
acquire some math phd credibility through the backing of world-class
cocksucking sick gamblers here... ;)

Gnugb has several PRN's and I gave you cocksuckers your choice of PRN
and your choice of starting seed value. It would be virtually
impossible to detect during even a 25-point long match, what PRN and
what seed you selected!

Surely not in a matter of seconds, that will take me to make my moves,
anyway...!!

If you want, we can add a twist to the bet by making the total time
used by each player during each match to a factor in determining the
pay offs...

But, you are not anybody who could possibly understand such simple
notions and thus not to be considered beyond your pathetic
entertainment value...

Why do you even exist mother fucking faggot...??

MK

muratk

2012-04-10 07:37:18 UTC

Post by GRondo
How strong is this sofware ?

They claim it's strong but if you play against it using its built in
dice rollers, it's more amusing than bozo the clown... :))

Your experience may vary if you have the patience to play with manual
dice...

Post by GRondo
Has it bugs ?

Way too many bugs for a commercial product. My guess is that it's
built on a stolen engine and the developer can't even wrap it up
in a functional package...

Post by GRondo
What about features ?

Tons of meaningless guilding the bullshit... :))

Post by GRondo
Is it worth the money ?

Not worth my money.

Post by GRondo
Would you prefer Snowie or GnuBB ?

Snowie seems slightly more realistic to me but it's outdated and
irrationally expensive. One must be a totally desperate sick
gambler to expect anything from it to justify it's cost.

Personally, I think gnubg is the most flexible, customizable and
realistic of the three.

I like its interface and features.

I think it's shy on hitting compared to XG and I don't know how
much this plays towards the claims that XG is stronger.

However, I don't think XG would necessarily do better against
humans based on this or that. Strong human players would quickly
figure the bots out and adjust their games accordingly.

I don't know if this helps you any but regardless, don't let these
sick assholes get to you. They are as slimy as they come...

MK

GRondo

2012-04-10 19:02:23 UTC

Post by GRondo
How strong is this sofware ?

They claim it's strong but if you play against it using its built in
dice rollers, it's more amusing than bozo the clown... :))
Your experience may vary if you have the patience to play with manual
dice...

Thank you Murat.
Though I am a beginner, I often beat it by 16 to 0 in a 5 points match at
default level. It always proposes and accepts double and take.
Extremeroller++ is too slow on my Core2Duo. I wonder how to increase its
strength.

Tim Chow

2012-04-10 19:16:41 UTC

Post by GRondo
Though I am a beginner, I often beat it by 16 to 0 in a 5 points match at
default level. It always proposes and accepts double and take.
Extremeroller++ is too slow on my Core2Duo. I wonder how to increase its
strength.

No matter how strong an opponent you play, human or computer, you're
going to win a sizable fraction of the time. That's the nature of
backgammon. Even if you had a computer that ran XGR++ quickly, it
would not be noticeably stronger than XGR+. You'd still be able to
win 16 to 0 in a five-point match sometimes. The same would be true
if you were to play the world's strongest human players.

---
Tim Chow

Bradley K. Sherman

2012-04-10 21:19:18 UTC

If "I often beat it by 16 to 0 in a 5 points match" means that
GRondo has beat one of the top BG programs in 16 straight
five-point matches more than once, then either the software
is malconfigured on GRondo's computer, or GRondo can't count.

Or both.

--bks

GRondo

2012-04-10 22:45:18 UTC

Post by Bradley K. Sherman

Post by GRondo
Though I am a beginner, I often beat it by 16 to 0 in a 5 points
match at default level. It always proposes and accepts double and
take. Extremeroller++ is too slow on my Core2Duo. I wonder how to
increase its strength.

I confirm :
Opponent : XG Roller
I ask for a five points match.
In *every* game XG Roller doubles, I take then I double, it takes ...
until cube is 8 !
The cube is 8.
I often make a gammon.
So 8X2 = 16

I built a bear off database for 11

Bradley K. Sherman

2012-04-10 22:47:48 UTC

Post by GRondo

...

Opponent : XG Roller
I ask for a five points match.
In *every* game XG Roller doubles, I take then I double, it takes ...
until cube is 8 !
The cube is 8.
I often make a gammon.
So 8X2 = 16

Thanks for the clarification. I don't have XG Roller,
so can't comment on the bug.

--bks

muratk

2012-04-11 09:55:47 UTC

Post by GRondo
Opponent : XG Roller
I ask for a five points match.
In *every* game XG Roller doubles, I take then I double, it takes ...
until cube is 8 !
The cube is 8.
I often make a gammon.
So 8X2 = 16

Thanks for the clarification. I don't have XG Roller,
so can't comment on the bug.

Thanking him for the "clarification" is not enough!

You insulted him by saying "either the software is malconfigured
on GRondo's computer, or GRondo can't count".

Apoliogize to him, motherfucking arrogant asshole...!!!

MK

x***@gmail.com

2012-04-11 19:18:09 UTC

Gronto, thank for trying out our program.

If in every single game XG doubles, I suspect you may have inadvertently downloaded XG from another site than ours (There are many site allowing download of supposedly cracked version XG, all of these will display that kind of behavior.)
I suggest you uninstall your current version and download it from our website : www.extremegammon.com

Xavier Dufaure de Citres
eXtreme Gammon

Paul

2012-04-11 21:57:45 UTC

On Apr 11, 8:18 pm, ***@gmail.com wrote:
....

Post by x***@gmail.com
If in every single game XG doubles, I suspect you may have inadvertently downloaded XG from another site than ours (There are many site allowing download of supposedly cracked version XG, all of these will display that kind of behavior.)

...

Playing against a beginner (Grondo's self-description), it could well
be that the probability of XG doubling is > 90%.

Grondo's argument is still totally unconvincing. If Grondo's arguing
that XG's cube action is weak, Grondo needs to post a position and
argue against XG's cube action in that specific situation.

Paul Epstein

Michael Petch

2012-04-12 00:17:14 UTC

Post by Paul
Playing against a beginner (Grondo's self-description), it could well
be that the probability of XG doubling is > 90%.
Grondo's argument is still totally unconvincing. If Grondo's arguing
that XG's cube action is weak, Grondo needs to post a position and
argue against XG's cube action in that specific situation.

I think there is more to it than that in this situation. The cracked
versions of XG will actually not play normally, in fact it'll play
pretty badly no matter what setting you use.

Did the player have a properly licensed copy of XG, or had he tried it
free for the trial period (downloaded from www.extremegammon.com )? I
would be curious to know if a trial or properly licensed version
exhibits this behavior. If the case is that it was legite, then there is
something amiss.

Michael Petch

2012-04-12 00:21:57 UTC

Post by Michael Petch
I think there is more to it than that in this situation. The cracked
versions of XG will actually not play normally, in fact it'll play
pretty badly no matter what setting you use.

If Grondo would like to send me a copy of one (or all) of the matches
I'd love to take a look.

Matches (XG or Mat files) can be emailed to me at ***@capp-sysware.com .

muratk

2012-04-13 11:16:05 UTC

The cracked versions of XG will actually not play normally, in
fact it'll play pretty badly no matter what setting you use.

How do you know this? Did you ever play with a cracked version
of XG?

Did the player have a properly licensed copy of XG, or had he
tried it free for the trial period (downloaded
fromwww.extremegammon.com)? I would be curious to know if a
trial or properly licensed version exhibits this behavior.

Personally I couldn't care less about what version of XG grondo
played with as much as whether you cocksucker ever played with
a cracked version of XG...?

Can you answer this? And an you tell me where did you downloaded
it so that I can verify "you" cocksucker's allegations...??

MK

Michael Petch

2012-04-13 14:41:18 UTC

Post by muratk
How do you know this? Did you ever play with a cracked version
of XG?

Yes

muratk

2012-04-15 08:46:04 UTC

On Apr 13, 8:41 am, Michael Petch <***@capp-sysware.com> wrote:
.
.
.
.

Post by muratk
How do you know this? Did you ever play with a cracked version
of XG?

.
.

Yes

.
.
Thanks for caring to answer the first part of my question but why
did you ignore the second part...?

Can you tell me where to download that cracked version of xg so
that I can duplicate your experience and believe you...?

Oh yeah, I have seen the pink aliens who landed in my backyard
in a stainless steel saucer also... :)

MK

Michael Petch

2012-04-15 09:32:03 UTC

Post by muratk
Can you tell me where to download that cracked version of xg so
that I can duplicate your experience and believe you...?

Google.

GRondo

2012-04-15 18:08:36 UTC

Post by muratk
Can you tell me where to download that cracked version of xg so
that I can duplicate your experience and believe you...?

Google.

Muratk
http://uploaded.to/file/9aivj1ii for instance. There are a lot of others
links. It seems this BG soft has been badly cr****d...

Michael Petch

2012-04-15 18:34:00 UTC

Post by GRondo

Post by muratk
Can you tell me where to download that cracked version of xg so
that I can duplicate your experience and believe you...?

Google.

Muratk
http://uploaded.to/file/9aivj1ii for instance. There are a lot of others
links. It seems this BG soft has been badly cr****d...

In all the versions I have seen online they have all managed to remove
the licensing info but were too dumb to actually realize the bot wasn't
playing normally.

As well some of these cracked products introduce code snippets to add a
backdoor to your system so that someone can eventually install malware
on your system remotely.

muratk

2012-04-15 20:02:39 UTC

On Apr 15, 12:34 pm, Michael Petch <***@capp-sysware.com> wrote:
.
.
.
.
.
.
.
.

Post by muratk
Can you tell me where to download that cracked version of xg so
that I can duplicate your experience and believe you...?

.
.

Post by Michael Petch
Google.

.
.

http://uploaded.to/file/9aivj1iifor instance. There are a lot of others
links. It seems this BG soft has been badly cr****d...

.
.

Post by Michael Petch
In all the versions I have seen online they have all managed to remove
the licensing info but were too dumb to actually realize the bot wasn't
playing normally.

.
.
Do you mean the author of the software put some code in it to detect
if it is cracked and make it play badly if it is?

I would find this interesting. Actually I am totally ignorant about
such matters. So, maybe this is a general defense against cracking?

Post by Michael Petch
As well some of these cracked products introduce code snippets to add a
backdoor to your system so that someone can eventually install malware
on your system remotely.

.
.
Yes, for that reason alone I have never used cracked version of any
software. I assumed that most people like you here would know better
not to do so also. And I was surprised when you answered that you had
actually played with a cracked version of XG. But maybe you used an
isolated test computer?

Anyway, my reason for asking was to find out whether anyone of you
guys
were able to duplicate and determine what grondo reported was due to
using cracked XG. Do we know yet?

MK

Michael Petch

2012-04-15 20:16:09 UTC

Post by muratk
Do you mean the author of the software put some code in it to detect
if it is cracked and make it play badly if it is?

Very astute.

muratk

2012-04-16 10:28:22 UTC

On Apr 15, 2:16 pm, Michael Petch <***@capp-sysware.com> wrote:
.
.

Post by muratk
Do you mean the author of the software put some code in it to detect
if it is cracked and make it play badly if it is?

Post by Michael Petch
Very astute.

Are we supposed to redefine "astute" for the sake of some
cocksucker...?

I have always been able to uninstall/reinstall software like Snowie,
eXtreme Garbage, etc. in a matter of seconds and keep using them in
trial mode indefinitely. I also posted articles here about how it can
be done so that nobody would need to resort to using cracked versions
of such commercial garbage built on technology "stolen" from other
products...

How about adding an "s" as a compromise and call it "ass-tute"...??

MK

mpetch

2012-04-16 00:03:33 UTC

Yes, for that reason alone I have never used cr*cked version of any
software. I assumed that most people like you here would know better
not to do so also. And I was surprised when you answered that you had
actually played with a cr*cked version of XG. But maybe you used an
isolated test computer?

Xavier knows I am a licensed user of XG. I do a fair amount of XG
support among players on SafeHarborGames. I may develop on GNUBG, but I
support Backgammon in general. I had an inquiry about a cr*cked version
from a user a couple months ago that was causing them issues so I
happened to download a copy. The person now owns a licensed copy!

Yes, when I install cr*cked software I assume that it is likely
malicious in nature. I use a sandboxed system on a separate segment of
my internet connection as to not cause any issues with my internal network.

Anyway, my reason for asking was to find out whether anyone of you
guys
were able to duplicate and determine what grondo reported was due to
using cr*cked XG. Do we know yet?

I requested a copy of some of the XG (or MAT files) from Grondo for this
purpose. I haven't heard anything. It will be obvious once I get one of
the files as to whether the issue was the bot deliberately playing bad.
One just has to analyse how the bot played by using a properly licensed
copy.

muratk

2012-04-16 10:21:04 UTC

On Apr 15, 6:03 pm, mpetch <***@gmail.com> wrote:
.

Post by mpetch
I requested a copy of some of the XG (or MAT files) from
Grondo for this purpose. I haven't heard anything. It will
be obvious once I get one of the files as to whether the
issue was the bot deliberately playing bad.

Why go to long the long way to determine this, instead of asking
Xavier...?

If grondo emails you the mat files, are you sure that you can
determine whether he played with a legitimate or cracked version of
XG...?

Apparently, you have some rapport with the developer of XG. So, why
don't you just simply ask cocksucking Xavier if what reported by
grondo is indeed what the "astute" code he implemented in his own
product, when his product is a detected cracked version...?

Much simpler, non...? ;) (Come dise en espanol?)

MK

muratk

2012-04-13 11:11:46 UTC

Grondo's argument is still totally unconvincing. If Grondo's
arguing that XG's cube action is weak, Grondo needs to post
a position and argue against XG's cube action in that specific
situation.

I agree that we should not take grondo's word for it without
him providing some evidence.

However, he may decide to go about it by betting money against
you/us all that he can substanciate his claims...

What would you say then...?

MK

muratk

2012-04-13 11:03:26 UTC

Post by x***@gmail.com
If in every single game XG doubles, I suspect you may have
inadvertently downloaded XG from another site than ours

I would like to "advertently" download a cracked XG

Post by x***@gmail.com
(There are many site allowing download of supposedly cracked
version XG, all of these will display that kind of behavior.)

Would you be kind enough to give me a "few" of those "many"
sites...?

And if I download a cracked copy of XG from the site(s) you
will specify, and if it doesn't display "that kind of behavior",
can I shove it up your lying ass...? :)

MK

Paul

2012-04-11 08:02:47 UTC

Post by Bradley K. Sherman
If "I often beat it by 16 to 0 in a 5 points match" means that
GRondo has beat one of the top BG programs in 16 straight
five-point matches more than once, then either the software
is malconfigured on GRondo's computer, or GRondo can't count.

...

Bradley,

That's a blatant misreading. He said "in a 5 points match". He
didn't say "16 straight five-point matches."
This clearly means that when GRondo plays XG in a 5 point match, the
result is often 16-0 to GRondo.

**********************************************************************************
Nothing GRondo has said is the slightest evidence of any type of
"bug".
**********************************************************************************

If GRondo plays poorly, then XG will usually have a correct initial
double. If GRondo continues to play poorly, GRondo may well decide to
always redouble, no matter what the checkers situation is.
If GRondo's abysmal play continues, XG will double again to 8 whenever
doubling to 8 is the correct play.
Then GRondo will often win a gammon -- that could well be true. No
anomaly there. It's often correct to redouble from 4 to 8 in a 5
point match when your opponent has gammon chances.

I'm sure XG beats GRondo at 5 point matches far more often than GRondo
beats XG.

Paul

Bradley K. Sherman

2012-04-11 13:11:42 UTC

For very small values of often, unless there is a bug in XG.

--bks

Paul

2012-04-11 15:39:05 UTC

Post by Bradley K. Sherman

For very small values of often, unless there is a bug in XG.
--bks

Well, if "often" means 15%, there's nothing to suspect. Also, there
might be a small sample.

Anyway, it's completely obvious that GRondo's complaint is just an
ignorant first impression from someone who doesn't understand the
game. There's no reason to discuss bugs in this context.

Winning a gammon is irrelevant when the cube is at 8 but GRondo
doesn't appear to realise that.

Paul Epstein

GRondo

2012-04-11 17:56:05 UTC

...
Bradley,

Well, if "often" means 15%, there's nothing to suspect. Also, there

might be a small sample.
Paul Epstein

A sample ?
When I say I often beat it 16-0, it means I win the 5 points match in
90%, yes, 9 out of 10 games is a 8-0 or 16-0 win for me. In order to lose
I must play against myself :)
With jellyfish analyer (dice from a text file), snowie 4 or the latest
gnubb, this is the opposite, I lose 90% of my 5 points matches...

Bradley K. Sherman

2012-04-11 18:02:41 UTC

Post by GRondo

...
Bradley,

Well, if "often" means 15%, there's nothing to suspect. Also, there

might be a small sample.

Well, something's definitely wrong!

--bks

muratk

2012-04-13 10:58:42 UTC

Post by Bradley K. Sherman
Well, something's definitely wrong!

Nawh, it's probably just grondo's being a stupid newcomer... ;)

MK

Frank Berger

2012-04-12 21:51:18 UTC

Post by GRondo
A sample ?
When I say I often beat it 16-0, it means I win the 5 points match in
90%, yes, 9 out of 10 games is a 8-0 or 16-0 win for me. In order to lose
I must play against myself :)
With jellyfish analyer (dice from a text file), snowie 4 or the latest
gnubb, this is the opposite, I lose 90% of my 5 points matches...

In this case I recommend a fresh install. IIRC there has been a
similar issue discussed some month ago, where a fresh install helped.

ciao
Frank

muratk

2012-04-13 10:56:34 UTC

Post by Paul
Anyway, it's completely obvious that GRondo's complaint
is just an ignorant first impression from someone who
doesn't understand the game.

What is completely onvious to me is that you belong to
some underdeveloped "believer" species...

Post by Paul
Winning a gammon is irrelevant when the cube is at 8 but
GRondo doesn't appear to realise that.

Maybe eXtreme Garbage doesn't know the difference between
losing a match to 5 by a gammon when the cube is at 4 or
when the cube is at 8...?

Years ago I used to get a kick out of this and believed
that it was significant also.

If you are going to get gammoned and lose the 5-point
match by 8 to 0, why bother redoubling to 8 and lose it
by 16 to 0...? :))

That's what you dumb scums should be answering...

MK

muratk

2012-04-11 09:41:01 UTC

Post by GRondo
Thank you Murat.
Though I am a beginner, I often beat it by 16 to 0 in a 5 points match at
default level. It always proposes and accepts double and take.
Extremeroller++ is too slow on my Core2Duo. I wonder how to increase its
strength.

I wonder if you may be an experienced player making an ass of XG
roller?

But I have to take your word that you are a beginner and that you
caught
a bug/weakness in the "extreme garbage"...

Either way, none of the resident cocksucking scientists/gamblers of
rgb
will respond to you directly on the specific issue.

They are sick pack of dogs in denial and will gang up on any newcomer
who
question their religion, with irrelevant counter-comments... (I can't
even
say counter-arguments because their blabberings never amount to even
half
of an argument).

One previously missed comment I could make to you is that fetching
dice
from a file will not guarantee that the bot will not "cheat". (Refer
to
my article about how gnubg cheats).

This is exactly why the cocksuckers who develop and peddle various
bots
will not submit to measuring their products against a static yard
stick
like fossilized jellyfish...!

MK

Walt

2012-04-10 16:38:49 UTC

Post by GRondo
How strong is this sofware ?
Has it bugs ?
What about features ?
Is it worth the money ?
Would you prefer Snowie or GnuBB ?

As others have said, it's the strongest and fastest and has many nice
features. My hunch is that most casual players will be satisfied with
the free gnubg instead of paying the ~$50 for XG; whether it's worth the
money depends on how serious you are and how much you'll miss the fifty
bucks.

Snowie is obsolete.

muratk

2012-04-11 09:24:15 UTC

Post by Walt
As others have said, it's the strongest and fastest

Strongest by what measure? Did it play against other bots 4 billion
games...?

Even if it had, you still couldn't tell since each bot is supposedly
different.

Thus, you would need a common yard stick to measure them against each
other. Which would require real scientific approach...

But the cocksucking sick gamblers here posturing as PHD's don't even
have a clue about how to take scientific measurements... :(

Long ago, I suggested jellyfish as a "static" measuring stick. After
it was caught cheating with its pants down, it could at least serve a
purpose, as a never to be improved/modified bot in the future...

So, you make snowie play 4 billion games against jellyfish, extreme
gammon play 4 billion games against jelly fish, bgblitz or whatever
other cocksucking piggybacking robot play 4 billion games against
jellyfish...

And then you compare their achievements against a static opponent!

As simple and as scientific as it gets, right...?

But, nooo... We are not dealing with honest and/or scientist people
here. We are dealing with sick cocksucking gamblers trying to leech
off on unsuspecting beginners to pay for their worthless bots, books,
web pages, etc...

Post by Walt
and has many nice features.

I would agree with that only if you call bugs features, as microsoft
does...

Analysing ad nauseam based on arbitrary assumptions and constants are
plain unscientific and worthless!

Post by Walt
My hunch is that most casual players will be satisfied with the free
gnubg instead of paying the ~$50 for XG; whether it's worth the
money depends on how serious you are and how much you'll miss the fifty
bucks.

This is a typical cocksucking sales pitch. If you are a casual player,
gnubg should be good enough for you but otherwise you should be stupid
enough to pay $50 for a bot based on a stolen neural net engine from
gnubg...

Frankly, I thought the snowie guys were smarter to charge $400 for
their
sbot based on a stolen neural net engine from jellyfish... (Well,
unless
the cocksucking liar eventually became a partner in snowie??).

Post by Walt
Snowie is obsolete.

Just like jellyfish. And it's only a matter of time before all of its
offsprings become obsolete and discredited for the better future of
"backgammon"...

MK

Michael Petch

2012-04-11 09:52:44 UTC

Post by muratk
So, you make snowie play 4 billion games against jellyfish, extreme
gammon play 4 billion games against jelly fish, bgblitz or whatever
other cocksucking piggybacking robot play 4 billion games against
jellyfish...
And then you compare their achievements against a static opponent!

You may wish to look at the Deprelli studies.

Michael Petch

2012-04-11 10:03:39 UTC

You may wish to look at the Deprelli studies.

The current "Benchmark" by which other bots are now measured is XG2
3-ply Checker XG Roller Cube minimum 1296 trials no maximum until error
is less than 0.005.

It took five of us about 1.5 months to generate the Benchmark data.

Using the new benchmark data, we then run all the positions on a number
of the current bots (on different analysis levels) to find out how they
differ from the benchmark.

The 2012 data can be found here:

http://www.bgonline.org/forums/webbbs_config.pl?noframes;read=114338

muratk

2012-04-11 10:31:27 UTC

Post by muratk
And then you compare their achievements against a static opponent!

You may wish to look at the Deprelli studies.

Okay, I will. Give me a little time to digest it.

Post by Michael Petch
The current "Benchmark" by which other bots are now measured is XG2
3-ply Checker XG Roller Cube minimum 1296 trials no maximum until error
is less than 0.005.

Is XG measured against itself then? I have a little problem with the
measured being the measuring unit... But, I will look into it anyway.

Post by Michael Petch
It took five of us about 1.5 months to generate the Benchmark data.

I don't understand what this means. Do you mean it took you 1.5 months
to create the measuring stick? Or to compile the results of your
measurings??

Post by Michael Petch
Using the new benchmark data, we then run all the positions on a number
of the current bots (on different analysis levels) to find out how they
differ from the benchmark.

I don't know if I can understand what you did even if I made an
effort.

My approach is much much more simple. Let the contestants play a
number of
games against a static opponent (using the same set of dice and
jellyfish
is also ideal for this since it allows not only the seed but also the
counter value to be set manually), and just declare the winner based
on
number of games won.

Anything beyond that makes me suspicious of smoke and mirrors
science...

Post by Michael Petch
http://www.bgonline.org/forums/webbbs_config.pl?noframes;read=114338

Okay, I'll try to read and understand what you guys tried to do there
before
I post again.

MK

Michael Petch

2012-04-11 19:26:07 UTC

Post by muratk
Is XG measured against itself then? I have a little problem with the
measured being the measuring unit... But, I will look into it anyway.

You have to have a baseline. XG is being compared to a deep rollout
using itself. In 2010 the baseline was GNUBG Rollouts.

Post by muratk
I don't understand what this means. Do you mean it took you 1.5 months
to create the measuring stick? Or to compile the results of your
measurings??

Sorry the rollouts to create the baseline data took about 1.5 months.

Post by muratk
I don't know if I can understand what you did even if I made an
effort.

And why wouldn't you want to make an effort?

Post by muratk
My approach is much much more simple. Let the contestants play a
number of
games against a static opponent (using the same set of dice and
jellyfish
is also ideal for this since it allows not only the seed but also the
counter value to be set manually), and just declare the winner based
on
number of games won.

Actually this is what the program "Dueller" is for, and there have been
bot vs bot shootouts in the past.

In Depreli's study we wish to actually quantify the difference in the
bots strength based on the actual difference in equity of the errors
themselves.

Our method also allows us to see how bot strength evolves. With a
benchmark like this we simply take a new bot (or a new method of
evaluation in a current bot) and run it across all the Depreli positions
and see how it compares to the baseline.

Post by muratk
Okay, I'll try to read and understand what you guys tried to do there
before
I post again.

Thanks. A good starting point regarding the Depreli studies themselves
(which date back starting in 2005) can be found here:

http://www.bkgm.com/articles/Keith/DepreliBotComparison/index.html

muratk

2012-04-13 11:06:55 UTC