Re: [Bug-gnubg] random dice generator? hahahaaa

[Massimiliano Maini]

I think there's no solution to the "I think GNUBG cheats" problem.
The kind of person that screams that is the kind of person that:
1. Plays seven  21-points face-to-face matches against a much better
player, with totally fair dice rolled by a 3rd person and using whichever
complex device to avoid dice manipulations.
2. Loses all of them by a wide margin.
3. Concludes that the other guy has been incredibly lucky and didn't
deserve to win at all.

So you can try to explain that GNUBG can use an external, impartial and
statistically sound RNG, but I doubt you'd achieve much.


On 22 August 2017 at 09:34, Chris Wilson  wrote:

>
>
> On Mon, Aug 21, 2017 at 11:42 PM, Philippe Michel  > wrote:
>
>> On Sat, 19 Aug 2017, tchow wrote:
>>
>> The other feature that might be interesting is to allow players to
>>> receive a handicap, not by having the computer make mistakes, but by
>>> having the dice rigged in their favor.
>>>
>>
>> IMHO, the only thing worth implementing at the gnubg level would be the
>> possiility to call an external RNG provided as a python function.
>>
>> I'm not familiar with the python-C interfaces, and currently this is used
>> the
>> other way, accessing internal gnubg routines and data structures from
>> python scripts, but I assume the reverse is possible.
>>
>> Then anyone interested enough could use his preferred algorithm, or a
>> driver for
>> this kind of device : http://ubld.it/truerng_v3
>>
>> Simple unbalanced dice (SHG style) would be easy to implement, schemes
>> like the
>> one described by Timothy would be trickier, needing to evaluate the 21
>> rolls
>> before choosing one, but certainly doable (we could get rid of the
>> ridiculous
>> "dice manipulation" option and offer it as an example python code :-).
>>
>>
> I think there's an easier solution to this well-worn topic and it should
> be easy to implement.
>
> I have a Windows program called Super_PI that was written in 1995. I've
> kept it around this long because I use it as an objective means of rating
> computer builds and overclocking tweaks. The program generates 33,554,432
> digits of PI. It currently takes 8 minutes and 15 seconds on my system and
> it dumps the output to a text file. Now, strip the digits 0, 7, 8, 9 and
> you have a perfect file of truly random numbers. Any number of methods can
> be used to retrieve a valid roll from the file. Offer the numbers in
> suitable text blocks available as an additional download from the website
> and trigger a program option to use those numbers and it should end the
> argument.
>
> Here are some comments that come with the ZIP file.
>
> "In August 1995, the calculation of pi up to 4,294,960,000 decimal digits
> was succeeded by using a supercomputer at the University of Tokyo. The
> program was written by D.Takahashi and he collaborated with Dr. Y.Kanada at
> the computer center, the University of Tokyo. This record should be the
> current world record. ( Details is shown in the windows help. ) This
> record-breaking program was ported to personal computer environment such as
> Windows NT and Windows 95. In order to calculate 33.55 million digits, it
> takes within 3 days with Pentium 90MHz, 40MB main memory and 340MB
> available storage. The software is free and the circulation of program is
> also free!"
>
> To think it used to take 3 DAYS to generate that many digits! :-)
>
> The archive also includes some C++ code. Contact me if you wish a copy of
> the program.
>
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> https://lists.gnu.org/mailman/listinfo/bug-gnubg
>
>
___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] windows 10

[Massimiliano Maini]

Hi,

I just stumbled over his :
http://www.pcadvisor.co.uk/forum/consumerwatch-2/documents-read-only-after-windows-10-install-4537987/?p=2

Could it be related ?

MaX.

On 8 November 2015 at 16:40,  wrote:

> I have a 5 year old HP Pavilion laptop which had Vista originally, but I
> upgraded to Windows 7 and now to 10.  I had no problems using GNU
> Backgammon before, but now with Windows 10 when I click on it the horn
> plays for 2 seconds and then stops. Nothing opens, it just goes dead.  I
> tried Troubleshooting and it comes up with Incompatible Program Detected.
> GNUB is incompatible.
>
>
>
> How can I fix this please?
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> https://lists.gnu.org/mailman/listinfo/bug-gnubg
>
>
___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Possible small UI simplification

[Massimiliano Maini]

Disclaimer: I've lost my interest in backgammon a long ago, so ...

On 22 August 2015 at 00:12, Philippe Michel  wrote:

>
> If noise is added to the evaluations GNU Backgammon will take a Box Müller
> transform of a point in the unit circle and add to each possible outcome
> probability. This means that the addition is random, but distributed so
> that it's more likely to have an noise addition close to zero than a noise
> addition close to the limit. The noise addition is limited to the number
> you put into the noise field in the dialog box.
>
> If you check the box Deterministic noise, the noise added to each
> evaluation will be based on a sum of the bytes in the hash of the board
> position, which (by the central limit theorem) should have a normal
> distribution. In that way you will always have that same noise amount to a
> position, since the noise added to the evaluation is only depending on the
> position itself.
>

Of the two, I think the non-det one is the one I'd keep: at any rate it's
the one closer to my idea of noise.

I may see why a det-noise can be handy, but as the non-det noise actually
uses gnubg RNG, it's not really non-det: if needed, you can make it (kind
of) det giving it its own RNG instance with a seed, no ?

MaX.
___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] REQ - gnubg autoroll dice when closed out / on the bar.png

[Massimiliano Maini]

Hi,

you have to take a cube decision, even if you're closed out. That's why it
doesn't auto-roll.

MaX.

On 10 February 2015 at 10:47, Wayne Joseph  wrote:

> Hello all,
>
> I have a feature request to improve the efficiency of gameplay when
> playing against the CPU.
>
> Please see
>
> http://i.imgur.com/oAIsA1u.png
>
> (Please note that 'autoroll' is already switched on in options as well -
> with no apparent effect in this situation)
>
> Is it possible to implement autoroll of the dice on your turn when you are
> 'closed out' against the opponents 6-prime in their homeboard?
>
> I personally find it quite irritating to have to repeatedly click the
> mouse to roll when you know you have precisely *zero* chance of getting in
> against this impenetrable formation. I would like to hear others thoughts
> on this functionality too? It feels pointless to me (apart from the fact
> that it it slows the game down and adds insult to injury not-LoL).
>
> Perhaps it should even autoroll once there is a gap in your opponents
> homeboard if you continue to dance, until you roll a number that you can
> come back in on from the bar.
>
> Please do consider adding an 'on / off toggle option' for this
> functionality for the next release if time permits.
>
> Many thanks
>
> : )
>
> sebalotek
>
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> https://lists.gnu.org/mailman/listinfo/bug-gnubg
>
>
___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Bug-gnubg Digest, Vol 129, Issue 7

[Massimiliano Maini]

On 17 August 2013 19:36, Frank Berger  wrote:

>
> @all Other: How often do you have convinced one of the conspiracy
> theorists? From my experience it's < 20%
>
>
20% looks uber-optimistic to me. I'd say 10%, and that's probably generous
too.

MaX.
___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Analyse -> Market Window question?

[Massimiliano Maini]

On 9 June 2013 15:34, sebalotek  wrote:

> I've attached another example from my practice game today -
> If I'm reading the screenshot correctly my Take Point for the offered cube
> should be above 30.079.
> My Current win percentage =  31.0%.
>
> So why is it classified as 'Bad' when I do Take at this stage? Is it me
> being n00by again or is this a bug?
>
> 
>
> 

Hi,
notice that on your screenshot you actually have two takepoints: dead cube
at 37% and fully live at 30%. Your real take point is somewhere between the
two.

As said by Ian, gnubg uses Janowski's formulas for cubeful equities and these
try to estimate the equity considering the "real' cube liveliness, which is
between "dead cube" and "fully live cube".

They are called "market windows" because the real take point (or other point)
lies somewhere in the range they show.

I think that mostly the market windows are (by advanced/expert players) to
understand how the window changes depending on the score: at some scores
the window is huge, at some others it's tiny.

Max.

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] feature request: tuple2string move converter python function (inverse parsemove)

[Massimiliano Maini]

+1

On top, using the same code grants that gnubg and the python function will
represent the same move in the same manner. Always.

MaX.

On 25 October 2012 22:35, Michael Petch  wrote:
> On 2012-10-25 14:30, Guido Flohr wrote:
>>
>> Maybe buggy, maybe inefficient, but I still like my function. ;)
>>
>
> I don't see why GNUBG would want to reinvent the wheel for functionality
> it possesses but doesn't expose to Python directly. Regarding Python and
> the tuple to string issue it really comes down to creating a thin
> wrapper around existing functionality that already exists in our code.
>
> --
> Michael Petch
> CApp::Sysware Consulting Ltd.
> OpenPGP FingerPrint=D81C 6A0D 987E 7DA5 3219 6715 466A 2ACE 5CAE 3304
>
>
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> https://lists.gnu.org/mailman/listinfo/bug-gnubg

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] How badly do we need something like the old Dueller ?

[Massimiliano Maini]

Thanks for the info Tony, it makes sense.

Personally, I think we could drop snowie in a potential new release of
Dueller: no
activity in a while and no support from the developpers, which makes Dueller job
harder. On top, gnubg is at least as strong as Snowie and for a much more
interesting price :)

It would be nice to have Dueller working with gnubg, bgblitz (these already OK
AFAIK) and mostly extreme gammon, the new king.

MaX.

On 7 February 2012 17:14, Tony Lezard  wrote:
> That's precisely what Dueller already does now. The tricky part is that in 
> the case of programs like Snowie, where I haven't had any assistance from the 
> developers, it has to simulate a Snowiebot implementing DBSI by carefully 
> observing screen state and synthesising mouse/keyboard actions.
>
> The gnubg module currently uses that sort of interface (with the CLI version) 
> as well.
>
> Regards,
>
> Tony Lezard
>
> -Original Message-
> From: Massimiliano Maini [mailto:maxma...@gmail.com]
> Sent: 07 February 2012 15:03
> To: Frank Berger
> Cc: bug-gnubg@gnu.org; Tony Lezard
> Subject: Re: [Bug-gnubg] How badly do we need something like the old Dueller ?
>
> Hi Frank/Tony,
>
> if we were to redo the whole thing from scratch, wouldn't make more sense to 
> consider the bots just as evaluation engines and let the Dueller handle the 
> match ?
>
> I mean:
>
> - Dueller initializes the two bots: connection, play level, match/session 
> rules (crawford, met, jacoby, beaver etc).
>
> - Dueller sends each decision (move or cube decision) to the bot, getting the 
> answer. It will need to send the board, match score (for match play) and roll 
> (for checker play).
>
> So the dueller code could be slightly more involved (keep track of the 
> match/session, generate dice, etc), but the interface with the bots would be 
> much simpler.
>
> MaX.
>
> On 2 February 2012 23:34, Frank Berger  wrote:
>> Hi all,
>>
>> as promised:
>>
>> here http://www.bgblitz.de/download/UBGI/DBSI%200.6.1.doc is a link that 
>> describes the interface between Tony Lezards Dueller and BGBlitz.
>> This can be used as a base to implement a "remote control" for bots.
>>
>> To make it a little easier I put the BGBlitz code implementing the
>> interface here
>>
>> http://www.bgblitz.de/download/UBGI/DBSI.zip
>>
>> as an example which can be used freely. The most important classes are 
>> Server, SetState and Command.
>>
>> I hope extreme Gammon will implement it and/or other bots like Palamedes. 
>> Maybe a new platform independent server will be implemented too.
>> Kudos to Tony for developing it and make it available.
>>
>> ciao
>> Frank
>>
>>
>>
>> ___
>> Bug-gnubg mailing list
>> Bug-gnubg@gnu.org
>> https://lists.gnu.org/mailman/listinfo/bug-gnubg
>

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] How badly do we need something like the old Dueller ?

[Massimiliano Maini]

Hi Frank/Tony,

if we were to redo the whole thing from scratch, wouldn't make more sense to
consider the bots just as evaluation engines and let the Dueller
handle the match ?

I mean:

- Dueller initializes the two bots: connection, play level,
match/session rules (crawford,
met, jacoby, beaver etc).

- Dueller sends each decision (move or cube decision) to the bot,
getting the answer. It will
need to send the board, match score (for match play) and roll (for
checker play).

So the dueller code could be slightly more involved (keep track of the
match/session,
generate dice, etc), but the interface with the bots would be much simpler.

MaX.

On 2 February 2012 23:34, Frank Berger  wrote:
> Hi all,
>
> as promised:
>
> here http://www.bgblitz.de/download/UBGI/DBSI%200.6.1.doc is a link that 
> describes the interface between Tony Lezards Dueller and BGBlitz.
> This can be used as a base to implement a "remote control" for bots.
>
> To make it a little easier I put the BGBlitz code implementing the interface 
> here
>
> http://www.bgblitz.de/download/UBGI/DBSI.zip
>
> as an example which can be used freely. The most important classes are 
> Server, SetState and Command.
>
> I hope extreme Gammon will implement it and/or other bots like Palamedes. 
> Maybe a new platform independent server will be implemented too.
> Kudos to Tony for developing it and make it available.
>
> ciao
> Frank
>
>
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> https://lists.gnu.org/mailman/listinfo/bug-gnubg

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] 1-ply vs 0-ply equity improvement

[Massimiliano Maini]

How badly do we need something like the old Dueller ?

Something that alows two bots (different ones or same bot on different
settings) to
play head to head and see who's stronger.

I don't remember if it was on this mailing list or on a bg forum (bgonline ?),
but somebody was talking about a general BG playing interface/protocol
to allow bots to play one against the other and/or on BG servers.

MaX.

2012/1/30 Øystein Schønning-Johansen :
> Sounds like a good answer to me.
>
> However. Keep in mind that this is only for the contact positions. How this
> number will change is you compare all classes 0-ply vs. 1-ply (crashed, race
> and bearoff in addition), is still guesswork. I guess it will be between
> 0.02 and 0.03 ppg.
>
> -Øystein

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] 0ply question

[Massimiliano Maini]

Hi all,

just stumbled on this. Take the position below: all the moves have
cubeful equity of -1,
as next turn, O would double and X will have to drop, independently on
which move
X makes now.

Question: how can 0ply get this right, know that next cube decision is a D/P ?

I was under the impression that 0ply should only feed the resulting
position in the net,
get the cubeless numbers and translate them to cubeful using janowski's stuff.

Does 0ply checker take into account the next cube decision ?

MaX.

GNU Backgammon  Position ID: UAAAQAUAAA
 Match ID   : cIno
+13-14-15-16-17-18--19-20-21-22-23-24-+ O: Alex Rietti
|  |   | O  O | OOO 0 points
|  |   |  | OOO
|  |   |  | OOO
|  |   |  | OO
|  |   |  | OO
v|  |BAR|  | 7 point match (Cube: 1)
|  |   |  | XX
|  |   |  | XX
|  |   |  | XX
|  |   |  | XXX Rolled 12
|  |   | X  X  X  | XXX 0 points
+12-11-10--9--8--7---6--5--4--3--2--1-+ X: Carlo Melzi

1. Cubeful 0-ply5/3 4/3  Eq.:  -1,000
   0,167 0,000 0,000 - 0,833 0,000 0,000
0-ply cubeful, noise 0,06 (d) [beginner]
2. Cubeful 0-ply6/3  Eq.:  -1,000 ( +0,000)
   0,155 0,000 0,000 - 0,845 0,000 0,000
0-ply cubeful, noise 0,06 (d) [beginner]
3. Cubeful 0-ply5/2  Eq.:  -1,000 ( +0,000)
   0,155 0,000 0,000 - 0,845 0,000 0,000
0-ply cubeful, noise 0,06 (d) [beginner]
4. Cubeful 0-ply6/4 5/4  Eq.:  -1,000 ( +0,000)
   0,141 0,000 0,000 - 0,859 0,000 0,000
0-ply cubeful, noise 0,06 (d) [beginner]
5. Cubeful 0-ply4/1  Eq.:  -1,000 ( +0,000)
   0,133 0,000 0,000 - 0,867 0,000 0,000
0-ply cubeful, noise 0,06 (d) [beginner]
6. Cubeful 0-ply6/5 4/2  Eq.:  -1,000 ( +0,000)
   0,126 0,000 0,000 - 0,874 0,000 0,000
0-ply cubeful, noise 0,06 (d) [beginner]

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] pubeval benchmark

[Massimiliano Maini]

Hi Joseph,

happy to see you work again on all this :)

Very interesting question you've raised here.

XG developer (Xavier Dufaure de Citres) is active on BGonline.com
(http://www.bgonline.org/forums/): if you ask your question (gnubg 0ply
vs XG 1ply) there, he may be willing to help.

MaX.

On 20 January 2012 00:18, Joseph Heled  wrote:
> Speaking of bot to bot comparisons...
>
> Anyone knows anything about the "raw" strength of the XG variants? By
> raw I mean a the 0-ply equivalent, i.e. without any lookup.
>
> I guess what I am asking is: how much of  XG strength comes from
> quality of implementation, and how much from the basic net evaluation?
>
> -Joseph

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Questions about gnubg-nn tools

[Massimiliano Maini]

That would be interesting indeed.
Do you have any description of what your own cube code does ?
At the time I just went as far as verifying that the current cube code is.
OK with respect to R.Janowksi's formulae (and it was).

Not sure I can help in integrating your code in gnubg, but I'm definitely
curious in understanding what's different with respect to the current one.

MaX.

On 5 January 2012 09:35, Joseph Heled  wrote:
> I wonder if this is just the net or the cube code contributes as well.
> As you know, the cube code in gnubg-nn is different from the one used in
> gnubg. (my own dewvelopment).
> I always thought it did better, but it will be interesting if someone
> verifies it, and offers it as a gnubg option  (especially with regard to the
> odd/even issue).
>
> -Joseph
>
> On 5 January 2012 21:17, Massimiliano Maini  wrote:
>>
>> On 5 January 2012 02:20, Joseph Heled  wrote:
>> > Seems like you got a very very slightly better race net, but I would be
>> > surprised if it makes a difference in real life.
>> >
>> > Would be much more interesting to
>> >   - get a better contact or crashed net
>> >   - expand the roll-out database for all categories (should be easy with
>> > the
>> > current availability of cycles)
>> >   - improve cube decisions (this is a hard one)
>> >   - improve back game evaluation and play (very hard one)
>> >
>> > -Joseph
>>
>> One thing that has always puzzled me is the strange behavior of the
>> strength
>> of gnubg at different plies. The last large scale study (done to compare
>> the
>> existing bots to the new extreme gammon), is resumed here:
>>
>>  http://www.extremegammon.com/studies.aspx
>>
>> Checker play is fine, error goes down as plies go up.
>> However, for cube actions, the situation is very strange:
>>
>> 3ply does much better than 2ply and 4ply on missed doubles and wrong
>> takes.
>> But 3ply does terribly worse than 2ply and 4ply on wrong doubles and
>> wrong passes.
>>
>> It has always been accounted to the so called "odd-even" effect, but
>> no other bot
>> seems to be affected as much as gnubg.
>>
>> MaX.
>>
>> ___
>> Bug-gnubg mailing list
>> Bug-gnubg@gnu.org
>> https://lists.gnu.org/mailman/listinfo/bug-gnubg
>
>

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Questions about gnubg-nn tools

[Massimiliano Maini]

On 5 January 2012 02:20, Joseph Heled  wrote:
> Seems like you got a very very slightly better race net, but I would be
> surprised if it makes a difference in real life.
>
> Would be much more interesting to
>   - get a better contact or crashed net
>   - expand the roll-out database for all categories (should be easy with the
> current availability of cycles)
>   - improve cube decisions (this is a hard one)
>   - improve back game evaluation and play (very hard one)
>
> -Joseph

One thing that has always puzzled me is the strange behavior of the strength
of gnubg at different plies. The last large scale study (done to compare the
existing bots to the new extreme gammon), is resumed here:

  http://www.extremegammon.com/studies.aspx

Checker play is fine, error goes down as plies go up.
However, for cube actions, the situation is very strange:

3ply does much better than 2ply and 4ply on missed doubles and wrong takes.
But 3ply does terribly worse than 2ply and 4ply on wrong doubles and
wrong passes.

It has always been accounted to the so called "odd-even" effect, but
no other bot
seems to be affected as much as gnubg.

MaX.

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] End game bug ?

[Massimiliano Maini]

Hi,
just noticed a strange bug. Enter this position:

Position ID: 3RYAAOxuGwABAA
Match ID: UQnvAAAE

The click "end game". The dialog showing the end of game (with updated
score) appears, but then nothing happens once you click on the
OK button. Impossible to move forward to the next game.

MaX.

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Analyse / Rollout - do we need it?

[Massimiliano Maini]

doubtful/bad/very bad decisions could be "Flagged" moves,
leaving "Marked" for user selected moves.

Max.

On 14 October 2011 20:34, Philippe Michel  wrote:
> On Fri, 14 Oct 2011, Michael Petch wrote:
>
>> On 14/10/2011 4:07 AM, Ian Shaw wrote:
>
>>> 2) Why is "Cmark" called "Cmark" and not simply "Mark"?  What information
>>> does the "C" convey?
>
>> I was curious about this naming convention as well. It has confused some
>> people in the past. They thought CMARK was to mark cubes. Simply going
>> with "Mark" makes sense to me.
>
> "Mark" is somehow already used for for the doubtful/bad/very bad decisions.
> See the red arrows vs. blue arrows in the GUI.
>
> "CMark" is still not very expressive, of course.
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> https://lists.gnu.org/mailman/listinfo/bug-gnubg
>

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] 16th Computer Olympiad, Tilburg, the Netherlands : November 18 to November 26, 2011

[Massimiliano Maini]

Hi all,

I'll reply to this with the bit of info I can provide.
I participated on behalf of gnubg team in a past edition of computer olympics.
Due to lack of participants (only Frank and me), the format was absolutely
free, just some time constranits (having it over before a given time).
Atmosphere is 100% relax.

I don't see any particular requirement beside:
- having a non-prehistoric computer
- knowing how to "operate" gnubg, but I'm sure Frank knows this :)

Frank is a great person and a true backgammon passionate.
We all know the title is purely symbolical: it's not a short session that is
gonna decide which bot is the best. But it's a funny experience and you
can always put a beer in the side pot to add some spice to the match :)

I think there's some kind of fee to participate.
When I participated, gnubg won and I've shipped the medal to one of
the main developers (think it was Jon).

MaX.

On 16 August 2011 21:26, boomslang  wrote:
> Hi Frank,
>
>
>
> What exactly are the requirements to participate? What format will the 
> matches be and how much time does it take approximately? Are there any costs 
> involved?
>
> Did anyone from the GNUbg team respond yet?
>
>
> Kind regards,
>
>
>
> boomslang

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
https://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] New win builds on gnubg.org (20110310 code)

[Massimiliano Maini]

As an additional advice, you may event want to uninstall your current gnubg
first: it will (dos) prompt to keep or not your preferences so that,
if you answer
yes and then install with Michael's installer, all your settings should be kept.

MaX.

On 10 March 2011 21:48, Michael Petch  wrote:
> I will be doing builds in MaX's absence. I decided to build today's
> release to fix a major problem that could cause instabilities with
> anyone using www.random.org for their dice generation. I highly
> recommend this update for anyone that uses Random.org .
>
> This release also gives us a chance to see if my builds (with slightly
> different GTK library versions) work as well as the versions MaX uses. I
> recommend using the full installer for this release, as one new DLL is
> needed (libpng14-14.dll).
>
> --
> Michael Petch
> CApp::Sysware Consulting Ltd.
> OpenPGP FingerPrint=D81C 6A0D 987E 7DA5 3219 6715 466A 2ACE 5CAE 3304
>
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] Windows builds

[Massimiliano Maini]

Hi guys,

I'll be moving to a new location for at least 4 months starting
mid-march and I won't be
able to create the windows builds on gnubg.org (believe it or not, I
won't have a PC at
hand, only a MacBook, sigh).

After that, either I come back home or I settle there for good, in
both cases I'll be able
again to create the builds.

Anyone interested in in doing this in my absence ?

MaX.

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] Re: Windows Build

[Massimiliano Maini]

Hi Michael, just built.

Cheers, MaX.

On 14 January 2011 12:58, Michael Petch  wrote:
> Howdy Max,
>
> Rudolf is having some trouble with random.org and I have committed a
> POSSIBLE fix. I don't have a windows build environment, and was wondering if
> you could build a new version for him to test?
>
> Thanks,
> Michael
>

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Re: Windows build of gnubg

[Massimiliano Maini]

On 28 December 2010 01:33, motiv4u  wrote:
> Someone knows how to open them
> (them = the sources on http://www.gnubg.org/media/sources/ with OS Windows)
>
> use 7zip ( http://www.7-zip.org/ freeware )
>
> in stead of winrar
>
> n.

Maybe, still freaky though ...

MaX.

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] Re: Windows build of gnubg

[Massimiliano Maini]

Oups, freaky: if I download via firefox the files are corrupted, while
via ftp they are OK.

Anyway, new build just uploaded.
Some new warnings (maybe not that important):

html.c: In function 'surlencode':
html.c:3732: warning: suggest parentheses around assignment used as truth value
set.c: In function 'SetXGID':
set.c:4141: warning: array subscript is above array bounds

MaX.

On 15 December 2010 09:37, Massimiliano Maini  wrote:
> Hi Philippe/all,
>
> sorry for being that lazy :)
>
> I have a problem: I tried to download the last 3 source archives
> (20101213/14/15) from here:
>
>   http://www.gnubg.org/media/sources/?C=N;O=D
>
> but they are all corrupt. Any idea ?
>
> MaX
>
> On 14 December 2010 23:48, Philippe Michel  wrote:
>> Dear Max,
>>
>> Would you mind making a new Windows build from the current state of gnubg
>> sources ?
>>
>> I have recently added fixes or committed code or patches submitted by
>> others, but have no further plans for now. Other contributors seem
>> relatively inactive at this time as well.
>>
>> Since some of these fixes will avoid crashes and assertion failures, I think
>> it would be useful to make them available in the Windows distribution.
>>
>> Best regards,
>>
>> Philippe Michel
>>
>>
>>
>

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] Re: Windows build of gnubg

[Massimiliano Maini]

Hi Philippe/all,

sorry for being that lazy :)

I have a problem: I tried to download the last 3 source archives
(20101213/14/15) from here:

   http://www.gnubg.org/media/sources/?C=N;O=D

but they are all corrupt. Any idea ?

MaX

On 14 December 2010 23:48, Philippe Michel  wrote:
> Dear Max,
>
> Would you mind making a new Windows build from the current state of gnubg
> sources ?
>
> I have recently added fixes or committed code or patches submitted by
> others, but have no further plans for now. Other contributors seem
> relatively inactive at this time as well.
>
> Since some of these fixes will avoid crashes and assertion failures, I think
> it would be useful to make them available in the Windows distribution.
>
> Best regards,
>
> Philippe Michel
>
>
>

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] New win builds on gnubg.org (20100816 code) - EOM

[Massimiliano Maini]

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] New Win install archive - timestamp 20100713

[Massimiliano Maini]

As always, on gnubg.org.

MaX.

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] (OT) Position ID documentation

[Massimiliano Maini]

Got the problem, you said:

> I think from your explanation, it should be:
>
> White = Dnm(25-lop+1, 15) - 1

When it's actually:

   White = Dnm(25-(lop+1)+1, 15) - 1

Now it's OK:

NC pos:1402609279900141, 1.40e+15

MaX.

On 8 June 2010 07:28, bgnj  wrote:
>
>
> Massimiliano Maini-3 wrote:
>>
>> NC_pos += D(M+1,15-1) * (D(25-M+1, 15) - 1)
>> I still get:
>> NC pos:        3085625541477601, 3.09e+15
>>
>
> I modified my program to compute using your formula:
>
> M=1, P1=15, P2=15471286559, P1*P2=232069298385, total=232069298385
> M=2, P1=120, P2=9364199759, P1*P2=1123703971080, total=1355773269465
> M=3, P1=680, P2=5567902559, P1*P2=3786173740120, total=5141947009585
> M=4, P1=3060, P2=3247943159, P1*P2=9938706066540, total=15080653076125
> M=5, P1=11628, P2=1855967519, P1*P2=21581190310932, total=36661843387057
> M=6, P1=38760, P2=1037158319, P1*P2=4020025640, total=76862099831497
> M=7, P1=116280, P2=565722719, P1*P2=65782237765320, total=142644337596817
> M=8, P1=319770, P2=300540194, P1*P2=96103737835380, total=238748075432197
> M=9, P1=817190, P2=155117519, P1*P2=126760485351610, total=365508560783807
> M=10, P1=1961256, P2=77558759, P1*P2=152112581441304, total=517621142225111
> M=11, P1=4457400, P2=37442159, P1*P2=166894679526600, total=684515821751711
> M=12, P1=9657700, P2=17383859, P1*P2=167888095064300, total=852403916816011
> M=13, P1=20058300, P2=7726159, P1*P2=154973615069700, total=1007377531885711
> M=14, P1=40116600, P2=3268759, P1*P2=131131497299400, total=1138509029185111
> M=15, P1=77558760, P2=1307503, P1*P2=101408311376280, total=1239917340561391
> M=16, P1=145422675, P2=490313, P1*P2=71302628047275, total=1311219968608666
> M=17, P1=265182525, P2=170543, P1*P2=45225023361075, total=1356444991969741
> M=18, P1=471435600, P2=54263, P1*P2=25581509962800, total=1382026501932541
> M=19, P1=818809200, P2=15503, P1*P2=12693999027600, total=1394720500960141
> M=20, P1=1391975640, P2=3875, P1*P2=5393905605000, total=1400114406565141
> M=21, P1=2319959400, P2=815, P1*P2=1890766911000, total=1402005173476141
> M=22, P1=3796297200, P2=135, P1*P2=512500122000, total=1402517673598141
> M=23, P1=6107086800, P2=15, P1*P2=91606302000, total=1402609279900141
> M=24, P1=9669554100, P2=0, P1*P2=0, total=1402609279900141
>
> You can compare to your program and see where the two differ.
>
> - BGNJ
> --
> View this message in context: 
> http://old.nabble.com/Re%3A-%28OT%29-Position-ID-documentation-tp28666843p28813912.html
> Sent from the Gnu - Backgammon mailing list archive at Nabble.com.
>
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] (OT) Position ID documentation

[Massimiliano Maini]

Doh, I think you're right .. .went for a complicate way to solve a simple issue,
just do the -1  :)

Yet with your fix:

==
# Non-contact positions.
NC_pos = 0
for M in range(1,23+1):
# Black has 1 checker on point M, 15-1 checkers on points
# 0..M (0 meaning off).
# White can be on points (M+1)..25 (25 meaning off), but not
# all 15 are off, hence the -1. Nobody is on the bar.
NC_pos += D(M+1,15-1) * (D(25-M+1, 15) - 1)
==

I still get:

 NC pos:3085625541477601, 3.09e+15

Which is more than your 1.4E+15.

On 6 June 2010 07:57, bgnj  wrote:
>
>
> Massimiliano Maini-3 wrote:
>>
>> I did the computation this way:
>>
>> # Non-contact positions.
>> NC_pos = 0
>> for lop in range(1,23+1):
>>       # Black has 1 checker on point lop, 15-1 checkers on points
>>       # 0..lop (0 meaning off).
>>       # White can be on points lop+1..25 (25 meaning off), but at
>>       # least one is on points lop+1..24. Nobody is on the bar.
>>       NC_pos += Dnm(lop+1,15-1)*(Cnm(24-(lop+1)+1,1)*Dnm(25-(lop+1)+1,15-1))
>>
>> I get this:
>>
>> Non-contact positions: 9350831674578000 (9E+15).
>>
>
> Massimiliano, I think your formula is incorrect.  For white, you have:
>
> White = (Cnm(24-(lop+1)+1,1)*Dnm(25-(lop+1)+1,15-1))
>
> With the explanation:
>
> # White can be on points lop+1..25 (25 meaning off), but at
> # least one is on points lop+1..24. Nobody is on the bar.
>
> I think from your explanation, it should be:
>
> White = Dnm(25-lop+1, 15) - 1
>
> where the -1 accounts for the one case that all checkers are off.  For
> example, if lop = 22, and therefore white has 3 spaces to distribute 15
> checkers, your formula gives:
>
> White = Cnm(2, 1) * Dnm(3, 14) = 2 * 120 = 240
>
> But that number exceeds the total number of ways to distribute 15 checkers
> on 3 spaces, which is:
>
> Dnm(3, 15) = 136
>
> Your formula gives the number of ways to distribute 15 checkers with 1
> checker being distinguishable from all other checkers (and this special
> checker is not in the tray).  This number is larger than the case where all
> checkers are indistinguishable from each other.  I hope that made sense.
>
> - BGNJ
> --
> View this message in context: 
> http://old.nabble.com/Re%3A-%28OT%29-Position-ID-documentation-tp28666843p28793143.html
> Sent from the Gnu - Backgammon mailing list archive at Nabble.com.
>
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] (OT) Position ID documentation

[Massimiliano Maini]

2010/6/3 Øystein Johansen :
> On Sun, May 30, 2010 at 3:11 AM, bgnj  wrote:
>>
>> This is an interesting problem.  I counted all the non-contact positions
>> in
>> order to see if that would get the number of positions down to 64 bits but
>> unfortunately it did not.

I did the computation this way:

# Non-contact positions.
NC_pos = 0
for lop in range(1,23+1):
# Black has 1 checker on point lop, 15-1 checkers on points
# 0..lop (0 meaning off).
# White can be on points lop+1..25 (25 meaning off), but at
# least one is on points lop+1..24. Nobody is on the bar.
NC_pos += Dnm(lop+1,15-1)*(Cnm(24-(lop+1)+1,1)*Dnm(25-(lop+1)+1,15-1))

I get this:

Non-contact positions: 9350831674578000 (9E+15).
Not enough.

Btw, the positions that are game over (15 black or white checkers out) are
simply 2*(Dnm(24+2,15)-1). This gives: 80450690110 (8E+10), very far from
what we need.

> MaX: I like your idea as well. (And I like your python code). 11 points each
> should be enough for any normal position, but remember this is theoretical
> discussion of only academic interest, and you solution is more like a
> pragmatic solution rather than a theoretical perfect solution.  :-)

Indeed. But for the theoretical discussion we probably have the answer:
given he delta between the total number of pos and 2^64 (delta is 8.2E16)
it's unlikely that stripping game over and illegal will bring us under 2^64.

Any attempt to separate the pos in 2 or more 'sets' and use different encodings
is not an answer to the theoretical question.

MaX.

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] (OT) Position ID documentation

[Massimiliano Maini]

I've some python that computes the number of positions where both
players occupy no more than X
regular points (board points, 1..24).

The computation is a sophistication of W.Trice's one: you assume that
black occupies m regular points
(m being of course between 0 and you limit X) and you compute number
of positions for black checkers
in the same way as Walter. Then for white it's more complicate: you
can't use just D(26-m,15) as you
want to limit the number of occupied points, so you just do the same
reasoning you've just done for
black with the additional note that if black occupies m points, white
can't occupy more than the min
between X (your imposed limit), 15 (max # of checkers) and  24-m
(vacant points).

It turns out (if my computations are OK) that limiting black and white
to be holding no more than 11
regular points each, then the number of points is already less than 2^64.

Python output here below, code further below. I'm ashamed about the
naive implementation of Cnm,
but hey, it works and takes no time, screw efficiency  :)

MaX.


=

C:\Documents and Settings\mmaini\Desktop\BG
posID>d:\python26\python.exe bgposid.py
TOT1: 18528584051601162496,1.852858e+19
TOT2: 18528584051601162496,1.852858e+19
0,0.00e+00
False False
Lim,#Pos,in64
0,256,1
1,8041216,1
2,20822945536,1
3,9822623914496,1
4,1194790350513152,1
5,45681766196850176,1
6,629989668141016576,1
7,3551999654903397376,1
8,9550656424722482176,1
9,15092292700890794496,1
10,17700293897831133696,1
11,18404277853614314496,1
12,18517676380201988096,0
13,18528090608482589696,0
14,18528575503767612416,0
15,18528584051601162496,0

=

import math;

def Cnm(n,m):
return 
(math.factorial(long(n))/(math.factorial(long(m))*math.factorial(long(n-m

def Dnm(n,m):
return Cnm(long(n+m-1),long(m))


def BlackOnM(m):
# Black occupies m regular points (board points, 0..24).
T1 = Cnm(24,m)
# The remaining 15-m black checkers are distributed over
# the occupied m points + the bar and the off (m+2 points). 
T2 = Dnm(m+2,15-m)
# The 15 whie checkers are distributed over the 24-m
# available regular points, plus the bar and the off (26-m).
T3 = Dnm(26-m,15)
return long(T1*T2*T3)

def BlackOnM_WithLim(m,lim):
# Black occupies m regular points (board points, 0..24).
T1 = Cnm(24,m)
# The remaining 15-m black checkers are distributed over
# the occupied m points + the bar and the off (m+2 points). 
T2 = Dnm(m+2,15-m)
# White can occupy a number of points that is the minimum
# between 15 (number of checkers), 24-m (points left vacant
# by black) and lim (self-imposed limit).
T3 = 0
for p in range(0,min(lim,15,24-m)+1):
# White occupies p regular points out of the available 24-m.
T31 = Cnm(24-m,p)
# The remaining 15-p white checkers are distributed over
# the occupied p points + the bar and the off (m+2 points).
T32 = Dnm(p+2,15-p)
T3 += T31*T32
return long(T1*T2*T3)


def TotPos():
Tot = 0
for M in range(0,16):
Bm = BlackOnM(M)
Tot += Bm
return Tot

def TotPos_WithLim(lim):
Tot = 0
for M in range(0,lim+1):
BmL = BlackOnM_WithLim(M,lim)
Tot += BmL
return Tot


# Sanity check :)
TOT1 = TotPos()
TOT2 = TotPos_WithLim(15)
print 'TOT1: %d,%e' % (TOT1,TOT1)
print 'TOT2: %d,%e' % (TOT2,TOT2)
print '%d,%e' % (TOT1-TOT2,TOT1-TOT2)
print TOT1<(pow(2,64)), TOT2<(pow(2,64))

# Iterate on Lim.
print 'Lim,#Pos,in64'
for lim in range(0,16):
NP = TotPos_WithLim(lim)
print '%d,%d,%d' % (lim,NP,NPhttp://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] (OT) Position ID documentation

[Massimiliano Maini]

Maybe the easier thing is to go with python (native long int support).

MaX.

2010/5/28 Øystein Johansen :
>
>
> 2010/5/28 Massimiliano Maini 
>>
>> Strange, we don't get the same number as Trice ...
>>
>> you & I : 18,528,584,051,601,166,000
>> Trice    : 18,528,584,051,601,162,496
>>
>> There's something strange in googledocs: look at the value of the product
>> for the line with m=3 ... it ends in 808 ...
>>
>
> Really funny, Could be an internal bug in google docs. Each number is
> converted into a 32bit (or 64bit) real type and then multiplied and then
> converted back to integer. I wonder what I get it I add an decimal?
>
> Maybe Walter had the same problem?
>
> I may check with gmp.
>
> -Øystein
>

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] (OT) Position ID documentation

[Massimiliano Maini]

Strange, we don't get the same number as Trice ...

you & I : 18,528,584,051,601,166,000
Trice: 18,528,584,051,601,162,496

There's something strange in googledocs: look at the value of the product
for the line with m=3 ... it ends in 808 ...

MaX.



2010/5/28 Øystein Johansen :
> ??? Strange. I copied and pasted from a web site.
>
> function productRange(a,b) {
>   var product=a,i=a;
>
>   while (i++     product*=i;
>   }
>   return product;
> }
>
> function combinations(n,k) {
>   if (n==k) {
>     return 1;
>   } else {
>     k=Math.max(k,n-k);
>     return productRange(k+1,n)/productRange(1,n-k);
>   }
> }
>
> function distributions(N, M) {
>   return  combinations( N+M-1,M);
> }
>
> Maybe I should not trust everything I read. However, I can't immediately see
> any bugs in the above code
> Aha! Now I see! It's wrong when k = 0, I'll fix it. Hang on.
>
> May be a bad fix, but I think it works better now.
>
> -Øystein
>
>
> 2010/5/28 Massimiliano Maini 
>>
>> Now it works.
>>
>> Your combinations function seems to be wrong for m=0:
>>
>> combin(24,0) = 1
>> combinations(24,0) = 25
>>
>> Guess you just have to define distribution as a function of COMBIN ...
>>
>> MaX.
>>
>> 2010/5/28 Øystein Johansen :
>> > 2010/5/28 Massimiliano Maini 
>> >>
>> >> Google docs, right .. excel is so has been :)
>> >>
>> >> BTW, your link does not work.
>> >
>> > I see  I'll try to publish it as a web page:
>> >
>> >
>> > http://spreadsheets.google.com/pub?key=tqUcK-zFdgQC_byUJvbgqvQ&output=html
>> >
>> > -Øystein
>> >
>> >
>
>

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] (OT) Position ID documentation

[Massimiliano Maini]

Now it works.

Your combinations function seems to be wrong for m=0:

combin(24,0) = 1
combinations(24,0) = 25

Guess you just have to define distribution as a function of COMBIN ...

MaX.

2010/5/28 Øystein Johansen :
> 2010/5/28 Massimiliano Maini 
>>
>> Google docs, right .. excel is so has been :)
>>
>> BTW, your link does not work.
>
> I see  I'll try to publish it as a web page:
>
> http://spreadsheets.google.com/pub?key=tqUcK-zFdgQC_byUJvbgqvQ&output=html
>
> -Øystein
>
>

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] (OT) Position ID documentation

[Massimiliano Maini]

Google docs, right .. excel is so has been :)

BTW, your link does not work.

MaX.

2010/5/28 Øystein Johansen :
>
>
> 2010/5/28 Massimiliano Maini 
>>
>> Hi all,
>>
>> another criteria to separate usual positions from unusual ones could
>> be the number of occupied points. In Trice's formula you sum over the
>> range 0..15 for m:
>>
>>  C(24,m) X D(m+2,15-m) X D(26-m,15)
>>
>> m being the (regular) points occupied by black.
>>
>> Now, how frequent is that black holds 15 points ?
>
> Good idea!
>
>>
>> According to rough calculations (excel not having big integers),
>> all the positions where black occupies 10 or less (regular)  points are
>> already less than 2^64.
>
> True!
>
> Do not use Excel because it has limitations like the one you mention, and
> lot's of others. Use Google Docs Spreadsheet, and you can quite easily
> extend it with Javascript functions. And of course you can share your
> spreadsheet with others. I've made the Trice calculation in Google Docs
> here: https://spreadsheets.google.com/ccc?key=0AiXPuo_wblgydHFVY0stekZkZ1
>
>>
>> This can even be refined, as in the total there are positions where
>> white occupies more than 11 points (and impossible positions, game
>> over positions etc).
>
> Yes, I believe this is theoretical possible. Thanks for your suggestions.
> I then just have to find the perfect hash function. I guess that can be done
> in couple of hours. I'll just get a cup of coffee first. I'll post a follow
> up mail when I've found it. :-)
>
>
> -Øystein

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] (OT) Position ID documentation

[Massimiliano Maini]

Hi all,

another criteria to separate usual positions from unusual ones could
be the number of occupied points. In Trice's formula you sum over the
range 0..15 for m:

  C(24,m) X D(m+2,15-m) X D(26-m,15)

m being the (regular) points occupied by black.

Now, how frequent is that black holds 15 points ?

According to rough calculations (excel not having big integers),
all the positions where black occupies 10 or less (regular)  points are
already less than 2^64.

This can even be refined, as in the total there are positions where
white occupies more than 11 points (and impossible positions, game
over positions etc).

Regards,

2010/5/25 Øystein Johansen :
>
> Yes, this is the kind of position that will then need a special treatment.
> Not really 'unlikely'? Maybe we should make a different definition of
> 'unlikely'. Jon's idea is really good. Kind of like UTF-8. However, it's
> still only of pure academic interest.
>
> -Øystein
>
>
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>
>

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] Re: gnu hypergammon databases

[Massimiliano Maini]

Hi Rick,

sorry for late reply, busy period.

I don't know if hypergammon db can be found somewhere on the web,
but you can reconstruct it easily using the makehyper.exe file provided
in windows distribs (I think default parameters are just fine, use
makehyper --help to see them).

I've put the gnubg mailing list in copy of this mail as other may help/comment.
Notice that if you want to post to the mailing list (do not hesitate if you have
queastions similar to this one), you may have to register here first:

  http://lists.gnu.org/mailman/listinfo/bug-gnubg

MaX.

On 8 May 2010 03:31, Rick Janowski  wrote:
> Max,
>
> Would you be able to help me with a problem i have downloading
> hyperbackgammon gnu databases.
>
> I did this myself a few months ago but now cannot find the link.
>
> If you do not know, would you be able to find me someone who does?
>
> Kind regards,
>
> Rick
>

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Problems with doubling

[Massimiliano Maini]

I would even think the same should apply to match play: when you win
more points than you need to reach the score limit (e.g. you win 4 points
being at -2), gnubg should award you only the points you need (2).

If you lead 20-1 in a 21 points match and you win simple with the cube at 2,
gnubg should shoud that you won 1 point and the match ends at 21-1
instead of showing that you won 2 points ending the match at 22-1 (which
I think is the case now).

MaX.

On 18 March 2010 11:50, motiv4u  wrote:
> Robert does not point out the Cube - No Cube issue anymore (correct me
> if I'm wrong)
>
> If you play a 1-point match, the final score is 1-0 or 0-1.
> However, GNU Backgammon tells you you win - or lose - 1, 2 or 3 points
> and that is indeed confusing.
>
> Setting up a 1-point match and winning it with backgammon gives you 1
> point and the match. However, if you win a backgammon, gnubg tells
> you:
>
>  Player wins a backgammon and 3 points.
>  The score (after 1 game) is: gnubg 0, Player 3 (match to 1 point,
> post-Crawford play).
>  Player has won the match.
>
> Both the "backgammon and 3 points" and the "post-Crawford play" are
> confusing, especially for players not used to money sessions, match
> play, 1-point games. Not for the 'habitués' of course, but
> nevertheless...
>
> My 2 1/2 cents.
>
> N.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] New win builds on gnubg.org (20100220 code)

[Massimiliano Maini]

That's strange cause I downloaded and installed it and I see the met.

I also get a different md5 ( I use md5deep:
http://md5deep.sourceforge.net/) and different size:

D:\Documents\gnubg-build\distribs>md5deep.exe -z gnubg-MAIN-20100220-setup.exe

25130031  d78184ad7a81c3bc9668d47866a4a347
D:\Documents\gnubg-build\distribs\gnubg-MAIN-20100220-setup.exe

(first number is size).

In file properties I see:

Size:  23.9 MB (25.130.031 bytes)
Size on disk:23.9 MB (25.133.056 bytes)

MaX.

On 2 March 2010 10:29, motiv4u  wrote:
> The downloaded new install (9:27 GMT) doesn't have the met either.
> The size of the exe: 23,9 MB (25.128.960 bytes)
> MD5: e9fd83d962a7f0eec616e14d575f029e
> CRC: 9C79E7A4
>
> N.
>
> On Tue, Mar 2, 2010 at 9:54 AM, Massimiliano Maini  wrote:
>> Doh, dumb me !
>>
>> Please download it once again and reinstall, I fixed it (9am GMT on 2nd 
>> March).
>>
>> MaX.
>>
>> On 1 March 2010 17:12, Neil Robins  wrote:
>>>
>>> I have downloaded it but am not finding the new MET.
>>>
>>>
>>> Massimiliano Maini-3 wrote:
>>>>
>>>> Hi all,
>>>>
>>>> the new build has the latest code (even if the timestamp is not
>>>> themost recent one).
>>>>
>>>> Most important, it contains a new MET 9Rockwell-Kazaross) obtained
>>>> with gnubg supremo rollouts up to 15pt scores.
>>>>
>>>> @Achim/Christian: the cvs snapshots are down, most likely because of
>>>> lack of space.
>>>> I've deleted some useless snapshots.
>>>>
>>>> MaX.
>>>>
>>>>
>>>> ___
>>>> Bug-gnubg mailing list
>>>> Bug-gnubg@gnu.org
>>>> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>>>>
>>>>
>>>
>>> --
>>> View this message in context: 
>>> http://old.nabble.com/New-win-builds-on-gnubg.org-%2820100220-code%29-tp27744962p27745538.html
>>> Sent from the Gnu - Backgammon mailing list archive at Nabble.com.
>>>
>>>
>>>
>>> ___
>>> Bug-gnubg mailing list
>>> Bug-gnubg@gnu.org
>>> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>>>
>>
>>
>> ___
>> Bug-gnubg mailing list
>> Bug-gnubg@gnu.org
>> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>>
>
>
>
> --
> "He is no fool who gives up what he cannot keep to gain what he cannot
> lose." -Jim Elliot
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] New win builds on gnubg.org (20100220 code)

[Massimiliano Maini]

Doh, dumb me !

Please download it once again and reinstall, I fixed it (9am GMT on 2nd March).

MaX.

On 1 March 2010 17:12, Neil Robins  wrote:
>
> I have downloaded it but am not finding the new MET.
>
>
> Massimiliano Maini-3 wrote:
>>
>> Hi all,
>>
>> the new build has the latest code (even if the timestamp is not
>> themost recent one).
>>
>> Most important, it contains a new MET 9Rockwell-Kazaross) obtained
>> with gnubg supremo rollouts up to 15pt scores.
>>
>> @Achim/Christian: the cvs snapshots are down, most likely because of
>> lack of space.
>> I've deleted some useless snapshots.
>>
>> MaX.
>>
>>
>> ___
>> Bug-gnubg mailing list
>> Bug-gnubg@gnu.org
>> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>>
>>
>
> --
> View this message in context: 
> http://old.nabble.com/New-win-builds-on-gnubg.org-%2820100220-code%29-tp27744962p27745538.html
> Sent from the Gnu - Backgammon mailing list archive at Nabble.com.
>
>
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] New win builds on gnubg.org (20100220 code)

[Massimiliano Maini]

Hi all,

the new build has the latest code (even if the timestamp is not
themost recent one).

Most important, it contains a new MET 9Rockwell-Kazaross) obtained
with gnubg supremo rollouts up to 15pt scores.

@Achim/Christian: the cvs snapshots are down, most likely because of
lack of space.
I've deleted some useless snapshots.

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] New win builds on gnubg.org (20091229 code) - bug ?

[Massimiliano Maini]

Compiled with latest code (20091230) and modified config.h as yuo
explained. Thanks !!

Distrib available on gnubg.org.

MaX..

2009/12/30 Michael Petch :
>
> Another important thing to add to your windows config.h is as follows .
>
> On Autoconf systems a define is created called PACKAGE_BUGREPORT (this comes
> from configure.in during the autoconf process) . This define contains the
> email address for bug reports. I now use this define in the www.random.org
> code for the user agent. So you will need to add a line to config.h for non
> autoconf systems (windows/mingw) that reads:
>
> #define PACKAGE_BUGREPORT “bug-gn...@gnu.org”
>
> This should get Max through his next compile (cross fingers)


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] New win builds on gnubg.org (20091229 code) - bug ?

[Massimiliano Maini]

I get an error compiling:

gnubg.c: In function 'getDiceRandomDotOrg':
gnubg.c:4473: error: expected ',' or ';' before 'PACKAGE_VERSION'
mingw32-make: *** [obj/gnubg.o] Error 1

MaX.

2009/12/29 Michael Petch :
>
> Should be fixed now (buffer overrun). I have also added a user agent string
> per the request of random.org, and done some cleanup.
>
> On 29/12/09 7:08 AM, "Massimiliano Maini"  wrote:
>
>> Hi a new win distrib is avauialble on gnubg.org (20091229).
>>
>> It includes recent fixes, including the one for random.org.
>>
>> I have however a problem with that: as soon as I select random.org, 
>> everything
>> works fine but I can no longer quit gnubg (alt-f4, ctrl-q or the red
>> cross button
>> on the window).
>>
>> Strange ...
>>
>> MaX.
>>
>>
>> ___
>> Bug-gnubg mailing list
>> Bug-gnubg@gnu.org
>> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>>
>
>
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] New win builds on gnubg.org (20091229 code) - bug ?

[Massimiliano Maini]

Hi a new win distrib is avauialble on gnubg.org (20091229).

It includes recent fixes, including the one for random.org.

I have however a problem with that: as soon as I select random.org, everything
works fine but I can no longer quit gnubg (alt-f4, ctrl-q or the red
cross button
on the window).

Strange ...

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] i have a new feature request

[Massimiliano Maini]

Hi, what would you use this for ?

MaX.

2009/11/27 Khodam Inja 
>
> hello to gnubg developers, could u pls add the number of double rolls to the 
> checker play tab of match and game statistics? ty for your attention bye now


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Re: Rollout jsd, statsig etc. [LONG]

[Massimiliano Maini]

Hmm maybe it's just too complicate for me.

It seems to me that the vectors you update at each iteration contain
the probability distribution function of the "equity" of each play, somehow
estimated from rollout results up to the current trial. At the end, you
just integrate on the multivariate tail to compute "your" probability.

If the above is true, on top of assuming a uniform prior, you assume
that the real pdf are the ones you have on your last step.
This is not far from assuming that the real equities are the estimated
ones. Duh, I must be missing something ...

MaX.

2009/11/17 Timothy Y. Chow :
> On Tue, 17 Nov 2009, Massimiliano Maini wrote:
>> 2009/11/16 Timothy Y. Chow :
>> > The multivariate tail probability, for
>> > example, tells you only the probability that some strange event will occur
>> > *under the assumption that the equities are equal to the estimated
>> > equities*.  This is *not* the same as *the probability that the true
>> > equities are different from their estimated values*.
>>
>> Just for my understanding, the bayesian approach would be no different
>> with respect to that, right ?
>>
>> It will give you a probability that some strange event will occur under
>> the assumption that the estimated pdf are the ones you have on the
>> current trial/iteration (plus the initial assumption on those pdf),
>> right ?
>
> No.  Under the Bayesian approach, the reported probability that (for
> example) Play A is the best is the probability that Play A is the best,
> *assuming* that the equities of the various plays were all chosen
> uniformly at random before you saw any rollout results.  (I'm assuming you
> pick a uniform prior.)
>
> Here, the "strange event" you're measuring is, "the equity really was
> something very different from the estimated equity, yet somehow we managed
> to see these very non-representative results."  It should be clear that
> measuring this event is more meaningful than *assuming* that the true
> equity is the estimated equity, and computing the probability that you
> will get misleading results in the future.  If you're going to *assume*
> that the true equity is the estimated equity, why would you roll things
> out further?  What you're interested in is the possibility that the true
> equity is *not* the estimated equity, so you shouldn't *assume* that
> they're equal.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Re: Rollout jsd, statsig etc. [LONG]

[Massimiliano Maini]

2009/11/16 Timothy Y. Chow :
> The multivariate tail probability, for
> example, tells you only the probability that some strange event will occur
> *under the assumption that the equities are equal to the estimated
> equities*.  This is *not* the same as *the probability that the true
> equities are different from their estimated values*.

Just for my understanding, the bayesian approach would be no different
with respect to that, right ?

It will give you a probability that some strange event will occur under the
assumption that the estimated pdf are the ones you have on the
current trial/iteration (plus the initial assumption on those pdf), right ?

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Re: Rollout jsd, statsig etc. [LONG]

[Massimiliano Maini]

Tim, the point you made is very clear. In my original post I
mentionned myself that in the "usual" interpretation there's some
inaccuracy. Also, I said that your method is (quote) "by far the right
conceptual way to answer the question".

That said, it seems that under usual conditions the multiple measures
are very close and any of these is equally good for the goal: indicate
if a rollout can be trusted or needs to be extended.

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] Rollout jsd, statsig etc. [LONG]

[Massimiliano Maini]

Hi all,

there has been a looong thread on bgonline (seems to be down right
now) about the
right manner to present rollout results.

Gnubg shows the so called "JSDs": the idea behind is that for a JSD of
1.96 between
two plays there's 95% chance for the play with higher rollout equity
being really the
one with higher equity (there some inaccuracy in the above sentence,
but the concept
is there).

Remark #1:
Why don't we show the % instead of the JSD ? It's much more reasonable.
This will mean that at the end of a rollout, we'll have something like

Play 1+0.650   95.5%
Play 2+0.630 4.5%
Play 3+0.600 2.5%

Notice that the percentage shown aside the top play is the
"confidence"we have in it
being better than the 2nd best play. For the other plays, the
percentage shown is the
confidence we have in that play being better than the top play.
Percentages for play 1
and 2 must add to 100%.

Showing the result this was looks nicer and more easily understandable.
Also, we should allow to enter percentages intead of JSDs as stopping criteria.


Remark #2:
Confidence intervals of play1 vs playN does nt give you the right picture: most
of the time what you want to know is the chances of any play being
better than *all* the
others.

One easy approach is this: for each play you have an equity and a
stdev. You assume
the true equity of each play to have a normal distribution and you do
Nmc MonteCarlo
trials. For each trial, you sample the N normal distributions and you
find the highest value.
You add 1 to the total of the play with the highest value.

At the end, you divide each  play total by Nmc and this gives you the
chances of each
play being best.

The above is just MonteCarlo integration of multiple normal distributions.

Only things you need are erf (error function, part of glib) and its
inverse (not part of glib
but part, for example, of Gnu Scientific library, or you can find many
decent C/C++
implementations around).

50K montecarlo trials are enough, depending on how fast it is, you can do more.

It would be very easy to integrate this in gnubg and to compute the
values at the end
of the rollout (and show them aside the confidence interval ones).

More ambitious (and probably unnecessary) would be to do the MC stuff
at the end of
each trial and use its results as stopping criteria for the rollout.
But this may slow
down rollouts (TBC, see remark #4 below).


Remark #3:
Another criteria that could be used to stop the rollout is the product
of all the confidence
of play1 vs playN. In my example above, play1 better than play2 with
confidence 95.5%,
play1 vs play3 confidence (100-2.5) = 97.5%, hence overall confidnce
of play 1 being best
= 95.5% * 97.5% = 93.1%.

This value is in fact a lower bound (it assumes that the events 'p1
better than p2" and "p1
better than p3" are uncorrelated, which is wrong). Combined with the
confidence of play1 vs
play2 (which is an upper bound), they give a reasonable interval of
the true probability at
almost no effort.


Remark #4:
Tim Chow has proposed a bayesian approach which seems by far the right
conceptual way
to answer the question "how sure are we that play1 is best". However,
this methods too
slow downs rollouts (you have to execute steps at each trial) and does
not provide any
major advantage over the montecarlo method (which can be executed only
once at the
end of the rollout, if not used as stopping criteria).
In my *extremely* naive python implementation however, the bayes
method was much faster
than the MC method (executed at each trial).


So, I would say that we can easily implement points #1 and #2: let's
show % instead of jsds
and let's add a final MC computation of "true" probabilities.


MaX.

P.S.
I have some python code (ugly, but working) that:

- simulates a Nt trials rollout for an arbitrary number of plays (of
which you precise
the equity)
- runs the computations in all 3 flavors; the lower/upper bound based
on jsds/conf intervals,
the montecarlo stuff (at the end only or during the rollout trials, 2
flavor), the Bayesian stuff


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] Re: Compiling latest code

[Massimiliano Maini]

It was me not awake :)

I took the snapshot from 11/08 instead of 11/09. Latest seems to compile fine.

MaX.

2009/11/9 Massimiliano Maini :
> Maybe it's me not fully awake, but compiling the latest code I get
> some strange errors:
>
> Compiling cache.c
> In file included from ../backgammon.h:25,
>                 from ../multithread.h:16,
>                 from cache.c:33:
> ../analysis.h:154: error: expected declaration specifiers or '...'
> before 'matchstate'
> In file included from ../backgammon.h:27,
>                 from ../multithread.h:16,
>                 from cache.c:33:
> ../rollout.h:160: error: expected declaration specifiers or '...'
> before 'BasicCubefulRollout'
> ../rollout.h:160: warning: data definition has no type or storage class
> ../rollout.h:160: warning: type defaults to 'int' in declaration of
> 'EXP_LOCK_FUN'
> In file included from ../multithread.h:16,
>                 from cache.c:33:
> ../backgammon.h:112: error: expected specifier-qualifier-list before 'CMark'
> ../backgammon.h:213: error: expected specifier-qualifier-list before
> 'matchstate'
> ../backgammon.h:356: error: expected '=', ',', ';', 'asm' or
> '__attribute__' before 'ms'
> ../backgammon.h:406: error: expected declaration specifiers or '...'
> before 'va_list'
> ../backgammon.h:414: error: expected declaration specifiers or '...'
> before 'va_list'
> In file included from ../multithread.h:16,
>                 from cache.c:33:
> ../backgammon.h:435: warning: type defaults to 'int' in declaration of
> 'matchstate'
> ../backgammon.h:435: error: expected ';', ',' or ')' before '*' token
> ../backgammon.h:441: warning: type defaults to 'int' in declaration of
> 'matchstate'
> ../backgammon.h:441: error: expected ';', ',' or ')' before '*' token
> ../backgammon.h:452: error: expected declaration specifiers or '...'
> before 'matchstate'
> ../backgammon.h:481: error: expected ')' before '*' token
> ../backgammon.h:491: error: expected ')' before '*' token
> ../backgammon.h:493: warning: type defaults to 'int' in declaration of
> 'matchstate'
> ../backgammon.h:493: error: expected ';', ',' or ')' before '*' token
> ../backgammon.h:1003: warning: type defaults to 'int' in declaration
> of 'matchstate'
> ../backgammon.h:1003: error: expected ';', ',' or ')' before '*' token
> In file included from cache.c:33:
> ../multithread.h:53: error: expected specifier-qualifier-list before
> 'matchstate'
>
> MaX.
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] Compiling latest code

[Massimiliano Maini]

Maybe it's me not fully awake, but compiling the latest code I get
some strange errors:

Compiling cache.c
In file included from ../backgammon.h:25,
 from ../multithread.h:16,
 from cache.c:33:
../analysis.h:154: error: expected declaration specifiers or '...'
before 'matchstate'
In file included from ../backgammon.h:27,
 from ../multithread.h:16,
 from cache.c:33:
../rollout.h:160: error: expected declaration specifiers or '...'
before 'BasicCubefulRollout'
../rollout.h:160: warning: data definition has no type or storage class
../rollout.h:160: warning: type defaults to 'int' in declaration of
'EXP_LOCK_FUN'
In file included from ../multithread.h:16,
 from cache.c:33:
../backgammon.h:112: error: expected specifier-qualifier-list before 'CMark'
../backgammon.h:213: error: expected specifier-qualifier-list before
'matchstate'
../backgammon.h:356: error: expected '=', ',', ';', 'asm' or
'__attribute__' before 'ms'
../backgammon.h:406: error: expected declaration specifiers or '...'
before 'va_list'
../backgammon.h:414: error: expected declaration specifiers or '...'
before 'va_list'
In file included from ../multithread.h:16,
 from cache.c:33:
../backgammon.h:435: warning: type defaults to 'int' in declaration of
'matchstate'
../backgammon.h:435: error: expected ';', ',' or ')' before '*' token
../backgammon.h:441: warning: type defaults to 'int' in declaration of
'matchstate'
../backgammon.h:441: error: expected ';', ',' or ')' before '*' token
../backgammon.h:452: error: expected declaration specifiers or '...'
before 'matchstate'
../backgammon.h:481: error: expected ')' before '*' token
../backgammon.h:491: error: expected ')' before '*' token
../backgammon.h:493: warning: type defaults to 'int' in declaration of
'matchstate'
../backgammon.h:493: error: expected ';', ',' or ')' before '*' token
../backgammon.h:1003: warning: type defaults to 'int' in declaration
of 'matchstate'
../backgammon.h:1003: error: expected ';', ',' or ')' before '*' token
In file included from cache.c:33:
../multithread.h:53: error: expected specifier-qualifier-list before
'matchstate'

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] New win builds on gnubg.org (20091102 code) - EOM

[Massimiliano Maini]

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Rollout with 'Stop when result is accurate'

[Massimiliano Maini]

I tend to agree. The two criteria should be:

1. std dev smaller than a given (absolute) value

2. jsd smaller than a given value

I would also allow to combine them instead of allowing only to choose
between the 2.

For example:

max trials = 1296
min trials = 324
stop when jsd <= 2 AND sd <= 0.010

It may make sense to allow to select an OR between the two criteria.

MaX.

2009/10/31 Philippe Michel :
> On Sat, 31 Oct 2009, Yoshito Takeuchi wrote:
>
> PS: Jon, Christian, how about applying this change (or something to that
> effect), plus the corresponding update the UI ? I insist : the current
> behaviour of this criterion is just useless!
>
>
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] Bot comparison on BGonline - preliminary results

[Massimiliano Maini]

Here:

http://www.bgonline.org/forums/webbbs_config.pl?noframes;read=52114

The pattern of GnuBG 2/3/4 ply errors on wrong doubles/takes/passes is
just strange.
See my comment there:

http://www.bgonline.org/forums/webbbs_config.pl?noframes;read=52269

MaX.

P.S.
I think there are notes from Joseph about gnubg training somewhere on the net:
anyone has the address ?


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Position/Match id moved to status bar

[Massimiliano Maini]

Much better.
Needs just some space between the end of the leftmost text area
and the 'GnubgID' label. maybe also arounf the frame ...

Purely cosmetic.

MaX.

2009/10/15 Jonathan Kinsey :
> I've put a frame around the id, looks slightly better maybe?
>
> Jon


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] New win builds on gnubg.org (20091016 code) - EOM

[Massimiliano Maini]

Can you give the exact steps to reproduce ?
It seems to work for me ...

MaX.

2009/10/16 Yoshito Takeuchi :
> Hi all.
>
> I tried 20091016 code, and Ctrl-C that is Gnubg id to clipboard.
> And paste result is
>
> Position ID: AA
> Match ID: cAgA
>
> This is umlimited game, initial position.
>
> Obvious 'Position ID:' is wrong.
>
> My system is Windows 7 RC 64bit.
>
> Yoshito Takecuhi
>
>
>
> 2009/10/16 Massimiliano Maini :
>>
>>
>>
>> ___
>> Bug-gnubg mailing list
>> Bug-gnubg@gnu.org
>> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>>
>>
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] New win builds on gnubg.org (20091016 code) - EOM

[Massimiliano Maini]

___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: FW: [Bug-gnubg] Position/Match id moved to status bar

[Massimiliano Maini]

Hi Ian,

this is still possible, probably it's even better than before: no need
to click in the text field (which no longer exists), just do a crtl-v.

MaX.

2009/10/14 Ian Shaw :
>
>
> I fairly often cut and paste IDs from websites into my gnubg. I would
> like to be able to continue this practive.
>
> -- Ian


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Position/Match id moved to status bar

[Massimiliano Maini]

That's what i meant saying "it should stand out" a bit more.
Not editable, probably true: it's most likely to be faster to
build the actual position clicking than entering manually the ID.

MaX.

2009/10/14 Christian Anthon :
> I like the placement too. I would probably prefer it to be a text box
> rather than a label. Doesn't have to be editable.
>
> Christian.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Position/Match id moved to status bar

[Massimiliano Maini]

Hi Jon,
2 minor warnings:

gtkgame.c: In function 'PasteIDs':
gtkgame.c:1177: warning: unused variable 'bd'
gtkgame.c: At top level:
gtkgame.c:1164: warning: 'GetFocusedWidget' defined but not used

I like the ctrl-C/ctrl-V thing: I guess many will be surprised by the change
but I think this is how it should work.

I would just make the gnubgID stand out a bit more in the GUI:

- put it to the left of the status bar
- highlight it somehow (white background, different color, whatever)

MaX.

2009/10/13 Jonathan Kinsey :
> I've moved the id field into the status bar and changed ctrl+c/ctrl+v to act
> directly on the id, you can also right click on id and copy/paste from
> there.
> The obvious advantage is to make the playing area bigger and remove the
> slightly
> unnecessary box at the top of the board.
>
> Any thoughts?
>
> Jon
>
>
>
> 
> Have more than one Hotmail account? Link them together to easily access
> both.
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] New win builds on gnubg.org (20091013 code)

[Massimiliano Maini]

Including latest changes to Analysis and Players dialogs.

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] possible degrade with change set "protect movelist during autosave"

[Massimiliano Maini]

I've rebuilt including the fix and FTPed to gnubg.org.

If you have downoloadad the last 20091009 install/exes before now
(2:33 PM GMT), please redownload it.

MaX.

2009/10/9 Christian Anthon :
> This was just a mistake in the bug fix. Reversed the source and the
> destination in the movelist copy.
>
> Christian.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] New win builds on gnubg.org (20091009 code)

[Massimiliano Maini]

Includes the fix for the 'add the same match twice to the dB" problem.

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Feature request: Training tool "Randomize first N moves within 0.XXX equity."

[Massimiliano Maini]

For chequer play, I can see:

- favor the move than wins more gammons (when gammons count, maybe)

and/or (maybe the two overlap):

- favor/disfavor the move that hits (or the move that blots)

For cube action:

- cube aggressive/cube shy (double close no-doubles and take close passes /
no-double close doubles, pass close takes)

For each "presonality aspect", multiple levels could be defined, e.g.:

- gammon-go attitude: disfavor a lot / disfavor a bit / neutral /
favor a bit / favor a lot
- gammon-save attitude: disfavor a lot / disfavor a bit / neutral /
favor a bit / favor a lot
- double/nodouble attitude: very shy / shy / neutral / aggressive /
very aggressive
- take/pass attitude: very shy / shy / neutral / aggressive / very aggressive

Another thing that could be interesting is to avoid applying the "presonality"
all the time, on all the decisions: an "early doubler" will double
early randomly
in the match (e.g. with a given probability like 33% of the time, or something).

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Feature request: Training tool "Randomize first N moves within 0.XXX equity."

[Massimiliano Maini]

The "personalities" thing would be very interesting IMO:

- a small set of predefined personalities (eventually, ability to
define custom ones, on chequer play and cube decision)

- possibility to chose a given personality or to pick a random
one (not disclosed, unless requested).

MaX.

2009/10/8 Jonathan Kinsey :
> This reminds me of the discussion on "personalities", as the kind of
> approach
> you suggest could easily fit into the post-evaluation equity adjustments
> that
> are an easy way to add varied gnu-opponents.
>
> Jon
>
> Michael Depreli wrote:
>> I'm sure many users use gnubg simply to play against as self training.
>> One issue that for me doesn't help with training is that gnubg will
>> obviously always play the same opening move or reply etc in the same
>> given position.
>> In real life backgammon obviously this isn't the case.
>> As a simple example for money gnubg plays 52 13/8 13/11.
>> Rollouts show this to be an error, and I'm guessing most humans would
>> not play this move the majority of the time.
>> Also rollouts have shown many opening plays and replies are very close
>> in equity so as not be be considered an error.
>>
>> My request therefore is to have an option to randomize the first N moves
>> within equity X.
>>
>> I would suggest something like a move filter:
>>
>> E.g. Randomize first N moves within X plus additional N moves within Y.
>>
>> Hopefully I'm not the only user that would find this feature useful.
>>
>>
>> Michael
>>
>>
>>
>>
>>
>> 
>> Have more than one Hotmail account? Link them together to easily access
>> both.
>>
>>
>> 
>>
>> ___
>> Bug-gnubg mailing list
>> Bug-gnubg@gnu.org
>> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>
>
>
>
> 
> New! Receive and respond to mail from other email accounts from within
> Hotmail Find out how.
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] New win builds on gnubg.org (20090928 code)

[Massimiliano Maini]

Includes recent bug fix for movelist with autosave/cmark stuff.

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] 2 bugs and a question about database file location

[Massimiliano Maini]

2009/10/7 Michael Petch :
>
> The GnuBG that you download does not "learn from experience" (see pevious
> posts to Roy Crabtree on that subject).

Actually it is probably better if you just trust Michael/us on the subject and
you DON'T SEE the post from Roy ... :)

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Line-terminal mode

[Massimiliano Maini]

In command line, you have the option "set automatic game on/off".
Once current games end, you can issue your commands and then
use "new game" to start a new game in the session.

In the GUI you have the option: Settings/Options/Start new games
immediately If you disable the option,  then you have no other mean
to start a new game than using a "new game" command: there's no
way to start a new game by clicking on something (which is bad IMO).

MaX.

2009/10/4 Timothy Y. Chow :
> Occasionally I'm at a library or something where there's a public computer
> from which I can ssh to my home machine to use gnubg in line-terminal
> mode, but cannot bring up the GUI.
>
> I want to be able to play a money-game session and save all the games.
> Better still would be to analyze the session (say at 2-ply) and view/save
> the results of *that*, but I'd settle for saving the results and analyzing
> them later.  I can't figure out how to do that.  Part of the problem seems
> to be that as soon as a money game ends, gnubg starts up a new game
> immediately, with no opportunity to execute a command between the end of a
> game and the start of the next game.  If I type "save game" after the new
> game has started, it saves the new game and not the one that just
> finished.
>
> Tim
>
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Autosave questions

[Massimiliano Maini]

This is how I would make it:

1. If autosaving from a loaded file (eg .sgf), use something like
.0001.autosave.sgf, in the same dir. The 0001 part of the file
name can be used to ensure the file is unique (proper to the current
instance of gnubg). If .0001.autosave.sgf already exists, use
.0002.autosave.sgf: the higher the number, the most recent
the autosave (I means across different instances of gnubg or different
runs of rollouts with autosave).

2. If autosaving without a filename, prompt for one.

3. At the end of the rollout(s), the autosaved file could be deleted
since it is (normally) no longer necessary. Not seeing an autosave
file implies everything went OK. Maybe we could make this optional,
with an option like "Save results once rollouts are done" when
doing a CMark rollout.

MaX.

P.S.
The Analyse/Rollout (I mean the 2nd one, the one for Cmarked moves)
should probably be renamed to Rollout CMarked (or whatever we chose
instead of CMarked).


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] New win builds on gnubg.org (20091002 code)

[Massimiliano Maini]

Including the rollout bug fix.

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Forward from BGO - Problem with rollouts on XP (20091001 build)

[Massimiliano Maini]

Here it is (from 20091002 code).

I start a match, wait for my turn, before rolling click on analysis/rollout,
get the dialog with the strange message, click on OK, crash.


MaX.

D:\Documents\gnubg>gdb gnubg-gui-py-sse2-wmt-O3.exe-debug.exe
GNU gdb 6.8
Copyright (C) 2008 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later 
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "i686-pc-mingw32"...
(gdb) run
Starting program: D:\Documents\gnubg/gnubg-gui-py-sse2-wmt-O3.exe-debug.exe
[New thread 1384.0xba4]
GNU Backgammon 0.90-mingw 20090902
Copyright (C) 1999, 2000, 2001, 2002, 2003, 2004 by Gary Wong.
Copyright (C) 2009 by Gary Wong and the AUTHORS; for details type
`show version'.
This program comes with ABSOLUTELY NO WARRANTY; for details type `show
warranty'.
This is free software, and you are welcome to redistribute it under
certain conditions; type `show copying' for details.
[New thread 1384.0xf68]
[New thread 1384.0x738]

Program received signal SIGSEGV, Segmentation fault.
0x617d4d79 in gtk_container_add () from
D:\Documents\gnubg\libgtk-win32-2.0-0.dll
(gdb) thread apply all bt

Thread 3 (thread 1384.0x738):
#0  0x7c90e514 in ntdll!LdrAccessResource () from C:\WINDOWS\system32\ntdll.dll
#1  0x7c90df5a in ntdll!ZwWaitForSingleObject () from
C:\WINDOWS\system32\ntdll.dll
#2  0x7c8025cb in WaitForSingleObjectEx () from C:\WINDOWS\system32\kernel32.dll
#3  0x0790 in ?? ()
#4  0x in ?? ()

Thread 2 (thread 1384.0xf68):
#0  0x7c90e514 in ntdll!LdrAccessResource () from C:\WINDOWS\system32\ntdll.dll
#1  0x7c90df5a in ntdll!ZwWaitForSingleObject () from
C:\WINDOWS\system32\ntdll.dll
#2  0x7c8025cb in WaitForSingleObjectEx () from C:\WINDOWS\system32\kernel32.dll
#3  0x0790 in ?? ()
#4  0x in ?? ()

Thread 1 (thread 1384.0xba4):
#0  0x617d4d79 in gtk_container_add () from
D:\Documents\gnubg\libgtk-win32-2.0-0.dll
#1  0x0045766a in RolloutProgressStart (pci=0x22e064, n=1, aars=0x0,
prc=0x54cce0, asz=0x22e09c, multiple=0, pp=0x22e0fc) at progress.c:760
#2  0x005098c3 in CommandRollout (sz=0xe31f9ff "") at gnubg.c:2530
#3  0x0050ce61 in HandleCommand (sz=0xe31f9ff "", ac=0x5527a0) at gnubg.c:1163
#4  0x005125b4 in UserCommand (szCommand=0x589d34 "rollout") at gnubg.c:3588
#5  0x004ab53b in Command (p=0x0, iCommand=45, widget=0xcb0b680) at
gtkgame.c:473
#6  0x619e1618 in gtk_item_factory_callback_marshal () from
D:\Documents\gnubg\libgtk-win32-2.0-0.dll
#7  0x63a44074 in g_closure_invoke () from
D:\Documents\gnubg\libgobject-2.0-0.dll
#8  0x63a576d9 in signal_emit_unlocked_R () from
D:\Documents\gnubg\libgobject-2.0-0.dll
#9  0x63a588f3 in g_signal_emit_valist () from
D:\Documents\gnubg\libgobject-2.0-0.dll
#10 0x63a58e26 in g_signal_emit () from D:\Documents\gnubg\libgobject-2.0-0.dll
#11 0x6198c4c0 in gtk_widget_activate () from
D:\Documents\gnubg\libgtk-win32-2.0-0.dll
#12 0x6187977a in gtk_menu_shell_activate_item () from
D:\Documents\gnubg\libgtk-win32-2.0-0.dll
#13 0x6187b4a7 in gtk_menu_shell_button_release () from
D:\Documents\gnubg\libgtk-win32-2.0-0.dll
#14 0x61868e63 in _gtk_marshal_BOOLEAN__BOXED () from
D:\Documents\gnubg\libgtk-win32-2.0-0.dll
#15 0x63a44074 in g_closure_invoke () from
D:\Documents\gnubg\libgobject-2.0-0.dll
#16 0x63a573ec in signal_emit_unlocked_R () from
D:\Documents\gnubg\libgobject-2.0-0.dll
#17 0x63a5878b in g_signal_emit_valist () from
D:\Documents\gnubg\libgobject-2.0-0.dll
#18 0x63a58e26 in g_signal_emit () from D:\Documents\gnubg\libgobject-2.0-0.dll
#19 0x61985ec1 in gtk_widget_event_internal () from
D:\Documents\gnubg\libgtk-win32-2.0-0.dll
#20 0x61860e82 in gtk_propagate_event () from
D:\Documents\gnubg\libgtk-win32-2.0-0.dll
#21 0x61861fec in gtk_main_do_event () from
D:\Documents\gnubg\libgtk-win32-2.0-0.dll
#22 0x6c37014a in gdk_event_dispatch () from
D:\Documents\gnubg\libgdk-win32-2.0-0.dll
#23 0x685ea27b in g_main_context_dispatch () from
D:\Documents\gnubg\libglib-2.0-0.dll
#24 0x685ed185 in g_main_context_iterate () from
D:\Documents\gnubg\libglib-2.0-0.dll
#25 0x685ed574 in g_main_loop_run () from D:\Documents\gnubg\libglib-2.0-0.dll
#26 0x618625dc in gtk_main () from D:\Documents\gnubg\libgtk-win32-2.0-0.dll
#27 0x004bf835 in RunGTK (pwSplash=0x0, commands=0x0,
python_script=0x0, match=0x0) at gtkgame.c:3593
#28 0x00511d66 in main (argc=1, argv=0x3e5410) at gnubg.c:4816
(gdb)


2009/10/2 Michael Petch :
> I just tried it. I took a random position (tried this twice on 2 different
> ones) that were previously analyzed. selected the top 3 moves from the
> analysis pane and then hit the rollout button. I don't have debug info but
> it seems to crash libgtk-win32-2.0-0.dll (I know that's not really helpful)
>
> On 02/10/09 12:54 AM, "Christian Anthon"  wrote:
>
>>
>> More importantly, can anyone confirm this?
>>
>> Christian.
>>
>
>
>
>
> 

Re: [Bug-gnubg] Forward from BGO - Problem with rollouts on XP (20091001 build)

[Massimiliano Maini]

I can produce a crash this way:

- start a match, when it's your turn, before rolling, click on Analysis/Rollout
- I get a pop up with the following message:

"This action will roll out the current position before the dice are
rolled. It is
not possible to extend nor save this rollout ?"

And just and OK button (plus the "Don't show this again" check box)

First, the message is in poor english and unclear.
Second, there should probably be also a cancel button.
And third, clicking on OK it crashes.
It also crashes clicking on the top-right cross button to close the
pop up window.

Also strange, the message and the crash also happens when you click on
analysis/rollout *after* having rolled the dice.

Clicking on rollout from analysis panel crashes too.

MaX.

2009/10/2 Christian Anthon :
> Michael Petch wrote:
>> This was posted to BGO by W Womack:
>>
>> "Anyone else having a problem with build 20091001 crashing when you try to
>> do a rollout? XP machine, worked fine with previous builds."
>>
>> Has anyone seen this?
>>
>>
>
> More importantly, can anyone confirm this?
>
> Christian.
>
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] New win builds on gnubg.org (20091001 code)

[Massimiliano Maini]

I messed up a bit: the code was built from 20090930 snapshot,since the 20091001
one is not available (Achim ? space problem ? process down ?).

Rectified everything. Thx.

MaX.

2009/10/1 motiv4u :
> PDF is solved
> You just missed the latest added 3D design (PlayOK)
> On the website, the Snapshot date is the previous one: 20090928
>
> Thank you
>
> N.
>
> On Thu, Oct 1, 2009 at 11:02 AM, Massimiliano Maini  
> wrote:
>> Should solve some PDF/locale issues.
>> New button to stop cmark rollouts.
>> Improved 3d core (for poor OpenGL drivers).
>>
>> MaX.
>>
>>
>> ___
>> Bug-gnubg mailing list
>> Bug-gnubg@gnu.org
>> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>>
>
>
>
> --
> "He is no fool who gives up what he cannot keep to gain what he cannot
> lose." -Jim Elliot
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Any chance of Previous Marked and Cmarked buttons?

[Massimiliano Maini]

I'm for it: have prev buttons and no labels for toolbar buttons, as
far as we have the tooltips.

MaX.

2009/10/1 Christian Anthon :
> On Sun, Sep 27, 2009 at 1:18 PM, Michael Depreli
>  wrote:
>> Sorry to be difficult :-)  but is there any chance of Previous Marked and
>> CMarked buttons?
>>
>
> Without labels on the buttons you could. Perhaps the best solution is
> to have no labels on the navigation buttons...
>
>> Also I think it would be nice if you could see which moves have been CMarked
>> in the move list.
>>
>> I was wondering if putting a border around the move would work?
>>
>> Hopefully it wouldn't make the move list look too busy.
>>
>
> I'll try and see how it works.
>
> Christian.
>
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] New win builds on gnubg.org (20091001 code)

[Massimiliano Maini]

Should solve some PDF/locale issues.
New button to stop cmark rollouts.
Improved 3d core (for poor OpenGL drivers).

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] New win builds on gnubg.org (20090928 code)

[Massimiliano Maini]

Hmmm ... running dependency walker (http://www.dependencywalker.com/)
on my Vista PC (where PDF exports misses the numbers) seems to indicate
that gnubg.exe uses LIBPANGOCAIRO-1.0-0.DLL which in turns uses
LIBPANGOFT2-1.0-0.DLL which in turn uses FREETYPE6.DLL (all from
gnubg's install dir).

I'll check on my office PC (XP, PDF export works, I see the numbers).

MaX.

2009/9/29 Jonathan Kinsey :
> When I run here gnubg doesn't use the libpangoft2 dll, the freetype6.dll is
> being used though.
>
> Jon


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] New win builds on gnubg.org (20090928 code)

[Massimiliano Maini]

Running CLI version I got this error message:

(gnubg-cli.exe:5168): Pango-WARNING **: couldn't load font "Sans 8
Not-Rotated 0", falling back to "
Sans Not-Rotated 0", expect ugly output.
>e
(gnubg-cli.exe:5168): Pango-WARNING **: failed to create cairo scaled
font, expect ugly output. the
offending font is 'Tahoma 0'

(gnubg-cli.exe:5168): Pango-WARNING **: font_font status is: 

(gnubg-cli.exe:5168): Pango-WARNING **: scaled_font status is: out of memory
(MaX)

What's missing ?

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] New win builds on gnubg.org (20090928 code)

[Massimiliano Maini]

Hmmm  on my home PC I donùt see the number's in Nardy's PDF
and I don't see them neither when I export to a PDF.

Guess i'm missing something in the install archive, no idea what ...

MaX.

2009/9/29 motiv4u :
> On Mon, Sep 28, 2009 at 12:08 PM, Christian Anthon
>  wrote:
>> Could be a missing font. I'm too busy to look at it right now, but try
>> to exchange pdf's and see if and where they work.
>
> PDF file with missing numbers attached
>
> N.
>
>>
>> Christian.
>>
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] New win builds on gnubg.org (20090928 code)

[Massimiliano Maini]

I see all of them (1-24, numbers for more than 5 chequers, cube value)
when I export to PDF.

Could be a missing font issue, but I have no idea where to look ...

MaX.

2009/9/28 motiv4u :
> The export to PDF also works in this one, but the numbers 1-24, the
> value of the cube and the number if more than 5 checkers are still
> omitted. That is: there are empty cases.
>
> N.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] Re: GNU Bug

[Massimiliano Maini]

Hi Scotty,

I cannot reproduce, I can export well to PDF, numbers are shown.

Can you reinstall the most recent archive and double check.

MaX.

2009/9/25  :
> Max,
>
> Recently upgraded GNU via full archive install.
>
> Sent message below to Stick, then I realized your the man :-).
>
> Attached in word document is error message and by build.
>
> Please advise.
>
> ScottyK
>
>
> Stick,
>
> You ever seen this message in GNU?  Came when trying to save settings. Can't
> save settings, wtf is foing on. A bug in new build?
>
> I attached images of as a word document only with my latest version.
>
> Hope all is well,
>
> ScottyK
>
>
>
>
> --
> ScottyK
>
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] New win builds on gnubg.org (20090928 code)

[Massimiliano Maini]

I've also created an exe with a proposal for layout and naming of "Go"
toolbar buttons and menus.
Opinions on naming and layout welcome :)

The exe is here:

http://www.gnubg.org/media/windows/gnubg-buttons-experimental.exe


For Christian/Jon:
- I've changed only the GUI labels, not variable names
- I haven't changed the gnubg commands
- There's still an issue with the prev/next atgger (previously marked)
menus, they are not grayed out when relevant.

The modified code is in the zip here:

http://www.gnubg.org/media/windows/gnubg-buttons-experimental.zip

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] New win builds on gnubg.org (20090925 code)

[Massimiliano Maini]

Hi,

this new build includes the latest changes and a minor update on GTK
libs, so please reinstall everything.
It should solve the PDF export bug.

Notice that on my "new" PC I do have a problem: the area I have to
click on to pick up the dice
is very small and randomly positioned around the dice, very hard to find it.

I don't know if it's limited to my PC/graphics card (XP, silly Intel
Q35 integrated adapter).

If anybody can test and report ...

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] pdf export in windows-xp

[Massimiliano Maini]

Confirmed. But I have no idea what's wrong/missing :(

MaX.

2009/9/25 Christian Anthon :
> Here is a report from savannah. Seems like the windows build is
> missing some cairo component. At least it still works on linux.
>
> "When attempting to save a pdf file the error unable to create cario
> surface for [file path] appears. I was using the 20090721 build
> without this issue. The 20090913 and 20090924 builds both display this
> error."
>
> Christian.
>
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Re: Few recent issues (20090922 code)

[Massimiliano Maini]

Also, ni the CLI I have this:

(No game) help next
next - Step ahead within the game

Usage: next [game|roll|rolled|marked] 

"within the game" doesn't sound convncing for a "next game" :)

Also, what the hell is "next rolled" ?

MaX.

2009/9/24 Massimiliano Maini :
> Right. Logicallly it should be Mark (or Flag or tag) for what today is
> CMark, and Error for what today is Mark...
>
> 2009/9/24 Ian Shaw :
>>
>>
>>> -Original Message-
>>> From: bug-gnubg-bounces+ian.shaw=riverauto.co...@gnu.org
>>> [mailto:bug-gnubg-bounces+ian.shaw=riverauto.co...@gnu.org]
>>> On Behalf Of Massimiliano Maini
>>> Sent: 24 September 2009 11:08
>>> To: Christian Anthon
>>> Cc: bug-gnubg@gnu.org
>>> Subject: Re: [Bug-gnubg] Re: Few recent issues (20090922 code)
>>>
>>> We can keep the commands in the CL interface as they are, but
>>> we can still change the labels in the menus and toolbar.
>>>
>>> I'm for changing the commands in the CLI too, or at least to
>>> define aliases like next roll = next move, so that backward
>>> compatibility is kept and clearer commands are available too
>>> (in sync with GUI).
>>>
>>> Also, I really dislike not having the buttons for prev marked
>>> and prev cmarked. if the issue is space, we canmake things
>>> smaller or eventually remove the buttons for prev/next game,
>>> since we have the drop down list anyway.
>>>
>>
>> If you're changing labels, perhaps you should think about the Cmark
>> itself.
>>
>> I've worked out what it does, obviously, but I've no idea what "Cmark"
>> actually means, so not will many others.
>>
>> How about just "Mark", "Tag" or "Flag"?
>>
>> -- Ian
>>
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] New win builds on gnubg.org (20090924 code)

[Massimiliano Maini]

Should also solve some possible install issues recently discovered
(windows dll hell).

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Re: Few recent issues (20090922 code)

[Massimiliano Maini]

Seems fixed now.

I have a strange behavior since I changed my PC, don't know if related
to the changes
in the code or in the hardware: after rolling and moving, when i have
to "confirm" the
move picking up the dice, I have difficulties in clicking in the right
area. I have to click
randomly around the dices.

I'll upload a new distrib to see if others have the same issue.

MaX.

2009/9/22 Christian Anthon :
> On Tue, Sep 22, 2009 at 12:25 PM, Massimiliano Maini  
> wrote:
>>
>> Refresh of the new buttons is not working in multiple occasions: they are 
>> often
>> grayed out when they shouldn't.
>
> I think I know why: The match state is not "game playing" before the
> first move record.
>
> Christian.
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Re: Few recent issues (20090922 code)

[Massimiliano Maini]

We can keep the commands in the CL interface as they are, but
we can still change the labels in the menus and toolbar.

I'm for changing the commands in the CLI too, or at least to define
aliases like next roll = next move, so that backward compatibility
is kept and clearer commands are available too (in sync with GUI).

Also, I really dislike not having the buttons for prev marked and prev
cmarked. if the issue is space, we canmake things smaller or
eventually remove the buttons for prev/next game, since we have the
drop down list anyway.

MaX.

2009/9/22 Christian Anthon :
> On Tue, Sep 22, 2009 at 12:25 PM, Massimiliano Maini  
> wrote:
>> Other comments:
>>
>> labels on the toolbar are too long and a bit unclear (marked/cmarked).
>> I would:
>>
>> - use "Prev." instead of "Previous", shorter (we can leave previous in the 
>> menu)
>> - replace "Prev./Next Marked" with "Prev./Next Error" (in the menu too)
>>
>> Last thing: "Previous" and "Next" in the toolbar corrspond to "Previous roll"
>> and "Next roll" in the menu. We shold harmonize. Also, I don't like "roll"
>> as it moves through cube decisions too. "Prev.Next decision" would be fine,
>> but it's a bit long for the toolbar. I would settle for "Prev./Next move".
>>
>
> I more or less agrees with you. The reason for the words marked and
> roll is that they match with the commands used in the CL interface...
>
>> Refresh of the new buttons is not working in multiple occasions: they are 
>> often
>> grayed out when they shouldn't.
>
> I think I know why: The match state is not "game playing" before the
> first move record.
>
> Christian.
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Installation & Application Error Problem - Help Needed

[Massimiliano Maini]

I think I've found the issue.

I've changed my PC at office (still XP) and I've experiencing the same problem.

With gdb and a debug exe I got this:

D:\Documents\gnubg>gdb gnubg-nogui-py-sse2-nomt-O3.exe-debug.exe
GNU gdb 6.8
Copyright (C) 2008 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "i686-pc-mingw32"...
(gdb) run
Starting program: D:\Documents\gnubg/gnubg-nogui-py-sse2-nomt-O3.exe-debug.exe
[New thread 2980.0x208]
warning: LDR: LdrpWalkImportDescriptor() failed to probe
D:\Documents\gnubg\python26.dll for its manifest, ntstatus 0xc0150002

gdb: unknown target exception 0xc0150002 at 0x7c98

Program received signal ?, Unknown signal.

Program exited with code 03000522.
You can't do that without a process to debug.
(gdb) quit

I've reinstalled python and now I have the msvcr90.dll and also a file named:

  Microsoft.VC90.CRT.manifest

Adding this into gnubg's install dir seems to solve the problem.

I guess that on PCs where python was installed 9and the dll registerd
or whatever)
the problem was not occurring. I'll make it available in the next
install archive.

MaX.

2009/9/23 birol morgül 
>
>
> Hi
>
> I think my problem is not related with Vista.
>
>  My computer has XP and GNUBG do not work. Previous version (before 4 
> september 2009 ) GNUBGs work very well.
> Maybe there is problem in my XP. (But I do not think)
> All other programs work in my XP. I  did not meet any problem up to now.
>
> Other 2 computers: (of my children)
>
> In Computer with Vista, GNUBG works and the computer with Vista  has those 
> DLLs in it.
>
> In computer with XP , GNUBG works  and there are not those DLLs in it
>
> Maybe there is another problem
> maybe problem is not related with those DLLs.
> Maybe it is necessary to install XP in my computer again.
>
> Thank you
>
> BM
>
>
> Kimden: Massimiliano Maini 
> Konu: Re: [Bug-gnubg] Installation & Application Error Problem - Help Needed
> Kime: "birol morgül" , bug-gnubg@gnu.org
> Tarihi: 23 Eylül 2009 Çarşamba, 1:04
>
> Hi,
>
> most likely your Vista has a few problems.
> I checked my one and the dlls you miss (except msvcr90.dll) are all
> there and they are all Windows system stuff, they don't come from
> python or whatever else.
>
> I suggest you run a dos shell in administrator mode
> (start/programs/accessories/command prompt, right click and select
> "Run as admin"°,
> then at dos prompt type:
>
> sfc /verifyonly
>
> If it finds problems, you'll have to try this:
>
> sfc /scannow
>
> It should repair most common issues.
>
> MaX.
>
> 2009/9/22 birol morgül 
> >
> > Hi Max
> >
> > I determined below dlls using "dependence walker" program
> >
> >
> > Wer.dll,   Bcrypt.dll, Ncrypt.dll , İeshims.dll, Msvcr90.dll    (some of 
> > them are new -september.2009).
> >
> > I downloaded them from the net.
> >
> > These dll files were copied to system32 directory   (I do not know iti is 
> > right or not)
> >
> > then again there was error related with python. and I could not understand.
> > Error message is below
> >
> > Error: The Side-by-Side configuration information for "c:\program 
> > files\gnubg-i\PYTHON26.DLL" contains errors. This application has failed to 
> > start because the application configuration is incorrect. Reinstalling the 
> > application may fix this problem (14001).
> > Warning: At least one module has an unresolved import due to a missing 
> > export function in a delay-load dependent module.
> >
> >
> > my computer has XP
> > At home my children's computers have XP and Vista
> >
> > In these  computers , GNUBG works very well
> >
> > the computer with  Vista has abowe dll files (maybe not all of them)
> > but
> > the computer with XP has not these dlls   (but GNUBG works)
> >
> > it is strange.
> >
> > BM
>
> 
> Yahoo! Türkiye açıldı!
> Haber, Ekonomi, Videolar, Oyunlar hepsi Yahoo! Türkiye'de!
> www.yahoo.com.tr


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Installation & Application Error Problem - Help Needed

[Massimiliano Maini]

Hi,

most likely your Vista has a few problems.
I checked my one and the dlls you miss (except msvcr90.dll) are all
there and they are all Windows system stuff, they don't come from
python or whatever else.

I suggest you run a dos shell in administrator mode
(start/programs/accessories/command prompt, right click and select
"Run as admin"°,
then at dos prompt type:

sfc /verifyonly

If it finds problems, you'll have to try this:

sfc /scannow

It should repair most common issues.

MaX.

2009/9/22 birol morgül 
>
> Hi Max
>
> I determined below dlls using "dependence walker" program
>
>
> Wer.dll,   Bcrypt.dll, Ncrypt.dll , İeshims.dll, Msvcr90.dll    (some of them 
> are new -september.2009).
>
> I downloaded them from the net.
>
> These dll files were copied to system32 directory   (I do not know iti is 
> right or not)
>
> then again there was error related with python. and I could not understand.
> Error message is below
>
> Error: The Side-by-Side configuration information for "c:\program 
> files\gnubg-i\PYTHON26.DLL" contains errors. This application has failed to 
> start because the application configuration is incorrect. Reinstalling the 
> application may fix this problem (14001).
> Warning: At least one module has an unresolved import due to a missing export 
> function in a delay-load dependent module.
>
>
> my computer has XP
> At home my children's computers have XP and Vista
>
> In these  computers , GNUBG works very well
>
> the computer with  Vista has abowe dll files (maybe not all of them)
> but
> the computer with XP has not these dlls   (but GNUBG works)
>
> it is strange.
>
> BM


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Re: Few recent issues (20090922 code)

[Massimiliano Maini]

Maybe even somethig like "<<" for prev and ">>" for next is ok for the labels.
Anyway the menus and the tooltips are more explicit.

I do agree that it must fit 1024 width.

MaX.

2009/9/22 Philippe Michel :
> On Tue, 22 Sep 2009, Massimiliano Maini wrote:
>
>> Other comments:
>>
>> labels on the toolbar are too long [...]
>
> Yes! Keep it keep working on a 1024 pixels screen width, please. This is no
> longer the case with the recent changes
>
>
>
> ___
> Bug-gnubg mailing list
> Bug-gnubg@gnu.org
> http://lists.gnu.org/mailman/listinfo/bug-gnubg
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] Re: Few recent issues (20090922 code)

[Massimiliano Maini]

Other comments:

labels on the toolbar are too long and a bit unclear (marked/cmarked).
I would:

- use "Prev." instead of "Previous", shorter (we can leave previous in the menu)
- replace "Prev./Next Marked" with "Prev./Next Error" (in the menu too)

Last thing: "Previous" and "Next" in the toolbar corrspond to "Previous roll"
and "Next roll" in the menu. We shold harmonize. Also, I don't like "roll"
as it moves through cube decisions too. "Prev.Next decision" would be fine,
but it's a bit long for the toolbar. I would settle for "Prev./Next move".

Refresh of the new buttons is not working in multiple occasions: they are often
grayed out when they shouldn't.

MaX.

2009/9/22 Massimiliano Maini :
> - toolbar buttons for previous marked and previous cmarked are missing:
> there are missing lines in gtktoolbar.c:
>  - in the declaration of toolbarwidget
>  - in ToolbarUpdate
>  - in ToolbarNew
>
> - When a match is anlyzed an you go to the 1st game (or any other)
> using the drop down list, the buttons are grayed out. They return available
> as soon as you click on any move in a the game record.
>
> - Regarding Jon's changes to analysis settings:  remove my gnubgautorc,
> by default the anlaysis is at "expert". Now I change it to WorldClass and
> then back to expert, I get this warning:
>
> You must set a count of extra moves and a search tolerance (see `help
> set analysis movefilter').
>
> MaX.
>


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Installation & Application Error Problem - Help Needed

[Massimiliano Maini]

OK, I have to include msvc90.dll in the distrib (instead of old
msvcrt71.dll), but I
don't think the others are needed. I don't see them on my XP PC, I'll check at
home (Vista).

For the records, the python 2.6 msi installer seems to be buggy and it doesn't
inistall msvcrt90.dll .. strange. I'll fetch on the web and manually add it.

Anyway, bad idea to manually put the dlls in the system32 dir.
In case you need it, just put them in gnubg installation dir.

I'll add msvcrt90.dll in the next build.

2009/9/22 birol morgül 
>
> Hi Max
>
> I determined below dlls using "dependence walker" program
>
>
> Wer.dll,   Bcrypt.dll, Ncrypt.dll , İeshims.dll, Msvcr90.dll    (some of them 
> are new -september.2009).
>
> I downloaded them from the net.
>
> These dll files were copied to system32 directory   (I do not know iti is 
> right or not)
>
> then again there was error related with python. and I could not understand.
> Error message is below
>
> Error: The Side-by-Side configuration information for "c:\program 
> files\gnubg-i\PYTHON26.DLL" contains errors. This application has failed to 
> start because the application configuration is incorrect. Reinstalling the 
> application may fix this problem (14001).
> Warning: At least one module has an unresolved import due to a missing export 
> function in a delay-load dependent module.
>
>
> my computer has XP
> At home my children's computers have XP and Vista
>
> In these  computers , GNUBG works very well
>
> the computer with  Vista has abowe dll files (maybe not all of them)
> but
> the computer with XP has not these dlls   (but GNUBG works)
>
> it is strange.
>
> BM


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] Few recent issues (20090922 code)

[Massimiliano Maini]

- toolbar buttons for previous marked and previous cmarked are missing:
there are missing lines in gtktoolbar.c:
  - in the declaration of toolbarwidget
  - in ToolbarUpdate
  - in ToolbarNew

- When a match is anlyzed an you go to the 1st game (or any other)
using the drop down list, the buttons are grayed out. They return available
as soon as you click on any move in a the game record.

- Regarding Jon's changes to analysis settings:  remove my gnubgautorc,
by default the anlaysis is at "expert". Now I change it to WorldClass and
then back to expert, I get this warning:

You must set a count of extra moves and a search tolerance (see `help
set analysis movefilter').

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Feature Request: Gnu to use rollouts for playing

[Massimiliano Maini]

Ideal would be just to implement the feature and make it configurable.
Gnubg should be able to play (and do rollouts ?) at a configurable
rollout level.
Establishing what level (where to truncate, ply during play, how to
stop , VR on/off)
is good enough may be a personal matter. If the feature is available and
configurable, we should just agree on its default setting (which may
as well be the
same as XG, for comparison purposes).

MaX.

2009/9/18 Michael Depreli :
> I think if you're going to introduce this feature then what would impress
> the masses is a more accurate version
> of XGRoller rather than something based on speed (within reason).
>
> My tests using an @6 rollout WITH VR combined with stop on jsd produced some
> very impressive results
> for checker play comparison rollouts. (Relative equity rather than absolute)
>
> I haven't checked out cube decisions but it did occur to me that it wasn't
> necessary to have the
> same settings for cube and checker rollouts if it could be shown that
> different settings were better for each.
>
>
>
>> Subject: RE: [Bug-gnubg] Feature Request: Gnu to use rollouts for playing
>> Date: Fri, 18 Sep 2009 18:21:02 +0100
>> From: ian.s...@riverauto.co.uk
>> To: michaeldepr...@hotmail.co.uk; bug-gnubg@gnu.org
>>
>>
>> Michael Depreli wrote:
>>
>>
>> Are you sure that for 0-ply switching off VR is more accurate in
>> a given time?
>> Unless I'm doing something wrong I get the opposite results.
>>
>>
>> Michael, its me that's doing it wrong. I mis-read the SE output. Here is
>> a revised test (still only short), showing that VR leads to slightly
>> LOWER SE for the same rollout period.
>>
>> Time Without VR With VR
>> (sec) Trials SE Trials SE
>> ===
>> 60 8817 0.033 376 0.024
>> 120 16621 0.023 744 0.021
>> 180 25356 0.019 1112 0.017
>> 216 31131 0.017 1296 0.016
>>
>> So, at 0-ply, using VR or not is a wash. VR might be slightly better in
>> terms of a low SE, but to impress the masses it might be optimal to
>> choose no VR and lots of trials in a short time.
>>
>> -- Ian
>>
>> > Subject: RE: [Bug-gnubg] Feature Request: Gnu to use rollouts
>> for playing
>> > Date: Fri, 18 Sep 2009 13:12:12 +0100
>> > From: ian.s...@riverauto.co.uk
>> > To: bug-gnubg@gnu.org
>> >
>> >
>> >
>> > I've finally got around to running some tests.
>> >
>> > Test Environment:
>> > Intel Core2 Duo T8100 @ 2.10 GHz, 2 GB RAM Windows XP
>> Professional GNU
>> > Backgammon 0.90-mingw 20090914 Cache 5 MB
>> >
>> >
>> > The test position was opening 43: 13/9 13/10 in a money game
>> > 1296 trials, cubeful evaluation, truncation at database
>> (default
>> > databases installed)
>> >
>> > Thr VR Plies Trials/min Time
>> > ==
>> > 2 No 0 7776 00:00:09
>> > 2 Yes 0 390 00:03:30
>> > 1 No 0 4860 00:00:16
>> > 1 Yes 0 205 00:06:22
>> > 2 No 1 200 00:06:20
>> > 2 Yes 1 131 00:08:59
>> > 1 No 1 103 00:12:04
>> > 1 Yes 1 74 00:18:18
>> > 2 No XGR 15552 00:00:05
>> > 2 Yes XGR 1897 00:00:41
>> >
>> > Thr = number of threads
>> > XGR = Extreme Gammon XGRoller emulation: 0 ply rollouts
>> truncated at 6
>> > moves, first two cube decisions at 1-ply.
>> > Extreme Gammon does not use VR for XGRoller, but I have
>> repeated the
>> > test with VR on.
>> >
>> > It is clear that the speed hit is enormous when using VR for
>> 0-ply
>> > rollouts, by a factor of around 20.
>> >
>> > At 1-ply, the speed hit is still noticeable, but only by a
>> factor of
>> > around 1.5.
>> >
>> > At 5 seconds per move, the XGR setting is certainly viable for
>> play and
>> > analysis. And reading the BGOnline forums, there certainly
>> seems to be a
>> > demand for this feature. (There is no reason for gnubg to copy
>> XG's
>> > exact settings, if we think something else is superior. In
>> fact, I think
>> > it should be possible to specify a .rol file to use so that
>> people can
>> > choose. The default option, though, should be FAST.)
>> >
>> > High speed is great, but the reliability of the result is also
>> important
>> > (though oft overlooked). I also tested the Standard Error
>> reported after
>> > one minute for various settings. All tests used two threads.
>> >
>> > VR Plies Trials SE
>> > =
>> > Yes 1 136 0.016
>> > No 1 206 0.088
>> > Yes 0 373 0.016
>> > No 0 8836 0.002
>> >
>> > It appears that to achieve a low SE in a given time, VR is
>> worthwhile
>> > for 1-ply rollouts, but for 0-ply rollouts it is better to
>> specify a
>> > much larger number of trials and switch off VR.
>> >
>> >
>> > -- Ian
>> >
>> > > -Original Message-
>> > > From: Christian Anthon [mailto:christian.ant...@gmail.com]
>> > > Sent: 21 August 2009 22:33
>> > > To: Ian Shaw
>> > > Cc: bug-gnubg@gnu.org
>> > > Subject: Re: [Bug-gnubg] Feature Request: Gnu to use
>> rollouts for
>> > > playing
>> > >
>> > > Can you check the speed of the threaded/non-threaded
>> rollouts (say
>> > > 0ply eval 6ply truncation).
>> > >
>> > > Christia

Re: [Bug-gnubg] Feature Request: Gnu to use rollouts for playing

[Massimiliano Maini]

Could you add a column with the SE in the table below ?

MaX.

2009/9/18 Ian Shaw :
>
>
> I've finally got around to running some tests.
>
> Test Environment:
> Intel Core2 Duo T8100 @ 2.10 GHz, 2 GB RAM Windows XP Professional GNU
> Backgammon 0.90-mingw 20090914 Cache 5 MB
>
>
> The test position was opening 43: 13/9 13/10 in a money game
> 1296 trials, cubeful evaluation, truncation at database (default
> databases installed)
>
> Thr     VR      Plies   Trials/min      Time
> ==
> 2       No      0       7776            00:00:09
> 2       Yes     0       390             00:03:30
> 1       No      0       4860            00:00:16
> 1       Yes     0       205             00:06:22
> 2       No      1       200             00:06:20
> 2       Yes     1       131             00:08:59
> 1       No      1       103             00:12:04
> 1       Yes     1       74              00:18:18
> 2       No      XGR     15552           00:00:05
> 2       Yes     XGR     1897            00:00:41
___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Simplifying the analyse/evaluation dialogs

[Massimiliano Maini]

Looks great to me !

MaX.

2009/9/17 Jonathan Kinsey :
> I've just checked in the first part of this (merging the analysis/evaluation
> settings). Take a look and see what you think.
>
> I'll take a look at the players and rollouts settings in a bit (I'm off for
> a long weekend tomorrow).
>
> Jon


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] Minor win install improvement

[Massimiliano Maini]

Works just fine. Thx.

MaX.

2009/9/17 Jonathan Kinsey :
> It's trivial to create a mutex for this, I'll add one later if I remember,
> I'll
> call it: "GNU Backgammon Mutex"
>
> Jon


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] New win builds on gnubg.org (20090918 code)

[Massimiliano Maini]

A few new things:

- Jon's changes on the Analysis/Hint settings dialogs (I love them !)
- Windows installer now detects if gnubg is running when you run the installer
- The default number of thread is now set to the autodetectednumber of
proc/cores (unless you already have set it to something)

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] Minor win install improvement

[Massimiliano Maini]

1 -It seems it is possible to detect gnubg is runnig when the
InnoSetup installer is launched (check the section Application is
running):

http://www.vincenzo.net/isxkb/index.php?title=Application_considerations

Does gnubg create such a mutex ? If yes, what's the mutex name ?
If no, can we do it so that I can add the check to the installer ?

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

Re: [Bug-gnubg] New win builds on gnubg.org (20090915 code)

[Massimiliano Maini]

2009/9/16 Jonathan Kinsey :
> It's a pity you can't just install the correct version automatically. Is
> this a
> limitation with inno setup?
>
> On a side note; I like the idea of having an initial page in the install
> that
> you select either "standard install" or "custom install", and if you pick
> the
> first it just installs with all the default options (i.e. no more input
> required).
>
> Jon

Yeah, InnoSetup is a bit unflexible ... it supports something called
"Pascal scripting" that is supposed to allow you to do anything you want,
but I don't really want to dig into it. If anybody has a better (free) installer
I may take a look.

BTW, I added the 2-way check i.e. you get a warning if you install
SSE2 exes on a non-SSE2 PC and a warning if you install NOSSE
exes on a SSE2 PC.

MaX.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] New win builds on gnubg.org (20090915 code)

[Massimiliano Maini]

I've added the choice between SSE2 (default) and non-SSE executables.
If SSE2 is selected, a test is run and if SSE2 is not supported, a
pop-up tells the user to re-run the install selecting non-SSE exes.

Michael D. (or anybody else with non-SSE2 CPU) can you test it ?

1- run the installer on your recent PC, select SSE2 exes, and then
check that the installed exes are SSE2 (run gnubg, Help/About gnubg
menu, Build Info button, you should find "SSE/SSE2" supported amongst
the features).

2- run the installer on your old PC, select SSE2 exes, and a pop up
should appear warning you. Then re-run the installer and select
non-SSE exes and then check that the installed exes are non-SSE2 (run
gnubg, Help/About gnubg menu, Build Info button, you should *not* find
"SSE/SSE2" supported amongst the features).

MaX.

P.S.
Next time I'll add the check the other way round too: you select
non-SSE exes on a SSE2 PC, you get a warning too.


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg

[Bug-gnubg] Re: Win SSE2 detection exe

[Massimiliano Maini]

Extract them to any directory, then duble click on the only .exe file.

On your recent PC it should say SSE2 is supported, on your old one it
should say it is not.

I'm integrating this somehow in the win distribs I build.

MaX.


2009/9/15 Michael Depreli :
> What do I need to do with all the files in the archive?


___
Bug-gnubg mailing list
Bug-gnubg@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-gnubg