Not logged inRybka Chess Community Forum
Up Topic Rybka Support & Discussion / Rybka Discussion / Rybka Source Code!
- - By Harvey Williamson (*****) Date 2007-06-24 08:05
Can someone who speaks German have a look at this link http://f23.parsimony.net/forum50826/messages/168155.htm

Is this guy claiming he has Rybka's source code?
Parent - - By Lukas Cimiotti (Bronze) Date 2007-06-24 08:08
Yes, he says he got it mailed
Parent - By er (**) Date 2007-06-24 08:13
... and meantime he deleted it.
EGR
Parent - - By er (**) Date 2007-06-24 08:16
He says he got the Rybka source code and other files
by email from CONVEKTA.

EGR
Parent - - By M ANSARI (*****) Date 2007-06-24 08:34
Well if that is true ... what a terrible terrible mistake by Convekta.  At least the person was honorable and did the right thing by deleting all the files.  Maybe more security should be in place there and I am not sure that Vas should be sending unencrypted source code around.  Personally I would make sure that source code is only on a computer that has no communication gear whatsoever ... no modem no LAN.  Any files that are sent by email should be encrypted and the encryption only deciphered when it is on a non networked computer.  Maybe Vas should get the heads up on this one.
Parent - - By Harvey Williamson (*****) Date 2007-06-24 08:36
I very much doubt that Convekta would have the source code.
Parent - By Kapaun (****) Date 2007-06-24 08:48
That's indeed doubtable. But it's hard to make anything out of this peculiar thread...
Parent - - By Lukas Cimiotti (Bronze) Date 2007-06-24 08:52
He didn´t say he got it from Convekta - he just says he got it mailed and it was attached to that mail. Convekta sends links.
Parent - By Michael Waesch Date 2007-06-24 08:55
Even worse. Why would Convekta upload the source code to the net and then addionally send out links to it?

That does not make any sense.

Mike
Parent - - By Michael Waesch Date 2007-06-24 08:52
Seems like a joke for me. Rybka is surely programmed in C and should contain a whole lot of different files than shown in this dubious posting. And if one really assumes that it was the source of Rybka, actually sent out in error by Convekta, this one should ask himself several questions:

* Why should Convekta send "updates" as single mail and not bulk, so that other also received these files in error?
* Why does the guy claiming he got Rybka´s source not post the header of the mail as proof ?
* Why does he ask the public what to do with the files? He didn´t even do the obvious: Run a virus scanner over the files. The sender of an e-mail adress can easily be forged.

Addionally, he offered to post parts of the code as proof. He didn´t. And besides: Why should Convekta have the source code saved somewhere so openly on harddrive? Do they also work on it? Nah, only Vas does ...

He also was advised to state his case here, where most of the Rybka enthusiasts live and even Vasik shows up regularly. He didn´t ...

So it is most likely that someone just wanted to cause some alarm and enjoy the attention he would get from that. Well, it didn´t work.

Mike
Parent - - By Kapaun (****) Date 2007-06-24 09:06
That is indeed the most likely explanation. Unless the mail he got perhaps was not from Convekta (actually he didn't claim it was) but from someone else...
Parent - - By Michael Waesch Date 2007-06-24 09:10
If I had and engine which I had to make a living from, I would have the source files on a computer that is not connected to the internet as one of the simplest but effective measures to prevent accidential distribution, hacking attempts, downloader trojans etc ...

Mike
Parent - - By Harvey Williamson (*****) Date 2007-06-24 09:15
I agree. I doubt this guy has the source code. There is no need for Convekta to have it.

ChessBase have certainly never seen the Hiarcs source code.

Maybe they sent him the source code for a beta version of the new GUI ;-)
Parent - - By Kapaun (****) Date 2007-06-24 09:30
I'm thinking of the frequent claims someone having succeeded in decompiling Rybka. Maybe one of them really had success? I can't imagine, though, what he would gain by sending the stuff around...
Parent - By Michael Waesch Date 2007-06-24 10:06
Given that someone really might have had success in decompiling Rybka, (s)he has now an almost unreadable code on harddrive. That´s it. Getting this code structured again would be a real pain in the *** - and to do this, in order to achieve what? For just having it illegally on a harddrive to lie there and then go online and bragg about?

Either this hypothetical guy/gal is an excellent programmer and now goes into it for his/her own engine or some not so satisfied Convekta customer who now sends it to Chessbase.

But I don´t see any case where someone should make himself all this work, only for going online and stupidly ask what to do with the files right now ...

Mike
Parent - - By Uri Blass (*****) Date 2007-06-24 13:30
Your idea may be a safe way to prevent accidential distribution but it also make working on the source less convenient.

You need to use more than one computer if you want both to work on the code and read the internet.
If you want to test the exe file in games that are done by the internet
You need to copy the exe file from one computer to another computer.

I simply dislike it and I doubt if the risk of accidential distribution is worth it.

Even in case of accidential distribution I think that Vas has a clear advantage relative to the people who get the code because he worked on the code and understand it so even in that case I expect rybka to continue to dominate(toga is still better than many commercial programs inspite of being open source program and toga has no person who works full time job on improving it).

Uri
Parent - - By Michael Waesch Date 2007-06-24 14:04
You forget that in case of distribution competitors may annihilate the current advantage in strength of Rybka, rendering it only one engine amongst many others. I would take the "inconvenience" and only Vasik knows how things are actually handled with Rybka. Besides: Nowaday it´s very easy and comfortable to connect 2 computers to one monitor and keyboard and mouse to two computers residing on one and the same desk easy to switch ...

Mike
Parent - - By Uri Blass (*****) Date 2007-06-24 14:42
I think that things are not so easy.

If competitors plan to modify the source and make a rybka clone then it may be easy to detect it and learning the code may take a significant time and I think that the time may be enough for Vas to improve rybka so rybka will not be one out of many programs with similiar strength.

Uri
Parent - By Michael Waesch Date 2007-06-24 15:04
Well, if you are really that convinced of your opinion, the next logical step for you is ask Vasik to release his code.
Mike
Parent - - By Maria flor de luna Date 2007-06-24 11:34 Edited 2007-06-24 11:36
I checked the files mentioned in this forum and there is no Rybka source code in it.

autosave.txt
Chess.dll
Chess.pdb
Chess.xml
Shared.exe
Shared.pdb
Text.exe
Test.pdb
Test.vshost.exe
Tester Processing Controls.dll
Tester Processing Controls.pdb
TesterDialogs.dll
Utils.dll
Utils.pdb
XControls.dll
XControls.pdb
Rybkav2.3.2.config.x64.exe
Rybka_v2.3.2.config.x64.exe comments.txt


Or maybe the guy thinks that the last file named: Rybka_v2.3.2.config.x64.exe comments.txt is the source code but i don't believe because no programming language looks like this:

If we are in the main search with depth = 7 with a can-fail-low situation, and the post-move evaluation score is more than this away from alpha, and several additional post-move criteria are met, we allow the move to fail low without any further work......depth_7_cfl_optimism_margin_opening.....depth_6_cfl_optimism_margin_endgame.....If we are in the main search with depth = 6 with a can-fail-low situation, and the post-move evaluation score is more than this away from alpha, and several additional post-move criteria are met, we allow the move to fail low without any further work......depth_6_cfl_optimism_margin_opening.....depth_5_cfl_optimism_margin_endgame.....If we are in the main search with depth = 5 with a can-fail-low situation, and the post-move evaluation score is more than this away from alpha, and several additional post-move criteria are met, we allow the move to fail low without any further work......depth_5_cfl_optimism_margin_opening.....depth_4_cfl_optimism_margin_endgame.....If we are in the main search with depth = 4 with a can-fail-low situation, and the post-move evaluation score is more than this away from alpha, and several additional post-move criteria are met, we allow the move to fail low without any further work......depth_4_cfl_optimism_margin_opening.....depth_3_cfl_optimism_margin_endgame.....If we are in the main search with depth = 3 with a can-fail-low situation, and the post-move evaluation score is more than this away from alpha, and several additional post-move criteria are met, we allow the move to fail low without any further work......depth_3_cfl_optimism_margin_opening.....depth_6_cfh_optimism_margin_endgame.....If we are in the main search with depth = 6 with a can-fail-high situation, and the post-move evaluation score is more than this away from alpha, and several additional post-move criteria are met, we allow the move to fail low without any further work.....depth_6_cfh_optimism_margin_opening.....depth_5_cfh_optimism_margin_endgame.....If we are in the main search with depth = 5 with a can-fail-high situation, and the post-move evaluation score is more than this away from alpha, and several additional post-move criteria are met, we allow the move to fail low without any further work.....depth_5_cfh_optimism_margin_opening.....depth_4_cfh_optimism_margin_endgame.....If we are in the main search with depth = 4 with a can-fail-high situation, and the post-move evaluation score is more than this away from alpha, and several additional post-move criteria are met, we allow the move to fail low without any further work.....depth_4_cfh_optimism_margin_opening.....depth_3_cfh_optimism_margin_endgame.....If we are in the main search with depth = 3 with a can-fail-high situation, and the post-move evaluation score is more than this away from alpha, and several additional post-move criteria are met, we allow the move to fail low without any further work.....depth_3_cfh_optimism_margin_opening.....depth_2_cfh_optimism_margin_endgame.....If we are in the main search with depth = 2 with a can-fail-high situation, and the post-move evaluation score is more than this away from alpha, and several additional post-move criteria are met, we allow the move to fail low without any further work.....depth_2_cfh_optimism_margin_opening.....Applies to all 'depth_*_cf*_optimism_margin' weights....high_depth_optimism_margin_linear_root_phase....depth_6_cfl_pawn_cap_margin_endgame.............If we are in the main search with depth = 6 with a can-fail-low situation and the current evaluation score is more than this distance away from alpha, we do not consider pawn captures which do not come with check....depth_6_cfl_pawn_cap_margin_opening.....depth_5_cfl_pawn_cap_margin_endgame.............If we are in the main search with depth = 5 with a can-fail-low situation and the current evaluation score is more than this distance away from alpha, we do not consider pawn captures which do not come with check....depth_5_cfl_pawn_cap_margin_opening.....depth_4_cfl_pawn_cap_margin_endgame.............If we are in the main search with depth = 4 with a can-fail-low situation and the current evaluation score is more than this distance away from alpha, we do not consider pawn captures which do not come with check....depth_4_cfl_pawn_cap_margin_opening.....depth_3_cfl_pawn_cap_margin_endgame.............If we are in the main search with depth = 3 with a can-fail-low situation and the current evaluation score is more than this distance away from alpha, we do not consider pawn captures which do not come with check....depth_3_cfl_pawn_cap_margin_opening.....depth_5_cfh_pawn_cap_margin_endgame.............If we are in the main search with depth = 5 with a can-fail-high situation and the current evaluation score is more than this distance away from alpha, we do not consider pawn captures which do not come with check...depth_5_cfh_pawn_cap_margin_opening.....depth_4_cfh_pawn_cap_margin_endgame.............If we are in the main search with depth = 4 with a can-fail-high situation and the current evaluation score is more than this distance away from alpha, we do not consider pawn captures which do not come with check...depth_4_cfh_pawn_cap_margin_opening.....depth_3_cfh_pawn_cap_margin_endgame.............If we are in the main search with depth = 3 with a can-fail-high situation and the current evaluation score is more than this distance away from alpha, we do not consider pawn captures which do not come with check...depth_3_cfh_pawn_cap_margin_opening.....depth_2_cfh_pawn_cap_margin_endgame.............If we are in the main search with depth = 2 with a can-fail-high situation and the current evaluation score is more than this distance away from alpha, we do not consider pawn captures which do not come with check...depth_2_cfh_pawn_cap_margin_opening.....Applies to all 'depth_*_cf*_pawn_cap_margin' weights....high_depth_pawn_cap_margin_linear_root_phase....depth_9_cfl_futility_margin_endgame.....If we are in the main search with depth = 9 with a can-fail-low situation and the current evaluation score is more than this distance away from alpha, we consider only captures and some checks........depth_9_cfl_futility_margin_opening.....depth_8_cfl_futility_margin_endgame.............If we are in the main search with depth = 8 with a can-fail-low situation and the current evaluation score is more than this distance away from alpha, we consider only captures and some checks........depth_8_cfl_futility_margin_opening.....depth_7_cfl_futility_margin_endgame.............If we are in the main search with depth = 7 with a can-fail-low situation and the current evaluation score is more than this distance away from alpha, we consider only captures and some checks........depth_7_cfl_futility_margin_opening.....depth_6_cfl_futility_margin_endgame.............If we are in the main search with depth = 6 with a can-fail-low situation and the current evaluation score is more than this distance away from alpha, we consider only captures and some checks........depth_6_cfl_futility_margin_opening.....depth_5_cfl_futility_margin_endgame.............If we are in the main search with depth = 5 with a can-fail-low situation and the current evaluation score is more than this distance away from alpha, we consider only captures and some checks........depth_5_cfl_futility_margin_opening.....depth_4_cfl_futility_margin_endgame.............If we are in the main search with depth = 4 with a can-fail-low situation and the current evaluation score is more than this distance away from alpha, we consider only captures and some checks........depth_4_cfl_futility_margin_opening.....depth_3_cfl_futility_margin_endgame.............If we are in the main search with depth = 3 with a can-fail-low situation and the current evaluation score is more than this distance away from alpha, we consider only captures and some checks........depth_3_cfl_futility_margin_opening.....depth_6_cfh_futility_margin_endgame.............If we are in the main search with depth = 6 with a can-fail-high situation and the current evaluation score is more than this distance away from alpha, we consider only captures and some checks.......depth_6_cfh_futility_margin_opening.....depth_5_cfh_futility_margin_endgame.............If we are in the main search with depth = 5 with a can-fail-high situation and the current evaluation score is more than this distance away from alpha, we consider only captures and some checks.......depth_5_cfh_futility_margin_opening.....depth_4_cfh_futility_margin_endgame.............If we are in the main search with depth = 4 with a can-fail-high situation and the current evaluation score is more than this distance away from alpha, we consider only captures and some checks.......depth_4_cfh_futility_margin_opening.....depth_3_cfh_futility_margin_endgame.............If we are in the main search with depth = 3 with a can-fail-high situation and the current evaluation score is more than this distance away from alpha, we consider only captures and some checks.......depth_3_cfh_futility_margin_opening.....Applies to all 'depth_*_cf*_futility_margin' weights....high_depth_futility_margin_linear_root_phase....tip_futility_margin_endgame.............If we are in the main search with depth = 2 and the current evaluation score is more than this distance away from alpha, we consider only captures and some checks......tip_futility_margin_opening.....tip_futility_margin_linear_root_phase...depth_1_futility_depth..depth_1_futility_margin_endgame.........If we are more than this distance below alpha at depth <= 'depth_1_futility_depth' of the main search, then we leave the situation for a depth == 1 quiescence search...depth_1_futility_margin_opening.depth_1_futility_margin_linear_root_phase.......tip_evasion_margin_endgame..............If we are in check at depth <= 3 of the main search and the current score is more than this distance away from alpha, we consider only evasion moves which capture the checking piece...tip_evasion_margin_opening......tip_evasion_margin_linear_root_phase....q_evasion_minor_margin_endgame..If we are in check in any depth of the quiescence search and the current score is more than this distance away from alpha, we consider only evasion moves which capture a piece, not including captures of pawns, knights or bishops....q_evasion_minor_margin_opening..q_evasion_minor_margin_linear_root_phase........q_evasion_pawn_margin_endgame...........If we are in check in any depth of the quiescence search and the current score is more than this distance away from alpha, we consider only evasion moves which capture a piece, not including captures of pawns........q_evasion_pawn_margin_opening...q_evasion_pawn_margin_linear_root_phase.q_evasion_margin_endgame........If we are in check in any depth of the quiescence search and the current score is more than this distance away from alpha, we consider only evasion moves which capture a piece.........q_evasion_margin_opening........q_evasion_margin_linear_root_phase......positional_weaker_moves_margin_endgame..........If the current evaluation score is within this distance of alpha, we generate all quiet moves with all piece types at depth = 1 of the quiescence search. These moves are pruned using the piece-from-to tables.........positional_weaker_moves_margin_opening..positional_weaker_moves_margin_linear_root_phase........positional_moves_margin_endgame.........If the current evaluation score is within this distance of alpha, we generate all quiet king and knight moves at depth = 1 of the quiescence search. These moves are pruned using the piece-from-to tables......positional_moves_margin_opening.positional_moves_margin_linear_root_phase.......q_checks_margin_min_3_endgame...If the current evaluation score is within this distance of alpha, we try all 'reasonable' checks at depth = -3 or lower of the quiescence search........q_checks_margin_min_3_opening...q_checks_margin_min_2_endgame...........If the current evaluation score is within this distance of alpha, we try all 'reasonable' checks at depth = -2 of the quiescence search.........q_checks_margin_min_2_opening...q_checks_margin_min_1_endgame...If the current evaluation score is within this distance of alpha, we try all 'reasonable' checks at depth = -1 of the quiescence search.........q_checks_margin_min_1_opening...q_checks_margin_0_endgame.......If the current evaluation score is within this distance of alpha, we try all 'reasonable' checks at depth = 0 of the quiescence search..q_checks_margin_0_opening.......q_checks_margin_1_endgame...............If the current evaluation score is within this distance of alpha, we try all 'reasonable' checks at depth = 1 of the quiescence search..q_checks_margin_1_opening.......q_checks_margin_linear_root_phase.......The lowest depth of the quiescence search where non-capturing checks are still tried....q_checks_depth..q_rook_margin_endgame...If the current evaluation score is further than this distance below alpha, we do not try captures of rooks in all depths of the quiescence search.......q_rook_margin_opening...q_rook_margin_linear_root_phase.q_minor_margin_endgame..........If the current evaluation score is further than this distance below alpha, we do not try captures of minor pieces in all depths of the quiescence search........q_minor_margin_opening..q_minor_margin_linear_root_phase........q_pawn_margin_endgame...........If the current evaluation score is further than this distance below alpha, we do not try non-checking pawn captures in all depths of the quiescence search......q_pawn_margin_opening...q_pawn_margin_linear_root_phase.........# of half-plies by which all searches are extended; can be used to calibrate the value of more time for a given time control....root_extension..# of half-plies by which all checks outside the PV are extended; should be 0, 1 or 2....check_extension.........
Parent - - By Michael Waesch Date 2007-06-24 11:49
I checked the files mentioned in this forum and there is no Rybka source code in it.

How did you check them? Did you get those files also? The original poster was a German ...

Mike
Parent - - By fhub (**) Date 2007-06-24 12:28

>How did you check them? Did you get those files also? The original poster was a German ...


Well, in fact nobody got these files via Email by Convekta -
this package is available at RapidShare since about a week ...
Parent - - By Michael Waesch Date 2007-06-24 12:36
Now we are coming more close to the truth. And as what is the download advertised?

Mike
Parent - By fhub (**) Date 2007-06-24 12:44
Just as ´Some Rybka Stuff´ ...
Parent - By fhub (**) Date 2007-06-24 12:25

>Or maybe the guy thinks that the last file named:
>Rybka_v2.3.2.config.x64.exe comments.txt is the source code
>but i don't believe because no programming language looks like this:


>If we are in the main search with depth = 7 with a can-fail-low situation,
>and the post-move evaluation score is more than this away from alpha,
>and several additional post-move criteria are met, we allow the move to
>fail low without any further work......


Of course that´s a programming language -
it´s called HT (Human Thinking) ... ;-)
Parent - By Ernst (***) Date 2007-06-24 13:13 Edited 2007-06-24 13:20
So you have it too it seems?
Parent - - By grolich (***) Date 2007-06-24 16:21
Actually All those files are binaries and symbol files.
There is also the one .txt file.

So even in theory this could only be the Rybka project before it was linked into a single .exe (plus a few tests...), but after compilation/linking of separate modules.

So, it's not the source code of anything.

Whether or not those are Rybka's binaries at all can be tested, with some effort (The existence of the .pdb files make it a lot more feasible than with just having mixed binaries...).

Even if it is, though, That's a lot less severe than the source code being out.

----------------------------------------------------------
That said, I'd give one warning to all of you on this forum:

I would be VERY careful before running the .exe files in there (Actually, I would never run them on my computer unless it's in a VM or another contained environment, and even then I'd be a bit worried...).

The title those files are released with (Rybka stuff) convinces many curious people to execute them.
That's an excellent way for people to do whatever they want with your computer.
Might as well put up a desktop wallpaper that says: "Hackers welcome" in big, shiny, bright letters.
That's a common way to place backdoors into other people's computers,
and it's very popular with emule/bittorent/share networks in general.

Also, you cannot really be sure after you run it if something like this has happened, no matter what anti-virus / spyware detector you have,
there's always a way through,.

Be warned.
Parent - - By fhub (**) Date 2007-06-24 16:41

>I would be VERY careful before running the .exe files in there


Perhaps 95% (or more) people couldn´t even run those files,
because all of them are 64bit, so the danger isn´t that big.
Parent - By grolich (***) Date 2007-06-24 16:50
Hopefully, most people with 64 bit machines have them for a specific reasons, and they know better than to run those things.
However, you know what curiosity can do to people.

As to not many people being able to run 64 bit executables..... It's a CHESS program. That's one of the things in which malicious software creators (if that is indeed what it is) can be pretty sure they'll have many 64 bit "beta testers"...

Among people who are interested in the pre linked version of the (so called) Rybka binaries, that percentage of 64bit users would be even higher.
Parent - - By Michael Waesch Date 2007-06-25 00:54
I fully agree on that. That´s really bizarre that no one did the obvious and just ran a virus scanner over the files...

Mike
Parent - By grolich (***) Date 2007-06-25 07:27
If the files are what I suspect they might be, a virus scanner will not find anything.
If I can find the time, I'll look into it to make sure.

But yes, I agree, the basics should be checked first.
Parent - By Fulcrum2000 (****) Date 2007-06-25 16:49
Or maybe the guy thinks that the last file named: Rybka_v2.3.2.config.x64.exe comments.txt is the source code but i don't believe because no programming language looks like this:

If you look at the timestamp of the Rybka_v2.3.2.config.x64.exe file you see it's from *May 2006* A rybka called 2.3.2, created on 10 May 2006 (more then a year ago!). Seems like a big hoax to me...
Parent - - By Iptables (**) Date 2007-06-24 17:39
Hmm. There is a second guy who say he got a mail with code

http://f23.parsimony.net/forum50826/messages/168176.htm

with comments like this:

___________________________________________________________
member name="P:Chess.Position.WhiteKnightCount"
summary
Returns the square for the first instance of the specified piece.
Returns -1 if no instance of this piece is found.
summary
member
member name="P:Chess.Position.SomeCastlingPossible"
summary
True: at least one type of castling is possible in this position
/summary
value>
/member
_____________________________________________________________

Maybe Vas should take a look

Alfred
Parent - - By Lukas Cimiotti (Bronze) Date 2007-06-24 17:50
Yes he claims, he has got it. But he says it only has 5000 lines of code. Rybka must be much bigger. I programmed in assembler a long time ago - my lagest program had 120000 lines of code - and it was only an editor.
Btw. that was on Amiga :)
Parent - - By Kapaun (****) Date 2007-06-24 18:03
Besides that the mods seem to have talked with Vas who says it isn't anything of Rybka. So maybe it's material of some or other engine to make it plausible - which would backup the idea it might be some malicious code, a trojan or whatever, which is executed when starting the exe file...
Parent - - By Iptables (**) Date 2007-06-24 18:10
yepp this is now on the board.

btw. a look on the code =! running a binary
if someone run an unknown binary and he is infected its his fault.

Alfred
Parent - - By DamirD81 (***) Date 2007-06-24 18:25
I want YOU all to take a look at this one.
http://f23.parsimony.net/forum50826/messages/168158.htm
Parent - - By Fulcrum2000 (****) Date 2007-06-24 18:29
Parent - By shauntx (*) Date 2007-06-25 01:30
yes i have smae thing that i some how got when i recieved rybka 2.3.2 link but however its much weaker than the previous 2.3.1 i will post pictures of all if anyone is interested i got many more that says weights_used.config and weights_used _annotated.config
but i don't think its a source then the rybka 2.3.2 that i recieved from link too is Rybka v.2.3.2. config x64 only single at first i was like what is this!!! I thought maybe i could tune it or something but didn't look into it because i recieved another link.  but the link i believe was just a beta.
Parent - - By grolich (***) Date 2007-06-25 08:10
I have to concur.
Put it this way:

5000 lines isn't even the right order of magnitude.

As for LINES of code:
I have not seen even a SINGLE line of code in any language.
Just a silly looking human language text...

Even for a config file, it's too bothersome to parse anyway (possible, but annoying).
There are more convenient and standard structures for config files.

Then again, I have not seen a single line of code, in ANY programming language.
Parent - - By Leto (***) Date 2007-06-28 02:55
Why are there so many exe's?  Looks like they might be Worms. 
Parent - By grolich (***) Date 2007-06-29 17:51
That's a definite possibility.
Up Topic Rybka Support & Discussion / Rybka Discussion / Rybka Source Code!

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill