Author: Mridul Muralidharan
Date: 02:56:57 01/10/04
Go up one level in this thread
On January 10, 2004 at 05:20:52, Uri Blass wrote:
>On January 10, 2004 at 03:40:13, Mridul Muralidharan wrote:
>
>>On January 09, 2004 at 22:24:46, Uri Blass wrote:
>>
>>>On January 09, 2004 at 21:37:26, Mridul Muralidharan wrote:
>>>
>>>>On January 09, 2004 at 20:48:58, Ed Trice wrote:
>>>>
>>>>>On January 09, 2004 at 20:42:50, Michel Langeveld wrote:
>>>>>
>>>>>>Did you only use the bitboard code and SEE for making Gothic Vortex?
>>>>>
>>>>>You cannot download Crafty and just change a few parameters and have a fully
>>>>>working Gothic Vortex program.
>>>>>
>>>>>But, some entire source files were ported directly for use in the program.
>>>>>
>>>>>I did ask Dr. Hyatt about this, of coursse.
>>>>
>>>>So Dr Hyatt does not need to modify crafty to play Gothic chess - you seem to
>>>>have done it already for him !
>>>>Will you be making this open source like crafty ? ;)
>>>>
>>>>Ofcourse I know the answer - but the gall to use crafty code almost completely
>>>>and then charge others for "$1 license" !!!
>>>
>>>I also do not like the $1 license but I think that the only person that can
>>>complain if there is a problem is Hyatt and not you.
>>>
>>
>>Hmm , if you see injustice getting done , I think from your above response , you
>>would keep quiet ? - ok , in that case your opinion is fair.
>>
>>I dont -
>>This person charges $1 , then he gets into long arguements with Bob , challenges
>>him to write a Gothic Chess version , insults someone like him who has worked
>>for 20+ years in this feild by giving him a "$1 free license for 1 year" , and
>>in end , I see that most of this gothic chess program is crafty anyway !
>>
>>Bob Hyatt has every right to complain - that does not mean I cant.
>
>I do not think that Ed trice insulted Bob Hyatt.
>I think that you insult Ed trice for no reason.
>
It is true that I consider his idea good but his patent frivolous , but no
insults.
I am especially peeved off by his "mightier than thou" attitude by spelling out
all that "$1" , etc stuff - and especially trying to imply threat - well in most
cases openly threatening.
He should remember that most programmers here are very responsible people ,
handling this and other responsible jobs - and these kinds of coercive threats
are not in the right spirit.
Mentioning about a "$1" fee is good - but not stuff like I will take you to
court , etc repeatedly - everyone here understands what patent infringement
means.
And as usual , you are ever ready to jump to conclusions ("think" ????)
To Ed Trice : "I think that you insult Ed trice for no reason."
This is Uri's thinking (so ignore it) , not mine - I have/had no intention of
insulting you or your work.
>>
>>Did I touch a raw nerve here ?
>>Ah well , tscp author Tom Kerigham also might not have had any problems also I
>>guess ? ;)
>
>No
>
>Movei did not start from tscp.
>I started from a legal move generator(something that tscp never had).
>
>Movei also never had similiar evaluation to tscp.
hehehe :)
>
>>
>>>>
>>>>The earlier impression I got from reading your request was - you are "borrowing"
>>>>some small pieces which are tough for you to develop - looks like it was not
>>>>using crafty bits to write your program , but modifying it superficially to play
>>>>your new variant.
>>>>No wonder you make claims about its playing strength !
>>>
>>>Your impression is not relevant.
>>>The limit between borrowing some pieces of Crafty and modifying Crafty is not
>>>clear.
>>
>>Indeed - for you changing a few comments , chaning a few variables , etc would
>>mean new program.
>
>No but it is impossible to write a program that plays 8*10 game with different
>pieces by only changing a few varaibles.
>
You are joking right ?
How much change do you expect ?
a) movegenerator change - maybe with bitboards slightly more involved - but
otherwise say 1-3 day (in case programmer is feeling especially lazy:) ) ?
To start anyone off in bitboards , there were already discussions on this here
at CCC ...
b) eval , most of which will remain same - just need to add support for eval for
new pieces - as "author of game" he will have quiet good idea of the patterns
involved - atleast you can make quiet an educated guess here ... maybe say
another 1 - 3 days. (ofcourse , it might take years to get all this tuned up !)
So crafty playing this variant is is like 2 - 6 days.
"impossible" -> ignorance or incompetence or not given thought to problem ??
I hope you thought about it before writing this post ....
Understanding basics helps , 8x8 , 16x16 , 10x8 , etc - all are just different
board sizes - nothing special with any of those !
Please dont consider that the bitboard approach for move generation works only
for 8x8 boards !!!! It can be modified for any of these.
>
>>Most others consider the ideas behind the program , its tuned paramenters - not
>>the syntax , indentation , and other irrelevent things as the crux of the
>>program.
>>You take those , you have taken the core of the program !
>
>tuned parameters?
Now you are totally joking.
What is eval other than tuning ?
Write patterns , tune them - various mobility , pscq values , pawn structure ,
etc ,etc - all are tuning issues , nothing else !
I would suggest you to read some papers on DeepThought and KnightCap (I got the
name right here right ?) where they dont hand tune eval paramenters...
>evaluation of Crafty is for chess and not for a different game so nothing is
>tuned in crafty for a different game.
Refer above ....
BTW , I just got crafty 19.9.
If you look at crafty eval.c , it is structured in such a way that you can
easily add eval parameters for more pieces .... like the archbishop , etc that
gothic chess has.
>
>>
>>>
>>>I believe that the main reason that Hyatt gave permission is the fact that it is
>>>not a chess playing program.
>>>
>>
>>????
>>You dont consider what as a "not a chess playing program." here ??
>>crafty ? , this ersion that Ed Trice has released ?
>>Ed Trice's program is also a chess playing program - albeit a new variant that
>>he has created.
>
>No
>
>The game is not chess because the board is different and there are new pieces.
>
Similar ideas for shatranj , losers chess , frc , etc i guess ?
"Game is not chess because -
a) different starting position
b) different castling rules
c) different game rules
d) different piece movement rules , blah , blah"
I suggest you to look at what Gothmog is going to support according to Tord here
: http://www.talkchess.com/forums/1/message.html?341304
Are you going to call this also "not chess" ?
Even the basic board is different - Hexagonal !
Even Ed Trice is not claiming that it is another game !
This is a variant of chess - simple.
>>
>>>It is clear that you need to change the move generator and a lot of things when
>>>you have 8*10 board.
>>
>>Indeed you have to - we are not talking about anything very stupid here.
>>But basic ideas are the same.
>>If I take crafty , sit with a few GMs , handtune the eval parameters better ,
>>add a few more patterns - then I guess in your definition , it is a new program.
>
>It is a different case because it is a chess playing program.
>
IT IS SAME PROGRAM !
Different parameters , diferent patterns , but same nonetheless.
What if , in above example , I do all this to support say FRC ? or Losers chess
? or Gothic chess ?( ;) ) or Shatranj ?
Will your opinion change then ?
>
>>No in most of the people here - it is crafty , just another modification of it.
>>I think lots of people already do similar stuff like this - Mike Byrne , etc
>
>Mike Byrne said that he is not a programmer and he does not know to change
>Crafty to a ghotic playing program.
I think he has a SE version of crafty where he hand tunes paramenters to make it
play better/different.
Others also create crafty personalities by changing eval and other parameters ,
that crafty exposes , etc.
Just because you ripped apart the code , and reconstructed it in a different
order does not make it different program !
>
>It is not something easy to do it.
>
When in the end you make money out of it , "easy" does not come into picture.
By the way , made it to the magical 2000 lines ? or is that also not "easy" ??
>I do not think that you have the right to speak for most people.
Neither do you.
I was voicing my opinion , definitely not Dr. Hyatt's and in my opinion how he
can interpret his permission.
Please pay attention to details ...
>The only problem that I see is trying to protect an opening position by a patent
>and I think that it is wrong even if without taking parts from crafty.
Yes , nothing wrong with his patent (leaving aside my opinions).
Please refer to my earlier post to atleast understand what you are trying to
post against !
>
>I do not see problems of taking parts from crafty when it is allowed by the
>author.
>
With the condition that he gives it back ... read license.
If he is going to give it back , Dr Hyatt does not need to do anything to
support this ! Since Ed Trice will be releasing this anyway.
>I know the part that is written in main.c but Bob hyatt made it clear that this
>part was not intended for game of 8*10 with other pieces.
Please read his and the license again.
It is indeed possible for Dr Hyatt , as author , to waive this for Ed Trice's
request , but from what he posted earlier it is not intended that way in an
obvious manner.
>
>Uri
Regards
Mridul
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.