MP tablebase generator

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

Charles B.

Re: MP tablebase generator

Post by Charles B. »

Charles B. wrote: When compiling with Cygwin these errors show up:

Code: Select all

$ gcc -O2 -o tbgen tbgen.cpp
In file included from tbgen.cpp:321:
tbindex.cpp: In function `int FCheckExistance(char*, int, color)':
tbindex.cpp:4093: error: invalid conversion from `const char*' to `char*'
tbgen.cpp: In function `void VCreateTable(int, int)':
tbgen.cpp:2446: error: invalid conversion from `const char*' to `char*'

OK, It compiles error-free now.


I still need to learn a programming language but for this a shortcut was taken.

Comparing the compiling errors in question from this download:

http://cap.connx.com/chess-engines/new- ... /TBGEN.ZIP


With the corresponding entries (in the files) in this compilable download and the errors were fixed:

http://cap.connx.com/chess-engines/new- ... wtbgen.zip


(Probably not the recommended way of doing it but it worked - this time)

Thank You Again,

Actually, Thank You Very Much!!
User avatar
hgm
Posts: 28387
Joined: Fri Mar 10, 2006 10:06 am
Location: Amsterdam
Full name: H G Muller

Re: MP tablebase generator

Post by hgm »

Dr.Wael Deeb wrote:I am surprised that a skilled programmer like H.G.Muller is not aware of this fact....
Dr.D
It was more meant as a question. Obviously I cannot make any definite statements about a program I have never seen, and only have heard vague rumors about.

I have my own 5-men generator, which takes abot 2 minutes to generate a 5-men like KQRKQ in memory. And I know for sure that this generator is not CPU bound: it is waiting for DRAM access 95% of the time.

From what I heard of others, I had understood that the Nalimov generator is much slower, and I assumed that this was because it was using disk rather than RAM as storage medium. If that is a wrong assumption, I am at a loss as to why it should be so slow.
Omega-three

Re: MP tablebase generator

Post by Omega-three »

hgm wrote:
Dr.Wael Deeb wrote:I am surprised that a skilled programmer like H.G.Muller is not aware of this fact....
Dr.D
It was more meant as a question. Obviously I cannot make any definite statements about a program I have never seen, and only have heard vague rumors about.

I have my own 5-men generator, which takes abot 2 minutes to generate a 5-men like KQRKQ in memory. And I know for sure that this generator is not CPU bound: it is waiting for DRAM access 95% of the time.

From what I heard of others, I had understood that the Nalimov generator is much slower, and I assumed that this was because it was using disk rather than RAM as storage medium. If that is a wrong assumption, I am at a loss as to why it should be so slow.
Its probably slow because its an old chessbase program from 1999. it will only use 1 cpu and you cannot run more than one instance.

It came on my Hiarcs 7.32 disc. It is TBGen.exe v 1.0.0.1

If any one knows of a newer version that is available for download I would like to know.
User avatar
hgm
Posts: 28387
Joined: Fri Mar 10, 2006 10:06 am
Location: Amsterdam
Full name: H G Muller

Re: MP tablebase generator

Post by hgm »

That doesn't sound like an explanation. Old programs speed up just as much when you run them on a faster computer. We are not comparing running time on 1999 hardware with that of a Core 2 Duo. My generator also uses only a single CPU.