Plea for a computerchess beginners forum or FAQ

[mvanthoor]

Hey Guenther!

FAQ with good practices distilled and gotchas noted sounds like a fantastic idea...

The chess programming Wiki should be enough, but apart from looking through it for nostalgic purposes, I think it is often too general with regard of topic discussion, and it seems to have at least some mistakes (for example, a few days ago I found a discussion on transposition tables, where the pseudo-code on CPW was deemed to be wrong). The CPW is a great (historical) resource to get you going, but visiting it will only be a start.

One of the problems with chess programming is not only finding the information, but also finding the CORRECT information; making sure information is actually correct has been the hardest part of getting my engine going, and I assume that, the more advanced the subjects become, the sparser the information about them will be.... with, at the end, the only recourse being to try and read code of known-good engines.

[Guenther]

Hey Guenther!

FAQ with good practices distilled and gotchas noted sounds like a fantastic idea...

The chess programming Wiki should be enough, but apart from looking through it for nostalgic purposes, I think it is often too general with regard of topic discussion, and it seems to have at least some mistakes (for example, a few days ago I found a discussion on transposition tables, where the pseudo-code on CPW was deemed to be wrong). The CPW is a great (historical) resource to get you going, but visiting it will only be a start.

One of the problems with chess programming is not only finding the information, but also finding the CORRECT information; making sure information is actually correct has been the hardest part of getting my engine going, and I assume that, the more advanced the subjects become, the sparser the information about them will be.... with, at the end, the only recourse being to try and read code of known-good engines.

This is far too advanced and it seems you mixed up talkchess with a chess programmers forum?
Actually the chess programmers forum is just a small part of it.

Often beginners, who come to the talkchess (general) forum, neither know what pgn is, nor cmd.
They often don't know about their own hardware (e.g. instruction sets), or don't know what an uci option is, or that not every
'bin' file automatically is a polyglot book ;-) etc.
(I could add much worse examples than the mentioned ones)
I remember a case when I suggested to create a log or debug file for isolating a problem and the answer was sth like
'we are living in the 21st century it would be stupid, if we would need to create such files for solving problems'.

Still they have managed to play a random game on their random GUI, about which they have no clue what it is doing, nor how it should be set up
and wonder why program x always loses to program y (well they say always - sometimes they mean two games with always), when all sources say
program x should be stronger.
Later after lots of questions and tedious help it will be revealed the programs were using more hash than ram available, or one used mpv 10,
or a very strange tc, or one program played with much more threads and so on and on...

[mvanthoor]

This is far too advanced and it seems you mixed up talkchess with a chess programmers forum?
Actually the chess programmers forum is just a small part of it.

Oops... sorry. I basically come to Talkchess for the programming and engine-release part

Often beginners, who come to the talkchess (general) forum, neither know what pgn is, nor cmd.
They often don't know about their own hardware (e.g. instruction sets), or don't know what an uci option is, or that not every
'bin' file automatically is a polyglot book etc.

Not that I'm one for refusing people to participate and learn, but if you're looking for information about -using- chess software, there are (I think) much better chess forums and sites than this one.

(I could add much worse examples than the mentioned ones)
I remember a case when I suggested to create a log or debug file for isolating a problem and the answer was sth like
'we are living in the 21st century it would be stupid, if we would need to create such files for solving problems'.

As I said, I have a feeling that this is mainly a forum where programmers or very technically minded people hang out; not end-users of chess software. Many people who are passionate about sports cars may not actually know how they are built inside, apart from the spec sheets and numbers.

Still they have managed to play a random game on their random GUI, about which they have no clue what it is doing, nor how it should be set up
and wonder why program x always loses to program y (well they say always - sometimes they mean two games with always), when all sources say
program x should be stronger.
Later after lots of questions and tedious help it will be revealed the programs were using more hash than ram available, or one used mpv 10,
or a very strange tc, or one program played with much more threads and so on and on...

If you want to write an FAQ that covers everything that might be wrong with an end-user's setup, be my guest... see you in 2035 Maybe we could have an FAQ for the most common 10 problems or thereabouts, and that would be it... or you'd be writing hundreds of pages/posts.

[Dann Corbit]

http://adamsccpages.blogspot.com/p/cond ... under.html

[Gerd Isenberg]

The chess programming Wiki should be enough, but apart from looking through it for nostalgic purposes, I think it is often too general with regard of topic discussion, and it seems to have at least some mistakes (for example, a few days ago I found a discussion on transposition tables, where the pseudo-code on CPW was deemed to be wrong). The CPW is a great (historical) resource to get you going, but visiting it will only be a start.

Can you please provide a link to the wrong pseudo code in CPW? And a link to the discussion here as well.
I really hope, after more than 10 years of editing CPW, it is a little bit more than nostalgic stuff

Thanks & Regards,
Gerd

[Rebel]

The chess programming Wiki should be enough, but apart from looking through it for nostalgic purposes, I think it is often too general with regard of topic discussion, and it seems to have at least some mistakes (for example, a few days ago I found a discussion on transposition tables, where the pseudo-code on CPW was deemed to be wrong). The CPW is a great (historical) resource to get you going, but visiting it will only be a start.

Can you please provide a link to the wrong pseudo code in CPW? And a link to the discussion here as well.
I really hope, after more than 10 years of editing CPW, it is a little bit more than nostalgic stuff

Thanks & Regards,
Gerd

Best computer chess website on the Internet Gerd.

[mvanthoor]

Can you please provide a link to the wrong pseudo code in CPW? And a link to the discussion here as well.
I really hope, after more than 10 years of editing CPW, it is a little bit more than nostalgic stuff

Don't get me wrong, the CPW is a great resource to get started with things, and to get to know that concepts exist. I also like to browse it for historical information. For actually programming, I don't really use it.

The discussion / post I was referring to was this one, which I found while I was researching why my engine "forgot" how to mate after adding a TT:

http://www.talkchess.com/forum3/viewtop ... 99#p763944

From the code snippet you show here the CPW implementation looks weird indeed. Why does it store an upperbound each time when (val > alpha and val < beta) ?, storing an exact value at the end of the move loop without knowing whether alpha improved or not seems wrong too.

Normally you would do something like this ...

It's actually possible that this has been changed already. I haven't checked.

[Gerd Isenberg]

Can you please provide a link to the wrong pseudo code in CPW? And a link to the discussion here as well.
I really hope, after more than 10 years of editing CPW, it is a little bit more than nostalgic stuff

Don't get me wrong, the CPW is a great resource to get started with things, and to get to know that concepts exist. I also like to browse it for historical information. For actually programming, I don't really use it.

The discussion / post I was referring to was this one, which I found while I was researching why my engine "forgot" how to mate after adding a TT:

http://www.talkchess.com/forum3/viewtop ... 99#p763944

From the code snippet you show here the CPW implementation looks weird indeed. Why does it store an upperbound each time when (val > alpha and val < beta) ?, storing an exact value at the end of the move loop without knowing whether alpha improved or not seems wrong too.

Normally you would do something like this ...

It's actually possible that this has been changed already. I haven't checked.

OK, it is the concrete code of the CPW-engine by Pawel Koziol and Edmund Moshammer.

https://www.chessprogramming.org/CPW-Engine
https://www.chessprogramming.org/CPW-Engine_search

I intend not to change their original code, but indicating issues and errors for didactical reasons with comments inside or descriptions outside the code, including references, i.e. CCC postings. Pawel already fixed something in 2014 and put the CPW-engine on Git-hub, and is of course welcome to edit his original pages.

http://www.talkchess.com/forum3/viewtopic.php?t=54802
https://github.com/nescitus/cpw-engine/ ... search.cpp

Thanks for pointing that out!

Gerd

[Gerd Isenberg]

Best computer chess website on the Internet Gerd.

Thanks for the compliments, Ed. But don't exaggerate

[jefk]

just make a 'high level' thread with text 'computerchess beginners';
or newbies or something like that. it's obviously a difference
whether you are here for 30 yrs or just looking.
my 2 cnts
www.superchess.blogspot.com