Alaric new version

[Peter Fendrich]

I promised some days ago to publish a new Alaric this weekend.
Well...
I have to delay a few days.

New is a winbord version (in addition to uci) and my own book code.

/Peter

[Tony Thomas]

I promised some days ago to publish a new Alaric this weekend.
Well...
I have to delay a few days.

New is a winbord version (in addition to uci) and my own book code.

/Peter

We would rather have a winboard engine... Worth the wait, I say.

[Volker Pittlik]

I promised some days ago to publish a new Alaric this weekend....

Thank you! I also have decided to pay for another half year.

Volker

[Peter Fendrich]

Great Volker!
This is the only forum I visit regulary.
I like the atmosphere here!
/Peter

[Pradu]

Great Volker!
This is the only forum I visit regulary.
I like the atmosphere here!
/Peter

I agree!

[Volker Pittlik]

Great Volker!
This is the only forum I visit regulary.
I like the atmosphere here!
/Peter

I agree!

Thank you very much!

Although I have to admit that I'm a bit tired at the moment. I'm thinking about to open the forum to other topics like go or a subforum just for talking about everything. But I'm not so sure if I should do. There are bad examles especially for the latter.

Volker

[Pradu]

Thank you very much!

Although I have to admit that I'm a bit tired at the moment. I'm thinking about to open the forum to other topics like go or a subforum just for talking about everything. But I'm not so sure if I should do. There are bad examles especially for the latter.

Volker

I guess you can start up a poll and see what everybody else thinks.

[Patrick Buchmann]

I promised some days ago to publish a new Alaric this weekend.
Well...
I have to delay a few days.

New is a winbord version (in addition to uci) and my own book code.

/Peter

Hi Peter,

A logo in hope better by the specialists

Regards,
Patrick

[Tony Thomas]

Great Volker!
This is the only forum I visit regulary.
I like the atmosphere here!
/Peter

I agree!

Thank you very much!

Although I have to admit that I'm a bit tired at the moment. I'm thinking about to open the forum to other topics like go or a subforum just for talking about everything. But I'm not so sure if I should do. There are bad examles especially for the latter.

Volker

Something similar to CTF can destroy the atmospher of winboard forum..It would attract more people to the forum if you open a subforum were people can talk about any type of AI but like you said opening a forum were anyone can talk about anything is a disaster.

[Daniel Shawul]

(edited out) read the thread below

[Daniel Shawul]

I think that many people me included are getting a little bit tired of chess:) So it is probably good to open up a sub-forum where some chess programmers could migrate to. Been here for over four years now, I think that i know enough* of chess programming. Trying other AI games like Go would be so much fun since you are going to start from Ground 0. One might argue that why not try new algorithms on your chess programme. I don't like this because i am already "brain-washed" with things like alpha-beta is the best there is, are you crazy not using null move...etc
And they are probably right!

Daniel
* - only enough to satisfy my needs:) Are you guys not satisfied with rybka?

P.s:
To Volker: I still couldn't defeat GNUGo. Do you have a database of games with explanation that i could study.

[juan dela cruz]

Go is a great game. Maybe this link can help refresh this new habit.
http://senseis.xmp.net/?LinkCollection

[Volker Pittlik]

I think that many people me included are getting a little bit tired of chess:) So it is probably good to open up a sub-forum where some chess programmers could migrate to

...

P.s:
To Volker: I still couldn't defeat GNUGo. Do you have a database of games with explanation that i could study.

So possibly a Go sub forum (or similar) would be O.K. As long it is not opened post can be made here, mabe with a [Go] in the subject line.

Sorry, I don't have a Go game collection yet. Also I set the computer level to 4 or 5. I guess thiat is weaker . Even with that settings I can't win all the time.

[Tord Romstad]

Hello all,

I hope the Winboard Forum won't die. It remains the friendlies computer chess forum on the Internet. Without the Winboard forum, there wouldn't have been any Glaurung, and I am sure many other amateur authors could say the same about their engines.

I think that many people me included are getting a little bit tired of chess:) So it is probably good to open up a sub-forum where some chess programmers could migrate to.

I agree, but I suggest a broader theme than just computer go. How about a sub-forum about abstract board games in general (othello, shogi, hex, checkers, ...)?

Been here for over four years now, I think that i know enough* of chess programming. Trying other AI games like Go would be so much fun since you are going to start from Ground 0.

My feelings are similar, but I can't quite make up my mind about whether I want to try go or shogi first. Go has a certain abstract mathematical beauty which chess and shogi cannot match, and is definitely the most difficult programming challenge. On the other hand, computer go programs play incredibly dull and ugly go (I don't think I have ever seen an entertaining game by a go program), while shogi easily beats both chess and go for pure excitement of watching computer games. On the plus side for go, the current algorithms are obviously insufficient, which means that I can spend my time hunting for new algorithms with Lisp or Haskell instead of micro-optimizing and fine-tuning decades-old algorithms in C or C++.

Tord

[Michael Sherwin]

Hi Tord,

I have been thinking of trying a new language, so I have been reading about various ones and downloading interpreters and compliers for them. Have not found a Windows Haskell compiler yet, do you know of any? And what is haskell and why would one wish to use it?

Here is what I have downloaded so far:

O'caml

several different lisp:
Corman Lisp
Lisp in a box - GNU common lisp
Steel Bank common lisp - Windows beta

PTL Scheme with the MIT compiler

Win32 forth

Of these I do not like forth and I do not believe that it can be fast, as all values and functions are pushed onto their respective stacks and then removed from the stacks to execute. Plus postfix notation twist my mind in ways that it wasn't meant to be twisted.

Lisp/Scheme seem like extreamly expressive languages at the cost of a little speed and can make doing hard things a little easier. However, as you stated before, chess code is rather straight forwad and there is no benifit to using lisp/scheme.

O'caml is very interesting and does not seem very difficult to learn. Almost lisp like with out the parenthsis. Very C like with very high level features. To tell you the truth I do not know what O'cml is. Description and documentaition is sparse and terse. It looks like a very concise language that is designed so the compiler can do a good job optimizing and so the programmer can approach each task using a consistant methodology.

Any or all of what I have said can be wrong and any comments or corrections are welcome. Keep in mind that I am a speed freak and am not interested in a slow language.

Also Shogi is very chess like, so wouldn't your criticism about using lisp for chess also apply to Shogi?

Thanks,
Mike

[Ron Murawski]

Hi Michael,

If you're looking for a very fast interpreted languge, then look at Euphoria. It's open source. An unusual aspect to the language is that Euphoria can be translated into C and then compiled.

"More powerful than C++, simpler than Basic!"
http://www.rapideuphoria.com/

Ron

[Michael Sherwin]

Hi Michael,

If you're looking for a very fast interpreted languge, then look at Euphoria. It's open source. An unusual aspect to the language is that Euphoria can be translated into C and then compiled.

"More powerful than C++, simpler than Basic!"
http://www.rapideuphoria.com/

Ron

Thanks Ron,

Very interesting!

Mike

[Michael Sherwin]

Hi Michael,

If you're looking for a very fast interpreted languge, then look at Euphoria. It's open source. An unusual aspect to the language is that Euphoria can be translated into C and then compiled.

"More powerful than C++, simpler than Basic!"
http://www.rapideuphoria.com/

Ron

Here is my short report on euphoria.

Euphoria sounds incredibile (on their web site).

It would be simple to learn and would be a good introductory language.

It could be very fun to write code for. The complete freedom of data representation with no worries of ever having to set data types or array sizes and then having to experiance the related bugs would be the reason.

Their claims of speed are impressive, with translated code to C running up to five times faster than when its interpreted. However, in a real world app. the claim of speed will not hold up. The sequence is the main culpret, as it can hold and mix any kind of data as well as grow and shrink with out programmer control. Therefore a simple program that uses mostly simple integers can run very fast. Programs, like chess programs that need lots of arrays and hense must use lots of sequences will be terribly slow.

A case in point is TSCP, done in euphoria, only does 13,107 nodes per second interpreted (TSCP181.exe by Tom Kerrigan does 164,427 nodes per second). Compiled it will only get a small boost, because the C code must still use very slow sequences to do its work. Tom's compile of TSCP is not the fastest anyway, so realisticly tscp in euphoria would be lucky to run at 1/10th the speed, even when translated to C.

[Ron Murawski]

Hi Michael,

If you're looking for a very fast interpreted languge, then look at Euphoria. It's open source. An unusual aspect to the language is that Euphoria can be translated into C and then compiled.

"More powerful than C++, simpler than Basic!"
http://www.rapideuphoria.com/

Ron

Here is my short report on euphoria.

Euphoria sounds incredibile (on their web site).

It would be simple to learn and would be a good introductory language.

It could be very fun to write code for. The complete freedom of data representation with no worries of ever having to set data types or array sizes and then having to experiance the related bugs would be the reason.

Their claims of speed are impressive, with translated code to C running up to five times faster than when its interpreted. However, in a real world app. the claim of speed will not hold up. The sequence is the main culpret, as it can hold and mix any kind of data as well as grow and shrink with out programmer control. Therefore a simple program that uses mostly simple integers can run very fast. Programs, like chess programs that need lots of arrays and hense must use lots of sequences will be terribly slow.

A case in point is TSCP, done in euphoria, only does 13,107 nodes per second interpreted (TSCP181.exe by Tom Kerrigan does 164,427 nodes per second). Compiled it will only get a small boost, because the C code must still use very slow sequences to do its work. Tom's compile of TSCP is not the fastest anyway, so realisticly tscp in euphoria would be lucky to run at 1/10th the speed, even when translated to C.

Hi Michael,

Thanks for your report. I had been in the process of investigating using Euphoria for creating some opening book tools. For myself the complexity of writing a move gen in Euphoria outweighed the convenience of using sequences to hold book positions. It's going to be much easier to re-use my already-debugged move gen instead.

I remember reading that there is a way to speed up Euphoria integer-only and fixed-size sequences by declaring them in certain ways. I don't know if Norman Blais' TSCP translation took advantage of this or not.

If you are looking for an improved compiled language, then D is quite good. But the compiled code is not as fast as Microsoft's so you will never be able to match sheer speed. D is, in my opinion, what the C++ language definition *should* have been.
http://www.digitalmars.com/d/

For chess we need to stick to C or C++ for optimal execution speed. At the highest levels no other languages seem to be able to keep up. I find that fact to be unfortunate.

Ron

[Peter Fendrich]

New Alaric:

http://alaric.fendrich.se/Downloads.html

Have fun!