Test versions / private versions wanted for testing purposes

Discussion of chess software programming and technical issues.

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Martin Thoresen
Posts: 1833
Joined: Wed Jun 21, 2006 10:07 pm

Test versions / private versions wanted for testing purposes

Post by Martin Thoresen » Thu Apr 12, 2007 10:47 am

Greetings every engine coder/programmer,

If anyone of you want to have your engine included in the Hammer of Thor
rating list, please send me an email at mordor(at)halden.net . It can be
beta versions of course, but that particular version will stay in the list
for some time to get a proper amount of games.

Hammer of Thor is a tournament time-controlled rating list (40/120-20/60-+30).

Best Regards,
Martin

User avatar
smrf
Posts: 484
Joined: Mon Mar 13, 2006 10:08 am
Location: Klein-Gerau, Germany
Contact:

Re: Test versions / private versions wanted for testing purp

Post by smrf » Thu Apr 12, 2007 4:56 pm

Most testers are having tests done by a GUI. Thus my SMIRF engine (having its own GUI, 8x8 and 10x8) is rarely tested. If you are interested to have it participate your test, send me a mail to get the newest full version MS-167f of SMIRF.

Reinhard.

Martin Thoresen
Posts: 1833
Joined: Wed Jun 21, 2006 10:07 pm

Re: Test versions / private versions wanted for testing purp

Post by Martin Thoresen » Thu Apr 12, 2007 5:09 pm

smrf wrote:Most testers are having tests done by a GUI. Thus my SMIRF engine (having its own GUI, 8x8 and 10x8) is rarely tested. If you are interested to have it participate your test, send me a mail to get the newest full version MS-167f of SMIRF.

Reinhard.
Hello Reinhard and thank you for your reply,

Can your GUI handle Winboard and/or UCI engines? If so, I could
run your engine vs some of the WB/UCI engines in my list.

Best Regards,
Martin Thoresen

User avatar
smrf
Posts: 484
Joined: Mon Mar 13, 2006 10:08 am
Location: Klein-Gerau, Germany
Contact:

Re: Test versions / private versions wanted for testing purp

Post by smrf » Thu Apr 12, 2007 5:22 pm

Can your GUI handle Winboard and/or UCI engines? If so, I could
run your engine vs some of the WB/UCI engines in my list.
It currently would not make sense to support such protocols, as long as there are no 8x8 and 10x8 supporting GUIs using those protocols. Otherwise the real abilities of SMIRF would be skipped and ignored.

The moment there would be a working 18x8 and 10x8 UCI GUI there will be a UCI SMIRF very soon, of course.

Reinhard.

Martin Thoresen
Posts: 1833
Joined: Wed Jun 21, 2006 10:07 pm

Re: Test versions / private versions wanted for testing purp

Post by Martin Thoresen » Thu Apr 12, 2007 5:36 pm

Reinhard,

Sorry for being such a newbie, but unless I can run your engine vs
WB or UCI engines I can't include your engine in the rating list.

I am not an expert on the technical aspects of chess engines,
only configuring them and running them properly.

But I would like to thank you for your interest of course! :)

User avatar
Daniel Mehrmann
Posts: 855
Joined: Wed Mar 08, 2006 8:24 pm
Location: Germany
Full name: Daniel Mehrmann

Protocols controlling the computerchess world ?

Post by Daniel Mehrmann » Fri Apr 13, 2007 1:59 pm

Hi,

You can't handle doing moves manual on a chess GUI ? :shock:

Protocols are needed of course. They given a good standard, no question.
If it's possible, protocols should be used and supported anyway.

But, it can't be true that protocols starting to control and decide if a engine is going to be tested or not. What happend if this decision prevent a success of the product ?

If i'm looking back to old times Mephisto's , Saitek's and first PC chessprograms must be tested manual because lack of protocols.

So, i'm asking:

Why it was possible in old times and it's no longer possible yet ?

Ok, its acceptable that not so much games are possible of course, but for a first rating it should be enought...


just wonder,
Daniel

Stan Arts

Re: Protocols controlling the computerchess world ?

Post by Stan Arts » Fri Apr 13, 2007 4:00 pm

Daniel Mehrmann wrote:Hi,

You can't handle doing moves manual on a chess GUI ? :shock:

Protocols are needed of course. They given a good standard, no question.
If it's possible, protocols should be used and supported anyway.

But, it can't be true that protocols starting to control and decide if a engine is going to be tested or not. What happend if this decision prevent a success of the product ?

If i'm looking back to old times Mephisto's , Saitek's and first PC chessprograms must be tested manual because lack of protocols.

So, i'm asking:

Why it was possible in old times and it's no longer possible yet ?

Ok, its acceptable that not so much games are possible of course, but for a first rating it should be enought...


just wonder,
Daniel
It's also a lot of fun. When you make moves manually (maybe even on a separate chessboard infront of you) you really get into the game and see much more then you otherwise would.
But you need long timecontrols, and you need to be involved with it so you need to have an afternoon or evening of free time to spend.

I hope most people still take out their chessboard in combination with their/an engine sometimes. (it's also great to instantly spot positional weaknesses of your engine, things a 60/40/10 win/loss/draw result don't tell. 10 long games with you watching usually reveal the whole friggin' spectrum of disasters your engine is sensitive to. (like an OTB or CCT tournament or something always does. :? ) It's also nice to play against your own engine from time to time, then you realise your engine is better then engine-testers make you (ok, maybe only me) believe. :)

Stan

User avatar
hgm
Posts: 22588
Joined: Fri Mar 10, 2006 9:06 am
Location: Amsterdam
Full name: H G Muller
Contact:

Re: Test versions / private versions wanted for testing purp

Post by hgm » Fri Apr 13, 2007 4:32 pm

smrf wrote: It currently would not make sense to support such protocols, as long as there are no 8x8 and 10x8 supporting GUIs using those protocols. Otherwise the real abilities of SMIRF would be skipped and ignored.
As you discover now, it would actually make great sense for testing against other engines.... :roll:

There are plenty of engines that can play on an 8x8 board, and now you cannot even test that feature of your engine against those.

User avatar
smrf
Posts: 484
Joined: Mon Mar 13, 2006 10:08 am
Location: Klein-Gerau, Germany
Contact:

Re: Protocols controlling the computerchess world ?

Post by smrf » Fri Apr 13, 2007 4:38 pm

Hi Daniel,

once upon a time there have been testers of chess programs, today there (mostly) are merely owners of computers having some testing automatism run. Maybe, if those tested engines there would play tic-tac-toe instead of chess, those "testers" would not even notice. ;-)

As far as I could experience myself by manually testing (there is no common protocol in 10x8 Chess yet) SMIRF actually is top together with GothicVortex, which is a very specialized program compared to SMIRF, which in contrast is able to play a lot of variants including traditional Chess and Chess960. Thus I prosume SMIRF would have a strength of about 2500 Elo minimum.

Does anyone know, what happened to Ed Trice, the main author of GothicVortex? His website has not been updated for weeks and his online server for Gothic Chess is no longer answering.

Reinhard.

User avatar
smrf
Posts: 484
Joined: Mon Mar 13, 2006 10:08 am
Location: Klein-Gerau, Germany
Contact:

Re: Test ... / private versions ... Why SMIRF is different

Post by smrf » Fri Apr 13, 2007 5:05 pm

hgm wrote:
smrf wrote: It currently would not make sense to support such protocols, as long as there are no 8x8 and 10x8 supporting GUIs using those protocols. Otherwise the real abilities of SMIRF would be skipped and ignored.
As you discover now, it would actually make great sense for testing against other engines.... :roll:

There are plenty of engines that can play on an 8x8 board, and now you cannot even test that feature of your engine against those.
It seems to me that you have misunderstood the main point: SMIRF is not just another 8x8 chess engine. It is a different and more flexible approach to computer chess and compatible rule supersets like Chess960, Capablanca Random Chess, Gothic Chess, Embassy Chess, Optimized Chess, Janus Chess and a lot more. Because I have had to put a lot of time in creating an own matching GUI instead of tuning the engine alone, there are even more reasons, why it is not appropriate to merely test it against 8x8 traditional chess specialized engines for to make up ones mind. At least one should see and inspect the complete approach.

This does not mean, that SMIRF would have made only a very weak entry into the computer chess scene, but it discovers, that the so called testers are not at all interested in new approaches or new abilities of chess programs. SMIRF uses an open and free own modern chess protocol TMCI, which has not had any follower, because of the dominating two protocols, which do not cover more flexible variant approaches and are neither compatible nor sufficient even for Chess960.

Thus I have learned, that there is a need for to compatibly extend the old UCI-2 protocol and to write an own and more flexible GUI based on that. But I have not got the time to do that as quick as it would be needed.

Moreover I am just migrating to Mac OS X, because with the start of Vista I am no longer a follower of Windows. Thus it would be an interesting idea to have SMIRF started in its terminating Windows version, before it will be completely rewritten from the scratch for the Mac. That version is offered to all serious chess program testers.

Regards, Reinhard.

Post Reply