Win32 based "professional" chess software
Chris Smith
Anyone aware of high end chess programs that are Win32 based? And
possibly mulitithreaded as well to take advantage of multiple
processors when run under Windows NT. Any coming soon?
Thanks.
Chris
Regards,
Chris
Komputer Korner
I am only aware of 8 commercial chess engine programs that are able to
run in WIN NT.
Genius 4, Extreme/Fritz4, Shredder, Corel, Champ, CM5000 , Diep and
Virtual Chess.
Tascbase with the King engine can but you have to do some prior setup.
The King is in
CM5000 anyway. Of these only Shredder, Corel, CM5000 and the Champ are
32 bit.
I am not sure about Diep because I don't have the program.
Crafty with Winboard is a free 32 bit program that runs in WIN NT. Gnu
Chess and
Arasan also are free and run in WIN NT.
As far as multithreading is concerned, Shredder is multi threaded but
it still runs only in
1 processor according to the documentation.
Chris Smith wrote:
--
Komputer Korner
The inkompetent komputer
Note that my true email is still kor...@netcom.ca
I don't often check the email of the sympatico address.
Komputer Korner
I forgot to mention that Virtual Chess is also 32 bit.
Komputer Korner wrote:
> I am only aware of 8 commercial chess engine programs that are able
> to
> run in WIN NT.
> Genius 4, Extreme/Fritz4, Shredder, Corel, Champ, CM5000 , Diep and
> Virtual Chess.
> Tascbase with the King engine can but you have to do some prior setup.
>
> The King is in
> CM5000 anyway.
--
Stefan Meyer-Kahlen
On Tue, 17 Jun 1997 22:28:30 +0000, Komputer Korner <kor...@netcom.ca>
wrote:
>As far as multithreading is concerned, Shredder is multi threaded but
>it still runs only in
>1 processor according to the documentation.
If you have two processors and WinNT the Shredder engine will run on
one processor while the graphical interface will run on the other,
but that's only a small improvement as far as the speed of the
engine is concerned.
Stefan
Komputer Korner
Hi Stefan,
How Small an improvement?
Stefan Meyer-Kahlen wrote:
--
Stefan Meyer-Kahlen
On Wed, 18 Jun 1997 20:30:08 +0000, Komputer Korner <kor...@netcom.ca>
wrote:
>Hi Stefan,
>
> How Small an improvement?
I don't know as I only have one CPU in my system :-(
It can't be a lot because normally when you play chess you don't move
or resize windows or open menus very often, so the interface thread is
mainly waiting for input and updating the clock once every second.
Stefan
Larry MacNeill
Komputer Korner wrote:
>
> I am only aware of 8 commercial chess engine programs that are able to run in WIN NT.
> Genius 4, Extreme/Fritz4, Shredder, Corel, Champ, CM5000 , Diep and
> Virtual Chess.
> Tascbase with the King engine can but you have to do some prior setup.
> The King is in
> 32 bit.
The installation program of Chessmaster 5000 refuses to run under NT.
Chessmaster 5000 also refuses to run. After I installed service pack 3
(which includes DirectX 3.0) I tried again with no change. The
technical support folks at Mindscape say that the program uses API calls
available under Win95 but not under NT and that the problem is more than
having DirectX available. I think I saw a simple patch of cm5000.exe
that allows it to run but without at least some of its custom controls
(contributed by Komputer Korner, maybe?) So I have a dual-boot computer
at home with Win95 installed for the sole purpose of running this chess
program.
I don't see much value in the custom controls (or the 3D decorator sets)
and I'd gladly buy a version of the program with standard controls for
the sake of portability. Or maybe buy another program. Can anyone rank
these programs on various attributes (strength, bells & whistles,
flexibility, portability, teaching facilities)?
I don't see Chessmaster mentioned here very much. Why is that?
Komputer Korner
Well, even though we respect CM5000's engine as fairly strong and the
bells and whistles
very good, the 2D boards and screen layout in CM5000 are bad. Have you
tried the following?
With a hex editor search in file CM5000.EXE for the byte pattern (in
hex):
83 bd 88 fe ff ff 02
and change the "02" to something different (if you change it to 01,
Chessmaster won't work under Windows 95 anymore).
That's it.
The offset should be:
0xcf11 for version 1.0.0
0xd3a5 for version 1.0.2
Here are a few notes:
- You have to install Chessmaster 5000 under Windows 95.
- Sometimes the title screen hangs eating up all the CPU. Simply remove
or rename title.exe in your CM5000 directory.
- You may have to install the CM5000.TTF font manually under NT.
- Radio buttons and check boxes won't display.
Larry MacNeill wrote:
>
>
> The installation program of Chessmaster 5000 refuses to run under NT.
> Chessmaster 5000 also refuses to run. After I installed service pack
> 3
> (which includes DirectX 3.0) I tried again with no change. The
> technical support folks at Mindscape say that the program uses API
> calls
> available under Win95 but not under NT and that the problem is more
> than
> having DirectX available. I think I saw a simple patch of cm5000.exe
> that allows it to run but without at least some of its custom controls
>
> (contributed by Komputer Korner, maybe?) So I have a dual-boot
> computer
> at home with Win95 installed for the sole purpose of running this
> chess
> program.
>
> I don't see much value in the custom controls (or the 3D decorator
> sets)
> and I'd gladly buy a version of the program with standard controls for
>
> the sake of portability. Or maybe buy another program. Can anyone
> rank
> these programs on various attributes (strength, bells & whistles,
> flexibility, portability, teaching facilities)?
>
> I don't see Chessmaster mentioned here very much. Why is that?
--
Komputer Korner
The inkompetent komputer
If you see a 1 in my email address, take it out.
Scott Rynd
Does anyone have an opinion on Sierra's PowerChess software? (I know
that is a loaded question...)
Scott
> ----------
> From: Larry MacNeill[SMTP:Larry.M...@UCHSC.edu]
> Posted At: Thursday, June 19, 1997 2:57 PM
> Posted To: ics
> Conversation: Win32 based "professional" chess software
> Subject: Re: Win32 based "professional" chess software
>
> Komputer Korner wrote:
> >
> > I am only aware of 8 commercial chess engine programs that are able
> to run in WIN NT.
> > Genius 4, Extreme/Fritz4, Shredder, Corel, Champ, CM5000 , Diep and
> > Virtual Chess.
> > Tascbase with the King engine can but you have to do some prior
> setup.
> > The King is in
> > CM5000 anyway. Of these only Shredder, Corel, CM5000 and the Champ
> are
> > 32 bit.
>
brucemo
Stefan Meyer-Kahlen wrote:
>
> On Wed, 18 Jun 1997 20:30:08 +0000, Komputer Korner <kor...@netcom.ca>
> wrote:
>
> >Hi Stefan,
> >
> > How Small an improvement?
>
> I don't know as I only have one CPU in my system :-(
> It can't be a lot because normally when you play chess you don't move
> or resize windows or open menus very often, so the interface thread is
> mainly waiting for input and updating the clock once every second.
Sounds miniscule.
bruce
Chris Smith
Thanks to all for the assistance. It would seem I'm left with trying
to determine the relative strength of the non-multitasking 32 bit
commercial programs. Anyone have that info?
How about Shredder? Haven't turned it up in a search.
Regards,
Chris
Regards,
Chris
Stefan Meyer-Kahlen
On Fri, 20 Jun 1997 17:29:43 -0700, brucemo <bru...@seanet.com>
wrote:
I don't know the exact meaning of miniscule and have no dictionary
handy, but I am afraid you are right.
Stefan
Komputer Korner
There is one other slight problem, WIN NT 4 does not support asymmetric
processing.
Stefan Meyer-Kahlen wrote:
--
Komputer Korner
Shredder is from 100-150 points weaker than the top programs. I said
150 points because
Hiarcs 6 is 50 points stronger than all the others. Also Shredder is
weaker relatively on speed
chess. its strength is slow chess, but at that level it still can't
match the top programs.
However it has fantastic 4 and 5 man endgame CDROMS that come with the
program..
Chris Smith wrote:
--
Komputer Korner
32 bit chess programs (windows based) strength
1) Genius 5
2) CM5000
3) Schach Champ
4) Virtual Chess Win95 version
5) Power Chess
6) Shredder
Komputer Korner
Whoops, Genius 5 is only 16 bit but since it seems to multitask quite
well with other programs
and since Genius 4 runs well in WIN NT 4, the only drawback to those 2
programs as far
as not being a 32 bit program is concerned is that you can't use long
filenames. Contrast
this with CM5000 which hogs all the CPU and won't let you multitask. I
think that is too much
of a price to pay just for having long filenames. CM5000 was written
very poorly as regards
multitasking. Also if you use Crafty with Winboard you have a 32 bit
program. So the
revised list should be:
1) CM5000
2) Schach Champ
3) Virtual Chess
4) Power Chess
5) Shredder
6) Crafty with Winboard
Honourable mention to Genius 5 the strongest of all Windows programs
that multitasks quite
well. Hiarcs6 engine in Fritz/Extreme is strongest windows program but
does not multitask with other
programs.
Komputer Korner wrote:
> 32 bit chess programs (windows based) strength
>
> 1) Genius 5
> 2) CM5000
> 3) Schach Champ
> 4) Virtual Chess Win95 version
> 5) Power Chess
> 6) Shredder
>
>
Komputer Korner
I didn't mention Corel Chess because it is on my KRASH LIST, but to be
technical Corel would also belong on the 32 bit list, because the CD has
2 programs on it, the version for Win 3.1 and another version for WIN
95.
It would have been last in strength anyway.
Komputer Korner wrote:
> Whoops, Genius 5 is only 16 bit but since it seems to multitask quite
> well with other programs
> and since Genius 4 runs well in WIN NT 4, the only drawback to those 2
>
> programs as far
> as not being a 32 bit program is concerned is that you can't use long
> filenames. Contrast
> this with CM5000 which hogs all the CPU and won't let you multitask. I
>
> think that is too much
> of a price to pay just for having long filenames. CM5000 was written
> very poorly as regards
> multitasking. Also if you use Crafty with Winboard you have a 32 bit
>
>
--
Andy Monks
Chris Smith (smy...@ic.net) wrote:
: Anyone aware of high end chess programs that are Win32 based? And
: possibly mulitithreaded as well to take advantage of multiple
: processors when run under Windows NT. Any coming soon?
: Thanks.
: Chris
I know Robert Hyatt, the author of crafty is going to be making crafty
utilize multiple processors. However, this will be only under unix. Time
to load linux maybe!.
Not sure when Robert will have finished this version (perhaps he'll reply)
--
Best Regards,
Andy Monks HP-UX WTEC Engineer
Hewlett-Packard Ltd. Email andy_...@hp.com
-------------------------------------------------------------------
I am Pentium of Borg. Division is futile, you will be approximated.
De' ll' Sovlu'DI' chaq Do'Ha' or Knowledge of useful information
may be unfortunate.
In 1492, an unknown and geographically incompetent sailor from Italy
would lead a small Spanish fleet to the Americas.
Stefano Gemma
Now that my 32bit multi-thread program works, I've found that it is 20%
slower than the single thread, do to the different memory addressing
required for handling global data.
Before i was handling global data this way:
1) add Value, delta
now i should use this:
2) add [ebp.Value], delta
This seems to be slower than before but maybe there's some mistake in the
implementation. Somewhere i've loose some cycle because i haven't the ebp
register free anymore, maybe i can do better.
GoodBye!
brucemo
Komputer Korner wrote:
>
> I didn't mention Corel Chess because it is on my KRASH LIST,
> It would have been last in strength anyway.
I can't believe it's weaker than Microsoft Chess, but to be
fair, that's 16-bit and they give it away now.
bruce
Stefan Meyer-Kahlen
>I didn't mention Corel Chess because it is on my KRASH LIST, but to be
>technical Corel would also belong on the 32 bit list, because the CD has
>
>2 programs on it, the version for Win 3.1 and another version for WIN
>95.
>It would have been last in strength anyway.
Oh, so Shredder is at least stronger than Corel Chess.
Stefan
Chris Smith
mclane
Komputer Korner <kor...@netcom.ca> wrote:
32 bit chess programs (windows based) strength
1) Genius 5
mclane:
Genius is 32 Bit ? I thought 16 Bit engine.
2) CM5000
3) Schach Champ
4) Virtual Chess Win95 version
5) Power Chess
6) Shredder
mclane:
I would say Shredder is stronger than PowerChess ! Or ?
brucemo
Komputer Korner wrote:
>
> 32 bit chess programs (windows based) strength
>
> 1) Genius 5
> 3) Schach Champ
> 4) Virtual Chess Win95 version
> 5) Power Chess
> 6) Shredder
What is Schach Champ? I think I saw this posted, but I don't remember.
How did you arrive at this list? Is this SSDF or KK intuition?
bruce
Robert Hyatt
mclane (mcl...@prima.ruhr.de) wrote:
: Komputer Korner <kor...@netcom.ca> wrote:
: 32 bit chess programs (windows based) strength
: 1) Genius 5
: mclane:
: Genius is 32 Bit ? I thought 16 Bit engine.
: 2) CM5000
: 3) Schach Champ
: 4) Virtual Chess Win95 version
: 5) Power Chess
: 6) Shredder
: mclane:
: I would say Shredder is stronger than PowerChess ! Or ?
The above is wrong anyway. Shredder won the Jakarta event. Virtual
Chess was there. Wchess (power chess) wasn't, not sure about the others
either...
It certainly doesn't go at the bottom...
brucemo
Komputer Korner wrote:
>
> There is one other slight problem, WIN NT 4 does not support asymmetric
> processing.
What do you mean?
bruce
John
I don't believe Genius is 32 bit since it runs under Win 3.1 as well as 95
and from what I gather
it doesnt perform well on a Pentium Pro....
brucemo <bru...@seanet.com> wrote in article <33B187...@seanet.com>...
> Komputer Korner wrote:
> >
> > 32 bit chess programs (windows based) strength
> >
> > 1) Genius 5
> > 2) CM5000
> > 3) Schach Champ
> > 4) Virtual Chess Win95 version
> > 5) Power Chess
> > 6) Shredder
>
brucemo
Komputer Korner wrote:
>
> Whoops, Genius 5 is only 16 bit but since it seems to multitask quite
> well with other programs
> and since Genius 4 runs well in WIN NT 4, the only drawback to those 2
> programs as far
> as not being a 32 bit program is concerned is that you can't use long
> filenames. Contrast
> this with CM5000 which hogs all the CPU and won't let you multitask. I
> think that is too much
> of a price to pay just for having long filenames. CM5000 was written
> very poorly as regards
> multitasking. Also if you use Crafty with Winboard you have a 32 bit
> revised list should be:
Ok, what are you doing here? First, you make Genius an "honorary"
32-bit program because it "multitasks well", but even though CM5K, which
is a 32-bit program according to you (I don't know if it is or not),
*does not* multitask well, it is at the top of your list.
So what is this a list of?
The thing these applications all have in common is that they are native
Windows applications.
But you missed Extreme Chess here.
> 1) CM5000
> 2) Schach Champ
> 3) Virtual Chess
> 4) Power Chess
> 5) Shredder
> 6) Crafty with Winboard
How did you generate this ranking?
bruce
Komputer Korner
I know I know. It was a finger fehler that was corrected on the next
post.
John wrote:
> I don't believe Genius is 32 bit since it runs under Win 3.1 as well
> as 95
> and from what I gather
> it doesnt perform well on a Pentium Pro....
>
>
--
Komputer Korner
I corrected this in another post. It was a fingerfehler. Of course
Genius still is a
16 bit program. What fooled me is that Genius 4 runs directly in Win NT
4.
mclane wrote:
> Komputer Korner <kor...@netcom.ca> wrote:
>
> 32 bit chess programs (windows based) strength
>
> 1) Genius 5
>
> mclane:
> Genius is 32 Bit ? I thought 16 Bit engine.
>
> 2) CM5000
> 3) Schach Champ
> 4) Virtual Chess Win95 version
> 5) Power Chess
> 6) Shredder
>
> mclane:
> I would say Shredder is stronger than PowerChess ! Or ?
--
Komputer Korner
Stefan said that Shredder can use 2 processors and take advantage of
dividing the
work between IO and Chess processing. The problem is that in WIN NT 4,
the OS will decide the work allocation. You can't define it as a
programmer. The
argument was would Shredder be faster with 2 processors. Answer from
Stefan was
yes but a small amount. So since Shredder isn't a parallel chess engine,
this idea
wouldn't work in WIN NT 4.
brucemo wrote:
--
Komputer Korner
Extreme is not 32 bit and it doesn't multitask well.
I have played with them all, so that is the basis of ranking them. Does
Thorsten disagree with me? He has played with them all as well. If
Thorsten has a different order I will defer to his judgement.
brucemo wrote:snipped
> So what is this a list of?
>
> The thing these applications all have in common is that they are
> native
> Windows applications.
>
> But you missed Extreme Chess here.
>
> > 1) CM5000
> > 2) Schach Champ
> > 3) Virtual Chess
> > 4) Power Chess
> > 5) Shredder
> > 6) Crafty with Winboard
>
> How did you generate this ranking?
>
mclane
Komputer Korner <kor...@netcom.ca> wrote:
>Extreme is not 32 bit and it doesn't multitask well.
>I have played with them all, so that is the basis of ranking them. Does
>Thorsten disagree with me? He has played with them all as well. If
>Thorsten has a different order I will defer to his judgement.
Oooops. To much honor.
From my point of view Virtual is higher than 2, maybe 1st rank.
Also Shredder should be better than Powerchess.
From features OF COURSE and playing strength,
The power chess features are not for SERIOUS players. The whole
programs seems to be designed for novice english language speaking
kids.
1.Virtual Win 95
2. CM5000
3. Schach Champ (I don't know much about this unique austrian chess
program, chrilly has shown it in AEGON, and I have seen some AVI files
showing chrilly in front of the PRATER telling some stupid sentences
like:
This is the prater.
The comments were read, not really spoken. Chrilly , please never try
to be actor or professional reader ! It sound very boing because you
don't raise or decrease your voice and the gaps are somehow
artificial.
4.Shredder
5.Crafty
6.Power chess
>> But you missed Extreme Chess here.
>>
>> > 1) CM5000
>> > 2) Schach Champ
>> > 3) Virtual Chess
>> > 4) Power Chess
>> > 5) Shredder
>> > 6) Crafty with Winboard
>>
>> How did you generate this ranking?
Good question. Anybody has its own ranking.
Mine is obviously different than KK's. I can live with this.
Thanks KK, but i prefer you create these lists.
>>
>> bruce
Komputer Korner
I agree with you except that the Power Chess engine is fairly strong. I
think it would
be a very close match between Shredder and Power Chess. It is the
W-Chess
engine 1995 version and is 26th on the SSDF list. And Schach is Nimzo
3.5 or later.
mclane wrote:
> Komputer Korner <kor...@netcom.ca> wrote:
>
> >Extreme is not 32 bit and it doesn't multitask well.
> >I have played with them all, so that is the basis of ranking them.
> Does
> >Thorsten disagree with me? He has played with them all as well. If
> >Thorsten has a different order I will defer to his judgement.
>
> Oooops. To much honor.
> From my point of view Virtual is higher than 2, maybe 1st rank.
> Also Shredder should be better than Powerchess.
> From features OF COURSE and playing strength,
> The power chess features are not for SERIOUS players. The whole
> programs seems to be designed for novice english language speaking
> kids.
>
> 1.Virtual Win 95
> 2. CM5000
> 3. Schach Champ (I don't know much about this unique austrian chess
>
>
> 4.Shredder
> 5.Crafty
> 6.Power chess
Robert Hyatt
Komputer Korner (kor...@netcom.ca) wrote:
: >
: > 1.Virtual Win 95
: > 2. CM5000
: > 3. Schach Champ (I don't know much about this unique austrian chess
: >
: >
: > 4.Shredder
: > 5.Crafty
: > 6.Power chess
Sorry to disagree, but I'm not convinced Crafty is better than Dave's
program... It looks very strong to me (at least wchessx on ICC)...
And I still think shredder belongs above the others above since it
finished ahead of all but cm5000 in a 12 round (nearly round-robin)
event...
Komputer Korner
These were Thorsten's rankings. However, they also look reasonable.
W-Chess X
is probably a lot stronger than the old version of W Chess which is in
Power Chess.
Crafty is hard to get a handle on. The strength seems to change greatly
from version
to version. It looks like a world beater in the Korrespondence Kup, but
version 12.x
seems to disappoint. As for Shredder's ranking, it plays much better at
slow chess in
comparison to the other programs than blitz. We need some human vs Win
32 program
tournaments.
Robert Hyatt wrote:
--
brucemo
Robert Hyatt wrote:
> Sorry to disagree, but I'm not convinced Crafty is better than Dave's
> program... It looks very strong to me (at least wchessx on ICC)...
It is very good at 5 0, at least.
> And I still think shredder belongs above the others above since it
> finished ahead of all but cm5000 in a 12 round (nearly round-robin)
> event...
What event was this?
bruce
Robert Hyatt
Komputer Korner (kor...@netcom.ca) wrote:
: These were Thorsten's rankings. However, they also look reasonable.
: W-Chess X
: is probably a lot stronger than the old version of W Chess which is in
: Power Chess.
: Crafty is hard to get a handle on. The strength seems to change greatly
: from version
: to version. It looks like a world beater in the Korrespondence Kup, but
: version 12.x
: seems to disappoint. As for Shredder's ranking, it plays much better at
: slow chess in
: comparison to the other programs than blitz. We need some human vs Win
: 32 program
: tournaments.
: Robert Hyatt wrote:
: > Komputer Korner (kor...@netcom.ca) wrote:
: > : >
: > : > 1.Virtual Win 95
: > : > 2. CM5000
: > : > 3. Schach Champ (I don't know much about this unique austrian
: > chess
: > : >
: > : >
: > : > 4.Shredder
: > : > 5.Crafty
: > : > 6.Power chess
: >
: > Sorry to disagree, but I'm not convinced Crafty is better than Dave's
: > program... It looks very strong to me (at least wchessx on ICC)...
: >
: > And I still think shredder belongs above the others above since it
: > finished ahead of all but cm5000 in a 12 round (nearly round-robin)
: > event...
: --
: Komputer Korner
: The inkompetent komputer
: If you see a 1 in my email address, take it out.
: Note that my true email is still kor...@netcom.ca
: I don't often check the email of the sympatico address.
12.6 seems to be better. It just set an all-time high for blitz on
ICC last night, 2989. It's not *that* good, but it seems to be playing
pretty decently of late. We'll see...
Rolf Tueschen
Komputer Korner <kor...@netcom.ca> wrote:
>These were Thorsten's rankings. However, they also look reasonable.
>W-Chess X
>is probably a lot stronger than the old version of W Chess which is in
>Power Chess.
>Crafty is hard to get a handle on. The strength seems to change greatly
>from version
>to version. It looks like a world beater in the Korrespondence Kup, but
>version 12.x
>seems to disappoint. As for Shredder's ranking, it plays much better at
>slow chess in
>comparison to the other programs than blitz. We need some human vs Win
>32 program
>tournaments.
Yes, agreed, holy korner. But for all ------
we need a komputer who writes readable texts within a certain continual
presentation.
What does that mean? Oh, it's quite simple. Look at your post above korner.
That's exactly how it appears on our displays. While writing it you surely
didn't break each line with a surplus of one and a half words into the next
line, no?
Look for a short comparison to Bob's fluency below. Do you detect the
difference?
BTW, also CW uses this wrong format for his posts. So that all our good style
is transformated into this sort of komputer-corrupted-fluency. It's perhaps a
virus of unknown sort? --- CW? KK? --- Please fix this problem.
Rol f <The aid of the expert human>
Komputer Korner
Watch out there Bob. Some say that God plays at 3000 level. Your program
is getting
dangerously close to it. And he who creates a GOD is tampering with the
ALMIGHTY.
I know I know, there were GOD threads a while back.
I can't remember what Elo everybody finally settled on. If somebody
remembers please
post and I will stand corrected in my admonition of Bob.
Robert Hyatt wrote:
> 12.6 seems to be better. It just set an all-time high for blitz on
> ICC last night, 2989. It's not *that* good, but it seems to be
> playing
> pretty decently of late. We'll see...
--
mclane
hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
>Jakarta. Nimzo, crafty and virtual chess were all there. Shredder finished
>first of course...
If you would HAVE Virtual chess at home, you would not say stuff like
the above.
Virtual is very strong. We can let shredder play against Power-chess,
if you want.
Also I have Nimzo Jakarta at home. We can let Virtual play against
Nimzo 3.5 if you want.
And more than 1 game.
Jakarta was nice, but to few games.
At home we have more time!
Charles Kamaras
I have several models available.
I'm looking for the old Prestige Challenger by Fidelity,or the Montreux
by Saitek.
brucemo
Robert Hyatt wrote:
> Jakarta. Nimzo, crafty and virtual chess were all there. Shredder finished
> first of course...
Aha. I was trying to figure out at what event CM5K finished higher than
Shredder, but now I realize you meant that CM5K wasn't there.
To be fair, Virtual Chess had a wimpier machine.
bruce
Komputer Korner
I do press a return on each line. It is the stupid Mozilla news reader
formatting
program that is used in Netscape. does anybody else have any ideas on
this?.
Rolf Tueschen wrote:
> Yes, agreed, holy korner. But for all ------
>
> we need a komputer who writes readable texts within a certain
> continual
> presentation.
>
> What does that mean? Oh, it's quite simple. Look at your post above
> korner.
> That's exactly how it appears on our displays. While writing it you
> surely
> didn't break each line with a surplus of one and a half words into the
> next
> line, no?
>
> Look for a short comparison to Bob's fluency below. Do you detect the
> difference?
>
> BTW, also CW uses this wrong format for his posts. So that all our
> good style
> is transformated into this sort of komputer-corrupted-fluency. It's
> perhaps a
> virus of unknown sort? --- CW? KK? --- Please fix this problem.
>
--
Rolf Tueschen
brucemo <bru...@seanet.com> wrote:
>Robert Hyatt wrote:
>> Jakarta. Nimzo, crafty and virtual chess were all there. Shredder finished
>> first of course...
>Aha. I was trying to figure out at what event CM5K finished higher than
>Shredder, but now I realize you meant that CM5K wasn't there.
It's intereting, young bruce, cause I wasn't there and therefore had to think
for myself and got the same result long before you asked for the second time.
How come? :)
Charles Sorgie
likely the portable models, maybe the tabletops. What about Saitek vs.
Novag vs. Whatever, as a company?
I currently have a Radio Shack Champion 2250XL tabletop. What I believe
is unique about it is that it actually has an LCD chessboard that shows
the pieces, as well as an LED indicated pressure sensitive chessboard.
I'm not sure of how its rated. How are chess computers rated?
--
Clear Skies,
Charles Sorgie
Deep Earth Music Website: http://home.earthlink.net/~stonebridge
Tim Mirabile
Komputer Korner <kor...@netcom.ca> wrote:
>I do press a return on each line. It is the stupid Mozilla news reader
>formatting
>program that is used in Netscape. does anybody else have any ideas on
>this?.
Ah, that explains it. Don't press return at all except to start a new
paragraph. Netscape is automatically formatting your posts, adding line breaks
somewhere between 70 and 80 characters. By pressing return slightly later, you
are adding additional line breaks, for example, after the word "formatting" in
your post above.
--
For USCF & FIDE rated chess on Long Island -> http://www.webcom.com/timm/
TimM on the Free Internet Chess Server - telnet://fics.onenet.net:5000/
Webmaster, tech support - ICD/Your Move Chess & Games: http://www.icdchess.com/
The opinions of my employers are not necessarily mine, and vice versa.
Robert Hyatt
mclane (mcl...@prima.ruhr.de) wrote:
: hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
: >Jakarta. Nimzo, crafty and virtual chess were all there. Shredder finished
: >first of course...
: If you would HAVE Virtual chess at home, you would not say stuff like
: the above.
: Virtual is very strong. We can let shredder play against Power-chess,
: if you want.
I never said "virtual is weak"... I simply said Shredder finished ahead
of it. I don't remember who beat who at Jakarta, but Crafty beat
Virtual Chess, and Shredder beat Crafty. After 12 rounds, it was a tough
event, with the best programs clustered at the top, slugging it out round
after round.
That's the only data I have to go on. However, general "common sense"
would rank programs based on a tournament like Jakarta, because winning
is *not* easy. Also Joel Rivat finished ahead of Virtual Chess in the
French Championship (Chess Guru). Again not to say that Virtual Chess
is weak, but others are definitely strong also.
: Also I have Nimzo Jakarta at home. We can let Virtual play against
: Nimzo 3.5 if you want.
: And more than 1 game.
the question is where Shredder comes in, not Nimzo or Virtual. Both
finished below Shredder in Jakarta...
: Jakarta was nice, but to few games.
Joe McCaughan
Charles Sorgie (stone...@earthlink.net) wrote:
: Charles Kamaras wrote:
: >
: > I have several models available.
: > I'm looking for the old Prestige Challenger by Fidelity,or the Montreux
: > by Saitek.
: Can you give me your opinion on chess computers? I'm interested in most
: likely the portable models, maybe the tabletops. What about Saitek vs.
: Novag vs. Whatever, as a company?
: I currently have a Radio Shack Champion 2250XL tabletop. What I believe
: is unique about it is that it actually has an LCD chessboard that shows
: the pieces, as well as an LED indicated pressure sensitive chessboard.
: I'm not sure of how its rated. How are chess computers rated?
You can get a rating by playing on FICS: telnet ics.onenet.net 5000 <cr>
Simply register as a computer (help register <cr>) and play a few
(hundred) games ... ;)
I have an RS 2050 (RazorSharp) that's got (approx) an 1850 rating on FICS.
: --
mclane
brucemo <bru...@seanet.com> wrote:
>To be fair, Virtual Chess had a wimpier machine.
Study the Virtual games and you see, it played strong.
Sometimes you lose, sometimes you win. All in all Virtual chess is/was
a strong program.
>bruce
brucemo
Komputer Korner wrote:
>
> Stefan said that Shredder can use 2 processors and take advantage of
> dividing the
> work between IO and Chess processing. The problem is that in WIN NT 4,
> the OS will decide the work allocation. You can't define it as a
> programmer. The
> argument was would Shredder be faster with 2 processors. Answer from
> Stefan was
> yes but a small amount. So since Shredder isn't a parallel chess engine,
> this idea
> wouldn't work in WIN NT 4.
I think that what Stefan is doing is that he has two threads, one for the
UI and one for the chess engine.
I don't know how NT works, but I've always assumed that in this case it
would assign an idle processor to a thread that came unblocked, if there
is one available. Since one would be doing chess engine almost
constantly, the other one would probably be available.
I've never had a two-processor machine, so I have zero experience with
this.
The next time I see Stefan I'm going to give him grief for starting this
whole mess in here, because this is a really dumb thread.
In a single-processor chess program, the engine will be going like crazy
and the UI will be *blocked* most of the time, meaning doing absolutely
nothing. When it is blocked it doesn't matter if you have one processor
or two. One will be assigned to chess. The others, if there are any,
will be idle anyway.
If you have two of them, perhaps it would be slightly faster to update the
clock every second, but if this is a serious issue, turn off the clock or
don't update it.
bruce
brucemo
mclane wrote:
>
> hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
>
> >Jakarta. Nimzo, crafty and virtual chess were all there. Shredder finished
> >first of course...
>
> If you would HAVE Virtual chess at home, you would not say stuff like
> the above.
> Virtual is very strong. We can let shredder play against Power-chess,
> if you want.
I am responding to Thorsten's post, but I'm also responding to Bob.
You guys are arguing chess program strength, but there are problems with either
argument. One of you is arguing based upon the result of an 11-round
tournament with non-uniform hardware, and the other is arguing based upon
subjective experience.
Either one of you could be wrong, but there is no way the other one is going to
prove it.
bruce
Rolf Tueschen
brucemo <bru...@seanet.com> wrote:
I thought the classical knock-out rules were still the actual ones. The
one on the floor :) unable to write another post has lost ...
>bruce
Jerome Campbell
On Tue, 1 Jul 1997, brucemo wrote:
> If you have two of them, perhaps it would be slightly faster to update the
> clock every second, but if this is a serious issue, turn off the clock or
> don't update it.
And then theres the OS stuff itself, two cpu's makes a significent
differece even if the application is single threaded---not having its
registers pushed and its cache dumped for OS stuff speeds up any
application.
chrisw
--
http://www.demon.co.uk/oxford-soft
Komputer Korner <kor...@netcom.ca> wrote in article
<33B6670E...@netcom.ca>...
> I do press a return on each line. It is the stupid Mozilla news reader
> formatting
> program that is used in Netscape. does anybody else have any ideas on
> this?.
I just bang iun into the Microsoft Explorer 3 software, just like this,
sometimes I do a CR and sometimes not. On my writer the line break was put
in automatically between the "and" and the "sometimes" above <carriage
return>
So what you see is what you get <carriage return>
Chris Whittington
David Gomboc
In article <5om1vb$m...@hpwin055.uksr.hp.com>,
Andy Monks <Andy_Monks@no_spam_hp.com> wrote:
>Chris Smith (smy...@ic.net) wrote:
>: Anyone aware of high end chess programs that are Win32 based? And
>: possibly mulitithreaded as well to take advantage of multiple
>: processors when run under Windows NT. Any coming soon?
>: Thanks.
>: Chris
>
>I know Robert Hyatt, the author of crafty is going to be making crafty
>utilize multiple processors. However, this will be only under unix. Time
>to load linux maybe!.
>
>Not sure when Robert will have finished this version (perhaps he'll reply)
I talked with someone who already has a parallel Crafty that runs
quite well. I don't know if that person wishes to remain anonymous or
not so I'll let them respond for themselves if they are interested.
It was kind of funny as I think Bob told him that his idea wouldn't
work that well, but it had already been implemented and worked
fine. :-) I don't have the knowledge to debate over how fine 'fine'
is, but other people tell me 'fine' in this case is actually rather
fine. :-)
That's all right though.. I'd rather read Bob's posts than many
others.. even if he dismisses something too casually on occasion.
Dave Gomboc
gom...@cpsc.ucalgary.ca
>--
>Best Regards,
>
>Andy Monks HP-UX WTEC Engineer
>Hewlett-Packard Ltd. Email andy_...@hp.com
>-------------------------------------------------------------------
>I am Pentium of Borg. Division is futile, you will be approximated.
>
>De' ll' Sovlu'DI' chaq Do'Ha' or Knowledge of useful information
> may be unfortunate.
>
>In 1492, an unknown and geographically incompetent sailor from Italy
>would lead a small Spanish fleet to the Americas.
Rolf Tueschen
Thanks, you did a good feedback below. Now I could see what's going on.
|just now after the point after "on" I hit CR| and I have installed 72 letters now I try to see where the automatical break is performed but now i
must stop with CR after the last "i" because of the formatting |. Note,
this is not a "I" but the symbol which is given at the keyb with a short
line above then a short break and another short line. Why this symbol
changes all the formatting process?
I state all the above breakes without CR as they appear on my display.
First line after "on." Second in the double long line I hit CR after the
minor "i", third break after "Note,", forth after "short", fifth after
"symbol", sixth after "process?" But there I hit CR to mark a new part.
Now apart the ugly second line which must make explode on most news
readers the displayed picture is nice.
I also thank Tim. Because the conscient hitting of CR seems to cause
this stupid picture of a single word in a line. So I conclude without
knowing better:
(i) We should keep the number of signs pro line low at 70, eventually 72
(ii) We should only hit CR for a ending of a whole paragraph, most
readers will perform like the text programs and stop the lines around
the chosen number mentioned in (i).
Could this be confirmed by anyone of Bill Gates I. jun class?
For all I want to know who's to blame? When I did it at 79, and the
picture is nice on my screen. But on the screen after Chris posted it
back in a quote it's horribly broken by these one word liners. See
below.
Is it then Chris who's responsible? HELP.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
mclane
brucemo <bru...@seanet.com> wrote:
>Either one of you could be wrong, but there is no way the other one is going to
>prove it.
>bruce
I find it always very ignorant/arrogant to discuss or to argue about a
data, that is not known to the guy who claims something.
e.g. a book. Somebody gives a value about a book he has not read just
because he don't likes what he has heard about the author.
Of course my judgement is subjective.
If you want we can let Virtual play against 12.6 crafty to show some
more data.
But before I judge, i normally study the data myself.
Also I am not the only guy thinking this. Peter Schreiner, who has
much chess knowledge and writes for Europa Rochade is also this
opinion. And some other friend I trust think the same. Virtual chess
is the most underestimated chess program we have.
Robert Hyatt
mclane (mcl...@prima.ruhr.de) wrote:
: brucemo <bru...@seanet.com> wrote:
: >Either one of you could be wrong, but there is no way the other one is going to
: >prove it.
: >bruce
: I find it always very ignorant/arrogant to discuss or to argue about a
: data, that is not known to the guy who claims something.
: e.g. a book. Somebody gives a value about a book he has not read just
: because he don't likes what he has heard about the author.
Could you perhaps explain what the above means? Somehow it slips past
my ability to figure it out.
: Of course my judgement is subjective.
: If you want we can let Virtual play against 12.6 crafty to show some
: more data.
What does Virtual Chess vs Crafty have to do with anything being
discussed? Play it all you like. Or continue with your CSTal games.
: But before I judge, i normally study the data myself.
: Also I am not the only guy thinking this. Peter Schreiner, who has
: much chess knowledge and writes for Europa Rochade is also this
: opinion. And some other friend I trust think the same. Virtual chess
: is the most underestimated chess program we have.
Note I'm not "subjectively" evaluating virtual chess. I *know* that
Shredder finished ahead of it in a tough 12-round tournament. That's
not speculation, nor supposition, nor guesswork... I simply posted that
I thought KK's ranking was wrong. I still do. But it *is* my opinion,
as I stated. It is very *interesting* just how much my opinion seems to
get you fired up. Is there a reason for that? Damned if I'm going to agree
with you "just because"... You can either get over it, get lost, or get
real...
Summary:
I maintain Shredder should be ranked above virtual chess, Crafty and
Nimzo, based on the Jakarta result. That's all I said. That's all I
meant. I didn't ask for a slew of games to justify Virtual being
better than Shredder.
Robert Hyatt
David Gomboc (gom...@cpsc.ucalgary.ca) wrote:
: In article <5om1vb$m...@hpwin055.uksr.hp.com>,
: Andy Monks <Andy_Monks@no_spam_hp.com> wrote:
: >Chris Smith (smy...@ic.net) wrote:
: >: Anyone aware of high end chess programs that are Win32 based? And
: >: possibly mulitithreaded as well to take advantage of multiple
: >: processors when run under Windows NT. Any coming soon?
: >: Thanks.
: >: Chris
: >
: >I know Robert Hyatt, the author of crafty is going to be making crafty
: >utilize multiple processors. However, this will be only under unix. Time
: >to load linux maybe!.
: >
: >Not sure when Robert will have finished this version (perhaps he'll reply)
: I talked with someone who already has a parallel Crafty that runs
: quite well. I don't know if that person wishes to remain anonymous or
: not so I'll let them respond for themselves if they are interested.
: It was kind of funny as I think Bob told him that his idea wouldn't
: work that well, but it had already been implemented and worked
: fine. :-) I don't have the knowledge to debate over how fine 'fine'
: is, but other people tell me 'fine' in this case is actually rather
: fine. :-)
: That's all right though.. I'd rather read Bob's posts than many
: others.. even if he dismisses something too casually on occasion.
: Dave Gomboc
: gom...@cpsc.ucalgary.ca
I remember the conversation. For the record, "fine" (my perspective)
means running about 3.7X faster on 4 processors, >7X faster on 8
processors. 12X faster using 16 processors. The old "standby" ideas
of splitting the search window into "strips" and searching different
ranges in parallel won't cut if for N>4. PVS won't cut it for N>4.
I don't have enough data on MTD(f) yet. It does have promise, but is
a special case of the first approach I mentioned above. I've implemented
all of 'em (not on crafty) and was not happy with anything except for the
DTS algorithm I used in Cray Blitz, and it was sagging badly on 32
processors and higher...
Komputer Korner
Tim Mirabile wrote:
> Ah, that explains it. Don't press return at all except to start a new
>
> paragraph. Netscape is automatically formatting your posts, adding
> line breaks
> somewhere between 70 and 80 characters. By pressing return slightly
> later, you
> are adding additional line breaks, for example, after the word
> "formatting" in
> your post above.
>
> --
> For USCF & FIDE rated chess on Long Island ->
> http://www.webcom.com/timm/
> TimM on the Free Internet Chess Server -
> telnet://fics.onenet.net:5000/
> Webmaster, tech support - ICD/Your Move Chess & Games:
> http://www.icdchess.com/
> The opinions of my employers are not necessarily mine, and vice versa.
Okay this reponse is a test to see if your suggestion works. I will
just let the line run on like this and see what happens. This could get
quite interesting. I remember in the earlier versions of netscape the
line would just run on . we will see if that is the case here.
Komputer Korner
Komputer Korner wrote:
> Tim Mirabile wrote:
>
> > Ah, that explains it. Don't press return at all except to start a
> new
> >
> > paragraph. Netscape is automatically formatting your posts, adding
> > line breaks
> > somewhere between 70 and 80 characters. By pressing return slightly
>
> > later, you
> > are adding additional line breaks, for example, after the word
> > "formatting" in
> > your post above.
> >
It works It works No more carriage returns from now on. I guess we are
out of the horse and buggy days anyway. But what about when I create a
document outside of Netscape and then post it here? Will it do the same
thing?
Rolf Tueschen
Komputer Korner <kor...@netcom.ca> wrote:
>Komputer Korner wrote:
>> Tim Mirabile wrote:
>>
>> > Ah, that explains it. Don't press return at all except to start a
>> new
>> >
>> > paragraph. Netscape is automatically formatting your posts, adding
>> > line breaks
>> > somewhere between 70 and 80 characters. By pressing return slightly
>>
>> > later, you
>> > are adding additional line breaks, for example, after the word
>> > "formatting" in
>> > your post above.
>> >
> It works It works
Come down again. _Your_ post works, I agree, but Tim's post is screwed
as before yours. See above ...
chrisw
--
http://www.demon.co.uk/oxford-soft
Komputer Korner <kor...@netcom.ca> wrote in article
> Watch out there Bob. Some say that God plays at 3000 level. Your program
> is getting
> dangerously close to it. And he who creates a GOD is tampering with the
> ALMIGHTY.
> I know I know, there were GOD threads a while back.
> I can't remember what Elo everybody finally settled on. If somebody
> remembers please
> post and I will stand corrected in my admonition of Bob.
>
> Robert Hyatt wrote:
>
> > 12.6 seems to be better. It just set an all-time high for blitz on
> > ICC last night, 2989. It's not *that* good, but it seems to be
> > playing
> > pretty decently of late. We'll see...
>
An excellent point, Korner. Only 11 points short of God :)
But, presumably God's Elo is found by the equation:
God's Elo = largest earthling's claimed / real / imagined elo + some
positive random number
And since God would never lose (he might draw occasionally ?), the positive
random number has got to be near 400 for the statistics to work, and,
worse, has no calculable upper bound ?
Chris Whittington
David Katelansky
>From: Charles Sorgie <stone...@earthlink.net>
>I currently have a Radio Shack Champion 2250XL tabletop. What I believe
>is unique about it is that it actually has an LCD chessboard that shows
>the pieces, as well as an LED indicated pressure sensitive chessboard.
>I'm not sure of how its rated. How are chess computers rated?
Those numbers usually imply that the computer plays at that strength; e.g.,
2250. Ages ago I bought a SciSys 16K that was supposed to be rated about
1800; it was more like 1600. The Chessmaster 2100 certainly wasn't a 2100
player; I have a massively plus score against it. It's hard to say how strong
a computer plays, since they play very unhumanly. I remember reading an
article in 1972 by Dr. Charles Kalme, where he put the best machines of the
time at about 1500 - some people were saying that chess computers of the time
were playing at expert level (back in 1967 Greenblatt's program was already
playing about 1600 to 1700 strength).
Dave Katelansky
herc...@edenbbs.com
David Regis
>>No more carriage returns from now on. I guess we are
>>out of the horse and buggy days anyway. But what about when I create a
>>document outside of Netscape and then post it here? Will it do the same
>>thing?
Don't rely on it! This horse and buggy user would still like to see
carriage returns at about line 70, without which posts are hard to
read and difficult to follow-up.
P.S. Is it Newtscape that is responsible for all the posts I see which
have the whole damn thing in HTML appended to the message? Now
*there's* a button whose default should be off. If software "improves"
any more I'm going to give up reading news...
--
May your pieces harmonise with your Pawn structure and
your sacrifices be sound in all variations
D _
/ "()/~ Dave Regis &8^D* Exeter Chess Coaching Page etc.:
|| \_/| = DrDave on BICS http://www.ex.ac.uk/~dregis/DR/chess.html
~\ / "...what else exists in the world but chess?"
_|||__SHEU ~/sheu.html -- NABOKOV "Contribute!" -- Doug Attig
mclane
hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
>Note I'm not "subjectively" evaluating virtual chess. I *know* that
>Shredder finished ahead of it in a tough 12-round tournament.
So - how often had virtual to play against crafty or shredder ? 2
times ? 3 times ? 1 time ?
This was a suisse tournament with people using different types of
hardware.
And I would say this data is nice, but not so relevant.
Of course fact is fact. But when I throw a coin 1 time, or when I
throw it 10 times, the chances occur different.
Here in germany Virtual chess has quite a good reputation due to
articles of Mr. Schreiner.
>That's
>not speculation, nor supposition, nor guesswork... I simply posted that
>I thought KK's ranking was wrong. I still do.
Right. I subscribe this. But is it important how KK ranks them ?
> But it *is* my opinion,
>as I stated. It is very *interesting* just how much my opinion seems to
>get you fired up.
I am not fired up.
>Is there a reason for that? Damned if I'm going to agree
>with you "just because"... You can either get over it, get lost, or get
>real...
No. The only reason is, that I think Virtual is stronger than you all
believe.
> I didn't ask for a slew of games to justify Virtual being
>better than Shredder.
What a pity. I mean: this would really PROVE it. So you are not
interested in proving. Aha.
This is of course your free will and I cannot say more to this.
Anyone has to live with his own selected data of the world (of
computerchess).
Robert Hyatt
mclane (mcl...@prima.ruhr.de) wrote:
: hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
: >Note I'm not "subjectively" evaluating virtual chess. I *know* that
: >Shredder finished ahead of it in a tough 12-round tournament.
: So - how often had virtual to play against crafty or shredder ? 2
: times ? 3 times ? 1 time ?
: This was a suisse tournament with people using different types of
: hardware.
: And I would say this data is nice, but not so relevant.
: Of course fact is fact. But when I throw a coin 1 time, or when I
: throw it 10 times, the chances occur different.
: Here in germany Virtual chess has quite a good reputation due to
: articles of Mr. Schreiner.
*Please* learn how to read. Surely you have schools over there that can
assist you somehow?
I did *not* say virtual chess was a weak program. I did *not* say
virtual chess was a lousy chess player. I did *not* say *anything*
bad about Virtual Chess at all. *period*...
Now read the following carefully. I am *not* "writing between the
lines." I am writing *exactly* what I mean:
Shredder is the current world microcomputer chess champion. It played
well in Jakarta and finished ahead of everyone there, including Virtual
Chess, Crafty, Ferret, Nimzo and a host of others. Based on *that* I
believe it deserves (has earned) the respect to be listed at the top of
the programs I mentioned. *period*. Virtual Chess also lost in the
French tournament to ChessGuru, further evidence that it is not significantly
better than the programs I mentioned.
There's no mystery, there's no back-room games played on oddball hardware,
there's no hidden agendas, there's no guessing, there's no interpretation,
there's no guesswork. Shredder finished ahead of Virtual Chess, Shredder
should be ranked ahead of Virtual Chess.
That is *all* I have said. I stand by it. I don't care what you think about
how it plays, about its reputation, or anything else. He won a 12 round
event by playing good chess. Shredder deserves the respect that win should
confer.
: >That's
: >not speculation, nor supposition, nor guesswork... I simply posted that
: >I thought KK's ranking was wrong. I still do.
: Right. I subscribe this. But is it important how KK ranks them ?
I simply pointed out my opinion. He doesn't have to support my opinion,
he's a grown boy it seems. If he disagrees with something I say, he should
post what he thinks. I simply suspected that he had forgotten about Jakarta
and who played who, and who won, and reminded him. He can rank things any
way he wants... And if I don't agree, I reserve the right to challenge the
rankings and give reasons to support why I think they should be different..
: > But it *is* my opinion,
: >as I stated. It is very *interesting* just how much my opinion seems to
: >get you fired up.
: I am not fired up.
: >Is there a reason for that? Damned if I'm going to agree
: >with you "just because"... You can either get over it, get lost, or get
: >real...
: No. The only reason is, that I think Virtual is stronger than you all
: believe.
OK. that's an opinion. I don't think Virtual is "weak". I simply think
Shredder is stronger than Virtual. If someone plays a bunch of games to
provide data, I can easily revise my opinion...
: > I didn't ask for a slew of games to justify Virtual being
: >better than Shredder.
: What a pity. I mean: this would really PROVE it. So you are not
: interested in proving. Aha.
not sure what it would prove, unless we get shredder on a decent game,
configured decently by the author, vs virtual on an equivalent machine,
with a decent configuration. Running on old hardware is not a good way
to test anything, as it can produce results that are misleading. Several
people run Crafty on very slow machines (that is all they have) and they
send me quite a bit of mail about positions where Crafty plays grossly due
to a null-move failure. In every case, pumping the NPS rate up to something
reasonable lets the depth reach a point where the null-move failures are less
of a problem, but on slow hardware, shallow searches can "kill." If you play
Crafty vs X, where X is a non-null-move program, and you use 386 machines,
Crafty will likely get killed. Because every game will be marred by null-
move failures due to shallow searches. Step both up to a P6/200, and the
results changed markedly, because that's where I develop and tune. Shredder
and/or virtual could behave the same way. So off-the-cuff games on inferior
hardware won't prove a thing. I saw the games Shredder played in Jakarta.
It played nice chess. I saw the game where Ferret beat Shredder and both
played good chess.
You can play all the games you want, but the best way would be a "author"
vs "author" match, like the ones Bruce and I play all the time on ICC. The
best Crafty I can do, vs the best Ferret Bruce can do, on the best (and equal)
hardware we can use (P6/200's).
If you look at all the "crafty" clones running on the servers, one thing that
I'm asked about over and over is "why is *your* crafty rated 200 points or
more above the next closest crafty? Are you holding back some of the new
code you've done?" The answer is *no*. I'm just careful with the book,
careful with the learning, and repair the book when something bad is discovered
by opponents. However, I'd maintain that *my* crafty is the best one, with
two exceptions for people that run the experimental versions I use to help me
test and debug. But it's best because it is running on the platform it is
developed on, using the book it is developed with, etc. That's a difference
between "crafty" and "Crafty-clones". If you want to see how "X" does against
Crafty, log on to ICC and play "crafty". If you want to see how "X" does
against Shredder, we need to get Shredder on ICC and let you play it. That
is the "best" test there is...
: This is of course your free will and I cannot say more to this.
Komputer Korner
Robert Hyatt wrote:
WE have all been guilty of mixing up
blitz results and Fischer 5 5 games, etc
with 40/2 chess.There is a significant
difference between the programs at the
different time controls. It is my
opinion that the most important time control
is Fischer 5 5. The reason is that no one
likes to flag on time and almost everyone
likes the machine to play quickly. Based on
results against IM Tom O'Donnell and against
Larry Kaufman , I rated Shredder lower. This
was unfair because Shredder is optimized for
40/2 which is the standard time control of
the world micro Championships and SSDF
testing. This is a big problem as hardly any
consumer ever plays 40/2 against their
program. It is almost always quick time
controls with the human cheating on the time
control. Basically Fischer 5 5 seems to be
the happy medium time control and this
should be recognized by the SSDF and
the WMCCC. I am not saying that 40/2
isn't important , but it is unrealistic to
place so much importance on it when
Fischer 5 5 is used the most. On the
servers, perhaps the time control average
is even quicker. Bob and Lonnie and
others know better what the actual average
is. The point is fast chess is what the
consumer wants and SSDF and WMCCC
have been a slave to 40/2 to the detriment
of what the people use their programs for.
Note that for long time analysis, 40/2 is
still a good indicator, so I am not belittling
the importance of 40/2, just I would like to
see a little more balance in time control tests.
--
Best regards
mclane
hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
>mclane (mcl...@prima.ruhr.de) wrote:
>: hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
>: >Note I'm not "subjectively" evaluating virtual chess. I *know* that
>: >Shredder finished ahead of it in a tough 12-round tournament.
>: So - how often had virtual to play against crafty or shredder ? 2
>: times ? 3 times ? 1 time ?
>: This was a suisse tournament with people using different types of
>: hardware.
>: And I would say this data is nice, but not so relevant.
>: Of course fact is fact. But when I throw a coin 1 time, or when I
>: throw it 10 times, the chances occur different.
>: Here in germany Virtual chess has quite a good reputation due to
>: articles of Mr. Schreiner.
>*Please* learn how to read. Surely you have schools over there that can
>assist you somehow?
Don't know why you go into the air now.
I was 19 years at schools.
This has NOTHING to do with READING.
KK was building a LIST !!
You talk about a chmapionship !
Thats something different.
The world-chess-champion could be the winner of a championship without
leading the list ! Lists like the swedish list, or erics lists, or
e.g. here kk's lists were build on different data than championships.
This is the case in ANY sport I know. Boris Becker could win Wimbelton
but mustn't be the first rank in the list.
Your idea that the world-computer-chess-champion has to be placed
FIRST in the list, or above the others woh participated in
championships is nice idea, but does not have any data behind despite
the fact of the jakarta event.
>I did *not* say virtual chess was a weak program. I did *not* say
>virtual chess was a lousy chess player. I did *not* say *anything*
>bad about Virtual Chess at all. *period*...
No. If you would have said this, my answers would have been a degree
more aggressive.
>Now read the following carefully. I am *not* "writing between the
>lines." I am writing *exactly* what I mean:
>Shredder is the current world microcomputer chess champion. It played
>well in Jakarta and finished ahead of everyone there, including Virtual
>Chess, Crafty, Ferret, Nimzo and a host of others. Based on *that* I
>believe it deserves (has earned) the respect to be listed at the top of
>the programs I mentioned. *period*.
You BELIEVE ! Exactly ! I don't know that any lists of ranking behave
with your ideas.
At a championship they play 1 game against each other.
This is , related to the 100 or more games the ssdf-people play out,
NOTHING!
If they would arrange their lists because of CHAMPIONSHIPS, than
Fritz3 would have been leader of the ssdf-list, just because YOU
BELIEVE in it, and Fritz3 has shown with hundreds of games in sweden
that it is unable to do this.
The same for Genius. It has won the championship in paderborn, but was
NOT leading the ssdf-list.
I have nothing against your interpretation, if you call it a BELIEVE.
I have ever said that you are as religious as any other guy here.
> Virtual Chess also lost in the
>French tournament to ChessGuru, further evidence that it is not significantly
>better than the programs I mentioned.
I don't know what ChessGuru is. Never heard of this.
Your idea that Virtual is NOT significantly stronger than the programs
YOU mention is based on how much data/games ???
>There's no mystery, there's no back-room games played on oddball hardware,
Why do you suggest mysteries ?
I spoke about different hardware usage in Jakarta. You speak about
mysteries. Strange...
>there's no hidden agendas, there's no guessing, there's no interpretation,
>there's no guesswork.
No interpretation. For me it seems you interpretate. But i forgot that
any of your statements is not interpretation but plain TRUTH and
FACT... :-)
>Shredder finished ahead of Virtual Chess, Shredder
>should be ranked ahead of Virtual Chess.
Based on this 1 game from jakarta where they had different hardware
(Computer Schach and Spiele writes Shredder used a PentiumPro200,
Virtual used a P5/166, others were "forced" to use only P5/133) ?
>That is *all* I have said. I stand by it. I don't care what you think about
>how it plays, about its reputation, or anything else. He won a 12 round
>event by playing good chess. Shredder deserves the respect that win should
>confer.
We don't attack this respect. I would do the hell to attack
Stefan-Meyer-Kahlen and his program. I like him, and I was in the same
SHREDDER-fever than any other guy who know about this nice program.
But this is not the question here. I do not speak about respect
concerning Shredder. Or respect towards Stefan.
You are mixing analysis and Respect !
The one thing is scientific, the other thing is feelings and
social-stamping on patterns.
What you try is a kind of hierarchical scheme based on championships.
This is nice, and maybe respectful, but is has nothing to do with the
data we all can produce, repeat, and prove. How many games with
shredder did you play yourself. I have played out several games with
shredder at home before I judge/evaluate it. You seem to base your
evaluation on championships.
Hm.
> And if I don't agree, I reserve the right to challenge the
>rankings and give reasons to support why I think they should be different..
You REASONS are not based on much data, they are based mainly on
feelings that somebody has won a title. Lists in all sports in the
world, tennis, chess, soccer, ... were NOT based on
championship-events.
They were based on repeated, often played out data.
Otherwise it would not be scientific.
>OK. that's an opinion. I don't think Virtual is "weak". I simply think
>Shredder is stronger than Virtual. If someone plays a bunch of games to
>provide data, I can easily revise my opinion...
Right. You have an opinion. I have an opinon. The opinons are
different. I asked on how much data you BASED your opinon. You said:
Jakarta.
Shredder and Virtual played 11 games in jakarta.
Your opinoin is therefore based on 11 games.
My opinon is based on more games. And I have both programs at home.
I don't need championships for that. And - as I said above -
Fritz3 is NOT leading the swedish-list, although it won in Hong-Kong.
And Genius is not leading it, although it won in Paderborn.
Mchess was leading it, after paderborn !!
Until Rebel came on the market.
Until Hiarcs came on the market.
>not sure what it would prove, unless we get shredder on a decent game,
>configured decently by the author, vs virtual on an equivalent machine,
>with a decent configuration. Running on old hardware is not a good way
>to test anything, as it can produce results that are misleading.
Aha. Running in Pentium 166 or 133 as in Jakarta ! Now I understand
what you mean. :-)
Bob - you prove my idea of you beeing ignorant.
> Several
>people run Crafty on very slow machines (that is all they have) and they
>send me quite a bit of mail about positions where Crafty plays grossly due
>to a null-move failure. In every case, pumping the NPS rate up to something
>reasonable lets the depth reach a point where the null-move failures are less
>of a problem, but on slow hardware, shallow searches can "kill." If you play
>Crafty vs X, where X is a non-null-move program, and you use 386 machines,
Aha. And WHERE do we or the swedish guys use 386-hardware ???????
I am using Pentium120 for a long time. I have tested shredder on this
hardware before the time it was released. I have discussed with the
author. Played out games. I do not speak about a software, without
having data.
And when, than I ask for more data.
Nobody uses 386 or 486. We all use pentiums.
I have bought an AMD K6 a week before. Moritz has done the same.
Others (Peter) has had a P5/200 Laptop at Aegon, I took my p5/120 to
Den Haag although there they gave us all a p6/200. Do you understand ?
ALL 50 programs got a p6/200. Of course if somebody who wanted a
faster machine, came with special tuned overclocked machines, like the
Nimzo-team.
But nobody uses the machines you mention because we want to produce
false result. I have never had a 386. I jumped from 286-16 to 486-33
to 486-100 to p5/120..
>Crafty will likely get killed. Because every game will be marred by null-
>move failures due to shallow searches. Step both up to a P6/200, and the
>results changed markedly, because that's where I develop and tune.
No doubt that this is important. If you want I can let Crafty run on
my AMDK6/200 with cstal or virtual run on p5/120 level. And the
results would be the same.
> Shredder
>and/or virtual could behave the same way. So off-the-cuff games on inferior
>hardware won't prove a thing.
I cannot subscribe to this. ANY game gives us more data.
Nobody tries to cheat.
We all try to make it optimum. Not anybody has the money to buy to
state-of the art machines. I have now 3 machines. 2 p5/120 and this
AMDK6/200. I would NOT say that people who have LOWER hardware cannot
speak with you in this discussion. That would be ignorant.
Any games proves something. People like you will always find an alibi
or an excuse why their programs lost the game. :-)
> I saw the games Shredder played in Jakarta.
>It played nice chess. I saw the game where Ferret beat Shredder and both
>played good chess.
I have no doubt about this. Stefan and I live almost in the same town
and I have seen nice shredder games too,
But this does not help in the discussion here.
>You can play all the games you want, but the best way would be a "author"
>vs "author" match, like the ones Bruce and I play all the time on ICC. The
>best Crafty I can do, vs the best Ferret Bruce can do, on the best (and equal)
>hardware we can use (P6/200's).
We should ask Stefan himself how strong Shredder is.
BTW: stefan does not use a P6/200 himself.
We have benchmarked our machines and I have benchmarked shredder on
several machines when I worked in the computer-company.
If only AUTHORS would be able to find out the truth, ... ha - nice
world you construct.
I like an open world. Anybody buys or downloads, and anybody shares
games and events and data with anybody. You seem to have a more CLOSED
scenario in mind.
I now understand why you don't want to see games of crafty here.
ONLY AUTHORS are able/capable to do it right.
You really seem to have an ICC complex.
You generate really high rating there. NO program of your ICC programs
has ever proved his high ranked elo rates here in europe in normal
games. Stobor, Ferret, crafty and Zarkov or others. When they come to
europe they do not play better than any of the NON-icc programs.
I have told you WHY we are not able to play online somewhere. Because
europe telephone - companies have (despite from what you talk about
capitalism and free market) a monopol.
Therefore the prices/costs are very high.
>If you look at all the "crafty" clones running on the servers, one thing that
>I'm asked about over and over is "why is *your* crafty rated 200 points or
>more above the next closest crafty? Are you holding back some of the new
>code you've done?" The answer is *no*. I'm just careful with the book,
>careful with the learning, and repair the book when something bad is discovered
>by opponents.
It could also be that YOUR OPPONENTS are not that strong. Or they are
strong rated but do play weak on the telephone, because they don't
play serious and just want to relax playing chess.
When Seiravan blitzes in Den Haag against Mchess in the nights, I
would not COUNT these games for rating purposes, no matter if Seiravan
playes 8-2 or 2-8. It is fun. Can it be that often people log on ICC
server of how these servers are called where people can log-in, and do
not always play serious, but more fun.chess and you make serious
statistics about ??
> However, I'd maintain that *my* crafty is the best one, with
>two exceptions for people that run the experimental versions I use to help me
>test and debug. But it's best because it is running on the platform it is
>developed on, using the book it is developed with, etc. That's a difference
>between "crafty" and "Crafty-clones". If you want to see how "X" does against
>Crafty, log on to ICC and play "crafty".
We will do this when the costs are in relation to the worth of the
data we get.
Look - nobody wants to cheat by using wrong settings. If we would have
the hardware and the money, we would all use a very fair and similar
hardware and would telephone you to dead in the us.
Do not insult the people just because they have not the same money or
access to the telephone-net.
How much do you pay for a local call in US in 1 hour ??
> If you want to see how "X" does
>against Shredder, we need to get Shredder on ICC and let you play it.
ICC is no better proval of anything.
> That
>is the "best" test there is...
No - that is wrong. That is the best test you suppose. Your rated elos
from ICC servers have never been proved somewhere else...
Robert Hyatt
Komputer Korner (kor...@netcom.ca) wrote:
: Robert Hyatt wrote:
: > You can play all the games you want, but the best way would be a
: WE have all been guilty of mixing up
: others know better what the actual average
: is. The point is fast chess is what the
: consumer wants and SSDF and WMCCC
: have been a slave to 40/2 to the detriment
: of what the people use their programs for.
: Note that for long time analysis, 40/2 is
: still a good indicator, so I am not belittling
: the importance of 40/2, just I would like to
: see a little more balance in time control tests.
: --
here we tend to agree. But I see *three* classes of games that are
played:
1. long games (40/2 is one example, but game/60 and so forth fit that
mold as well.
2. fast games (5 minute games and the like)
3. ICC games. I keep these separate because they are played in a
more relaxed atmosphere for most and the games are more casual. In
Crafty's case, it obviously is *dead serious* about playing all three
since it never relaxes, but humans on ICC/etc seem to play a little more
off-handed than in tournaments. This is *not* true of GM/IM players,
however (in my opinion) as they know there will be big crowds watching,
and they sweat bullets most of the time trying to play their best.
However, if you make a "blitz chess" list, then Ferret has to be at the
very top. Check out how it did at Jakarta by winning *every* game in the
blitz tournament.
Also, using a blitz ranking will turn the typical "program order" upside-
down, as the programs at the top of the SSDF don't play great fast chess at
all...
Robert Hyatt
: >mclane (mcl...@prima.ruhr.de) wrote:
: >: hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
: >: >Note I'm not "subjectively" evaluating virtual chess. I *know* that
: >: >Shredder finished ahead of it in a tough 12-round tournament.
: >: So - how often had virtual to play against crafty or shredder ? 2
: >: times ? 3 times ? 1 time ?
: >: This was a suisse tournament with people using different types of
: >: hardware.
: >: And I would say this data is nice, but not so relevant.
: >: Of course fact is fact. But when I throw a coin 1 time, or when I
: >: throw it 10 times, the chances occur different.
: >: Here in germany Virtual chess has quite a good reputation due to
: >: articles of Mr. Schreiner.
: >*Please* learn how to read. Surely you have schools over there that can
: >assist you somehow?
: Don't know why you go into the air now.
: I was 19 years at schools.
: This has NOTHING to do with READING.
: KK was building a LIST !!
: You talk about a chmapionship !
Yes, but I *did not* say anything negative about Virtual Chess. Rather,
I said something *positive* about Shredder. Very simple. Very easy to
understand. Very easy to justify. And impossible to interpret as "we
think Virtual Chess is better than you give it credit for being." I did
not imply (nor mean to imply) that Virtual Chess was a weak program. I
only stated that to win a 12-round bloodfest like Jakarta, the winner
is going to have to be more than "just lucky" with 12 rounds.
: Thats something different.
: The world-chess-champion could be the winner of a championship without
: leading the list ! Lists like the swedish list, or erics lists, or
: e.g. here kk's lists were build on different data than championships.
: This is the case in ANY sport I know. Boris Becker could win Wimbelton
: but mustn't be the first rank in the list.
: Your idea that the world-computer-chess-champion has to be placed
: FIRST in the list, or above the others woh participated in
: championships is nice idea, but does not have any data behind despite
: the fact of the jakarta event.
It has some data. Shredder beat Virtual Chess, Shredder beat Crafty,
Crafty beat Virtual Chess... and so forth. A conclusion *can* be drawn
from the above. Perhaps not with 100% confidence, but with pretty high
confidence.
: >I did *not* say virtual chess was a weak program. I did *not* say
: >virtual chess was a lousy chess player. I did *not* say *anything*
: >bad about Virtual Chess at all. *period*...
: No. If you would have said this, my answers would have been a degree
: more aggressive.
: >Now read the following carefully. I am *not* "writing between the
: >lines." I am writing *exactly* what I mean:
: >Shredder is the current world microcomputer chess champion. It played
: >well in Jakarta and finished ahead of everyone there, including Virtual
: >Chess, Crafty, Ferret, Nimzo and a host of others. Based on *that* I
: >believe it deserves (has earned) the respect to be listed at the top of
: >the programs I mentioned. *period*.
: You BELIEVE ! Exactly ! I don't know that any lists of ranking behave
: with your ideas.
: At a championship they play 1 game against each other.
: This is , related to the 100 or more games the ssdf-people play out,
: NOTHING!
: If they would arrange their lists because of CHAMPIONSHIPS, than
: Fritz3 would have been leader of the ssdf-list, just because YOU
: BELIEVE in it, and Fritz3 has shown with hundreds of games in sweden
: that it is unable to do this.
: The same for Genius. It has won the championship in paderborn, but was
: NOT leading the ssdf-list.
I'm not sure what this means. Does Kasparov want to (a) be the #1 ranked
player in the world or (b) win every tournament he enters regardless of (a)?
I suspect (b).
: I have nothing against your interpretation, if you call it a BELIEVE.
: I have ever said that you are as religious as any other guy here.
: > Virtual Chess also lost in the
: >French tournament to ChessGuru, further evidence that it is not significantly
: >better than the programs I mentioned.
: I don't know what ChessGuru is. Never heard of this.
: Your idea that Virtual is NOT significantly stronger than the programs
: YOU mention is based on how much data/games ???
about 20 games, all 40/2, all played with the program authors there running
the program in the best way they can. As I said before, that's *all* the
data I have, but it is a non-trivial amount of data between Jakarta and the
French computer chess championship. ChessGuru is similar to Crafty, written
by Joel Rivat. He uses bitmaps (we have compared notes quite frequently at
times) and he ran on a DEC alpha of some sort. Plays very strong chess...
: >There's no mystery, there's no back-room games played on oddball hardware,
: Why do you suggest mysteries ?
: I spoke about different hardware usage in Jakarta. You speak about
: mysteries. Strange...
the hardware disadvantage wasn't huge. P5/166 vs P6/200. certainly not a
factor of 2x, although it isn't equal. I don't know how virtual performs on
a P6, but if it is a 16 bit engine, the difference was even less...
: >there's no hidden agendas, there's no guessing, there's no interpretation,
: >there's no guesswork.
: No interpretation. For me it seems you interpretate. But i forgot that
: any of your statements is not interpretation but plain TRUTH and
: FACT... :-)
I'm notinterpreting, I'm just counting wins and losses in two tournaments.
Shredder finished with more 1's than Virtual Chess did. I then sort into
order based on wins/draws/losses and Voila! I have a ranking that is not
subjective at all...
: >Shredder finished ahead of Virtual Chess, Shredder
bikejr
Komputer Korner <kor...@netcom.ca> wrote in article
> others know better what the actual average
> is. The point is fast chess is what the
> consumer wants and SSDF and WMCCC
> have been a slave to 40/2 to the detriment
> of what the people use their programs for.
> Note that for long time analysis, 40/2 is
> still a good indicator, so I am not belittling
> the importance of 40/2, just I would like to
> see a little more balance in time control tests.
> --
>
> Best regards
> Komputer Korner
On the ICC most crafty's (including crafty itself) won't play any slower
blitz than 5 3 time control. Some won't play with an increment at all (cuts
down on computer cheaters for sure.). For the manual ops 2 12 seems the
most popular as then you have time to manually transfer the moves without
being in a fire drill status. For humans vs machines 3 0, 5 0, and 5 3 seem
popular. To me calling 3 0 blitz and 5 12 blitz is pushing it. I mean
there's a lot of difference between a 3 0 game and a 5 12 game, and certain
programs on certain machine will play one or the other better, but it's
still all considered blitz. Certainly on the servers no one plays 40/2, and
that has always been in my opinion on of the big limitations of the SSDF.
It doesnt test what is really happening. Why not start a ranking system for
blitz chess time controls (or does such an animal exist?). You could sure
get a lot more game data in a hurry, and perhaps provide something more
useful to what people are really out there playing. In all honesty being
also a manual op as well as a crafty clone hack on inferior hardware, when
I want a new chess program for server play. I don't care how good it plays
40/2 as I'm never gonna play that. I want the strongest program that can
play like a 2 12 control, or a 5 3, or whatever Im gonna play. As an aside
it would certainly be interesting if some of these commercial programs
could be automated for server use and put up against the handfull or less
of programs that are currently automated.
Rolf Tueschen
mcl...@prima.ruhr.de (mclane) wrote:
>hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
>>Note I'm not "subjectively" evaluating virtual chess. I *know* that
>>Shredder finished ahead of it in a tough 12-round tournament.
>So - how often had virtual to play against crafty or shredder ? 2
>times ? 3 times ? 1 time ?
>This was a suisse tournament with people using different types of
>hardware.
>And I would say this data is nice, but not so relevant.
>Of course fact is fact. But when I throw a coin 1 time, or when I
>throw it 10 times, the chances occur different.
>Here in germany Virtual chess has quite a good reputation due to
>articles of Mr. Schreiner.
>>That's
>>not speculation, nor supposition, nor guesswork... I simply posted that
>>I thought KK's ranking was wrong. I still do.
>Right. I subscribe this. But is it important how KK ranks them ?
No, KK is just a poster of this group. But in case you ask, not at all
of importance. What's important -- Czub and his concentration camps.
>> But it *is* my opinion,
>>as I stated. It is very *interesting* just how much my opinion seems to
>>get you fired up.
>I am not fired up.
>>Is there a reason for that? Damned if I'm going to agree
>>with you "just because"... You can either get over it, get lost, or get
>>real...
>No. The only reason is, that I think Virtual is stronger than you all
>believe.
What was proven at Jakarta. Ok. Point for Czub mclane.
>> I didn't ask for a slew of games to justify Virtual being
>>better than Shredder.
>What a pity. I mean: this would really PROVE it. So you are not
>interested in proving. Aha.
And he is crazy, ill, ignorant (Czub's original notions). How could you
expect him to be interested in something you wanted to prove? :)
Which one is always applauding you with this "Aha"? TC, mclane or simply
Czub?
>This is of course your free will and I cannot say more to this.
>Anyone has to live with his own selected data of the world (of
>computerchess).
If you will allow him ...
Remember, the ideas of concentration camps and signing people for
visible proofs of their disgusting character or behavior. Perhaps you
want to select the data for the people ...
Robert Hyatt
bikejr (bik...@serv.net.nospamme) wrote:
: Komputer Korner <kor...@netcom.ca> wrote in article
For "crafty" I manage three sets of time controls on the servers:
1. crafty plays any bullet.
2. crafty plays blitz up to 5 3 (or 5 5 on chess.net). If I let it
play any blitz, its rating will fluctuate enough that I can't tell whether
I'm going forward or backward in skill. IE compare a 3 0 blitz game with
a 5 12 blitz game and you see what I mean. Also, there is a *tremendous*
amount of cheating on the servers, where two or more humans will observe
a game played by another, and then relay the stuff Crafty whispers to the
player to help him play better. 5 3 makes this nearly imposible...
3. crafty plays any standard time control within reason, but doesn't
whisper anything, to circumvent some of the cheating. I don't let it
play *very* long standard games because a couple of idiots used to
match it in 900 900, and then let it sit all day, before it would flag
them after 900 minutes were up. I got lots of complaints about that
and have cut standard games off at 60 60, max.
One caveat, however, is that a few GM players, and a few IM's that have
asked, have a way of playing Crafty at *any* time control. GM Bill
Lombardy, for example, only plays 5 12 games against Crafty. I don't
worry about GM's cheating, because even if they did, I get as much out
of the game as I would under any circumstances. IM's seem to be the
same (except for one or two bad actors of course).
I did a "poll" a year+ ago, and asked the stronger ICC and chess.net players
what time controls they liked best, most GM's seemed to prefer 5 0 or 5 3,
period. Anything longer (they said) simply wasn't "blitz".
that's the "logic" behind crafty's formula..
I have been after ICC to adopt a new 'Cheating' policy. Now, when they catch
someone using a computer, they simply add a (C). I'm after them to *ban* that
player immediately, with no refund, and to make this policy known to anyone
that joins ICC. Computer cheating is a *huge* problem...
brucemo
mclane wrote:
> Based on this 1 game from jakarta where they had different hardware
> (Computer Schach and Spiele writes Shredder used a PentiumPro200,
> Virtual used a P5/166, others were "forced" to use only P5/133) ?
At these tournaments, hardware is provided, and you use it if you
want. Nobody is forced to use anything, but if you run best on
something they don't provide, you either have to bring your own or
suffer.
Virtual Chess wasn't on a P6. P5/166 sounds about right.
They borrowed a P6 for their playoff round vs Fritz, and they had one
in the blitz tourney, I believe.
bruce
brucemo
Robert Hyatt wrote:
> However, if you make a "blitz chess" list, then Ferret has to be at the
> very top. Check out how it did at Jakarta by winning *every* game in the
> blitz tournament.
I will agree that Ferret is a strong blitz program, but "very top" is too
strong. Genius is very strong at blitz, so is Fritz, and all of the rest
of them can bash you three out of four in any given match.
It is a very different world though when you compare blitz with 40/2. The
stack rank would likely be very different. I wonder if it would look more
like the Swedish list of maybe five years ago?
bruce
mclane
hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
>Yes, but I *did not* say anything negative about Virtual Chess. Rather,
>I said something *positive* about Shredder.
Ever heard of Galilei' principle of relativity ??
It is known since 1640.
>I
>only stated that to win a 12-round bloodfest like Jakarta, the winner
>is going to have to be more than "just lucky" with 12 rounds.
Nobody said something against the event of jakarta !
But if somebody wins in a suisse tournament, this does NOT tell us
that he is the strongest player in the ranking list.
I want to say that he had just luck. That is the opposite of what I
say.
I don't know why you guess the opposite of what I think, but you seem
to suggest that championships are a kind of measurement for playing
strength. I have to deniy this. It is not true.
Championships have their own rules.
Their own luck. And as I said: each program plays only 11 games.
Very often the strong program leads the ranking list, there is a
correlation, I don't want to deny this.
But I don't think that Shredder will lead the swedish elo list, or
eric's list, or shredder should be ranked in KK's list on the first
rank just because YOU think that a championship is much important.
That is my point. Experience and statistics could easily show that you
are wrong. But instead you make this a discussion if you said
something AGAINST virtual. The discussion was how do we rank the
programs. And KK mentioned a list and asked me. And I said, I would
say Virtual is ranked to low. Than you mentioned shredder is
world-champion and therefore higher to rank than nimzo, virtual and
crafty.
If statistics would be that easy, we could forget all about outplaying
games and life would also be easy. Just make some suisse tournaments
and anything would be easy to evaluate.
>: championships is nice idea, but does not have any data behind despite
>: the fact of the jakarta event.
>It has some data. Shredder beat Virtual Chess, Shredder beat Crafty,
>Crafty beat Virtual Chess... and so forth. A conclusion *can* be drawn
>from the above. Perhaps not with 100% confidence, but with pretty high
>confidence.
I cannot resist ! You should better go to school and read again what I
have written !!!
I said: no data behind despite the jakarta event. That is exactly what
you say ! I don't see any reason for your repetition.
Also PRETTY high would say:
Fritz is leading the swedish rating list.
Genius is leading it.
And in fact it was Mchess.
Also shredder will not lead it. Believe me.
>the hardware disadvantage wasn't huge. P5/166 vs P6/200. certainly not a
>factor of 2x, although it isn't equal. I don't know how virtual performs on
>a P6, but if it is a 16 bit engine, the difference was even less...
You have to mutilply these things with the chances of each round.
>I'm notinterpreting, I'm just counting wins and losses in two tournaments.
>Shredder finished with more 1's than Virtual Chess did. I then sort into
>order based on wins/draws/losses and Voila! I have a ranking that is not
>subjective at all...
Right. It is based on 2 tournaments. And ?
Thats not much.
Voila !
Rolf Tueschen
mcl...@prima.ruhr.de (mclane) wrote:
>hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
>>Yes, but I *did not* say anything negative about Virtual Chess. Rather,
>>I said something *positive* about Shredder.
>Ever heard of Galilei' principle of relativity ??
>It is known since 1640.
The pyramids are still much much older, Thorsten Czub.
>>I
>>only stated that to win a 12-round bloodfest like Jakarta, the winner
>>is going to have to be more than "just lucky" with 12 rounds.
>Nobody said something against the event of jakarta !
>But if somebody wins in a suisse tournament, this does NOT tell us
>that he is the strongest player in the ranking list.
>I want to say that he had just luck. That is the opposite of what I
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>say.
~~~~~~
Please write a little bit more "concentrated".
>I don't know why you guess the opposite of what I think,
ROTFL
>but you seem
>to suggest that championships are a kind of measurement for playing
>strength. I have to deniy this. It is not true.
>Championships have their own rules.
>Their own luck. And as I said: each program plays only 11 games.
>Very often the strong program leads the ranking list, there is a
>correlation, I don't want to deny this.
>But I don't think that Shredder will lead the swedish elo list, or
>eric's list, or shredder should be ranked in KK's list on the first
>rank just because YOU think that a championship is much important.
>That is my point. Experience and statistics could easily show that you
>are wrong.
Dear Master Softwaretester. Didn't you write that one single game could
be sufficient and much more of value than hundreds of tests and
stastistics?
>But instead you make this a discussion if you said
>something AGAINST virtual. The discussion was how do we rank the
>programs. And KK mentioned a list and asked me. And I said, I would
>say Virtual is ranked to low. Than you mentioned shredder is
>world-champion and therefore higher to rank than nimzo, virtual and
>crafty.
>If statistics would be that easy, we could forget all about outplaying
>games and life would also be easy. Just make some suisse tournaments
>and anything would be easy to evaluate.
>>: championships is nice idea, but does not have any data behind despite
>>: the fact of the jakarta event.
>>It has some data. Shredder beat Virtual Chess, Shredder beat Crafty,
>>Crafty beat Virtual Chess... and so forth. A conclusion *can* be drawn
>>from the above. Perhaps not with 100% confidence, but with pretty high
>>confidence.
>I cannot resist ! You should better go to school and read again what I
>have written !!!
How could he, Czub, you wrote he's "crazy", "mentally ill"?
>Right. It is based on 2 tournaments. And ?
>Thats not much.
But you are satisfied with one single game ...
Czub, please give us the newest data about Alzheimer. Hahaha.
>Voila !
I see, thanks.
Will you please take back your nazi-like propaganda of signing and
concentrating people in camps?
Thanks.
Rolf Tueschen
Rolf Tueschen
brucemo <bru...@seanet.com> wrote:
>mclane wrote:
>> Based on this 1 game from jakarta where they had different hardware
>> (Computer Schach and Spiele writes Shredder used a PentiumPro200,
>> Virtual used a P5/166, others were "forced" to use only P5/133) ?
>At these tournaments, hardware is provided, and you use it if you
>want. Nobody is forced to use anything, but if you run best on
>something they don't provide, you either have to bring your own or
>suffer.
Bruce, it's obvious, that I'm forced to do this:
do you want to say that the ICCA staff didn't sleep on Hilton beds but
on their own air mattraiss?
Daniel Kang
Robert Hyatt <hy...@crafty.cis.uab.edu> wrote
> I have been after ICC to adopt a new 'Cheating' policy. Now, when they
catch
> someone using a computer, they simply add a (C). I'm after them to *ban*
that
> player immediately, with no refund, and to make this policy known to
anyone
> that joins ICC. Computer cheating is a *huge* problem...
But how do you detect if someone is using a computer or not? And how do
you confirm it?
--
><><><><><><><><><><><><><><><><><><><><><><><
Daniel B. Kang - A 10th grader at SIS (Seoul International School)
who's interested in getting into college after completing junior
year
<><><><><><><><><><><><><><><><><><><><><><><>
My return address is intentionally invalid due to an increasing
number of unwanted junk mails. Please take out a period between
dan and kang. Thanks.
><><><><><><><><><><><><><><><><><><><><><><><
Visit Model World Government page at
http://www.geocities.com/CapitolHill/Lobby/2547/
Robert Hyatt
Daniel Kang (dan....@nuri.net) wrote:
: Robert Hyatt <hy...@crafty.cis.uab.edu> wrote
: > I have been after ICC to adopt a new 'Cheating' policy. Now, when they
: catch
: > someone using a computer, they simply add a (C). I'm after them to *ban*
: that
: > player immediately, with no refund, and to make this policy known to
: anyone
: > that joins ICC. Computer cheating is a *huge* problem...
: But how do you detect if someone is using a computer or not? And how do
: you confirm it?
It is actually not too difficult, but I'm not going to list the steps here
as it would simply provide cheaters with more information on how to avoid
getting caught. I will tell you that with enough games played by a suspected
cheater, it can be judged with nearly 100% reliability when you look at all
the "signatures"...
Robert Hyatt
: >I said something *positive* about Shredder.
: Ever heard of Galilei' principle of relativity ??
: It is known since 1640.
: >I
: >only stated that to win a 12-round bloodfest like Jakarta, the winner
: >is going to have to be more than "just lucky" with 12 rounds.
: Nobody said something against the event of jakarta !
: But if somebody wins in a suisse tournament, this does NOT tell us
: that he is the strongest player in the ranking list.
: I want to say that he had just luck. That is the opposite of what I
: say.
: I don't know why you guess the opposite of what I think, but you seem
: to suggest that championships are a kind of measurement for playing
: strength. I have to deniy this. It is not true.
: Championships have their own rules.
: Their own luck. And as I said: each program plays only 11 games.
: Very often the strong program leads the ranking list, there is a
: correlation, I don't want to deny this.
: But I don't think that Shredder will lead the swedish elo list, or
: eric's list, or shredder should be ranked in KK's list on the first
: rank just because YOU think that a championship is much important.
: That is my point. Experience and statistics could easily show that you
: are wrong. But instead you make this a discussion if you said
: something AGAINST virtual. The discussion was how do we rank the
: programs. And KK mentioned a list and asked me. And I said, I would
: say Virtual is ranked to low. Than you mentioned shredder is
: world-champion and therefore higher to rank than nimzo, virtual and
: crafty.
: If statistics would be that easy, we could forget all about outplaying
: games and life would also be easy. Just make some suisse tournaments
: and anything would be easy to evaluate.
: >: championships is nice idea, but does not have any data behind despite
: >: the fact of the jakarta event.
: >It has some data. Shredder beat Virtual Chess, Shredder beat Crafty,
: >Crafty beat Virtual Chess... and so forth. A conclusion *can* be drawn
: >from the above. Perhaps not with 100% confidence, but with pretty high
: >confidence.
: I cannot resist ! You should better go to school and read again what I
: have written !!!
: I said: no data behind despite the jakarta event. That is exactly what
: you say ! I don't see any reason for your repetition.
: Also PRETTY high would say:
: Fritz is leading the swedish rating list.
: Genius is leading it.
: And in fact it was Mchess.
: Also shredder will not lead it. Believe me.
Take a look at the KK list. It didn't include every program around. I
pointed out that Shredder should be ranked above Nimzo *and* Virtual
Chess, because it finished above them at Jakarta. I've not mentioned the
SSDF, nor genius, nor fritz, nor mchess, nor rebel... Just that it did *not*
belong at the bottom of the list which included two programs it had beaten
(Nimzo and Virtual Chess).
: >the hardware disadvantage wasn't huge. P5/166 vs P6/200. certainly not a
: >factor of 2x, although it isn't equal. I don't know how virtual performs on
: >a P6, but if it is a 16 bit engine, the difference was even less...
: You have to mutilply these things with the chances of each round.
I agree. But private games played on sub-optimal hardware have their own
set of issues and reasons to distrust them. At least at Jakarta, each
author had things set as strongly as he knew how...
: >I'm notinterpreting, I'm just counting wins and losses in two tournaments.
: >Shredder finished with more 1's than Virtual Chess did. I then sort into
: >order based on wins/draws/losses and Voila! I have a ranking that is not
: >subjective at all...
: Right. It is based on 2 tournaments. And ?
: Thats not much.
but it's better than any data KK had to order *his* list. And certainly
*any* data is better than *none*...
: Voila !
brucemo
Robert Hyatt wrote:
> Take a look at the KK list. It didn't include every program around. I
> pointed out that Shredder should be ranked above Nimzo *and* Virtual
> Chess, because it finished above them at Jakarta. I've not mentioned the
> SSDF, nor genius, nor fritz, nor mchess, nor rebel... Just that it did *not*
> belong at the bottom of the list which included two programs it had beaten
> (Nimzo and Virtual Chess).
KK made a stack rank of some programs, and he ranked them based upon Thorsten's
impressions.
The results from Jakarta are another stack rank.
Neither of these is necessarily an indication of how well the programs would do
at any given time control against any given set of opponents. Nor is Thorsten
bound to defer to the Jakarta results as being somehow more significant than
his own impressions.
bruce
Komputer Korner
Robert Hyatt wrote: However, if you make a "blitz chess" list, then
Ferret has to be at the
> very top. Check out how it did at Jakarta by winning *every* game in
> the
> blitz tournament.
>
> Also, using a blitz ranking will turn the typical "program order"
> upside-
> down, as the programs at the top of the SSDF don't play great fast
> chess at
> all...
You forget that Genius 5 wasn't at Jakarta.
--
Best regards
Komputer Korner
The inkompetent komputer
Komputer Korner
Robert Hyatt wrote:
>
>
> I have been after ICC to adopt a new 'Cheating' policy. Now, when
> they catch
> someone using a computer, they simply add a (C). I'm after them to
> *ban* that
> player immediately, with no refund, and to make this policy known to
> anyone
> that joins ICC. Computer cheating is a *huge* problem...
Does that mean that you want them to ban Lonnie?
--
Best regards
Komputer Korner
The inkompetent komputer
Komputer Korner
Robert Hyatt wrote:
>
>
> but it's better than any data KK had to order *his* list. And
> certainly
> *any* data is better than *none*...
>
> : Voila !
I made 2 lists. 1) 32 bit Windows programs and 2) all programs.
Komputer Korner
brucemo wrote:
> Robert Hyatt wrote:
>
> > Take a look at the KK list. It didn't include every program
> around. I
> > pointed out that Shredder should be ranked above Nimzo *and*
> VirtuaKK made a stack rank of some programs, and he ranked them based
> upon Thorsten's
> impressions.
>
> The results from Jakarta are another stack rank.
>
> Neither of these is necessarily an indication of how well the programs
> would do
> at any given time control against any given set of opponents. Nor is
> Thorsten
> bound to defer to the Jakarta results as being somehow more
> significant than
> his own impressions.
>
> bruce
The list that you guys are talking about was my list of 32 bit Windows
programs. I did make another list of the TOP 10 programs that nobody
disagreed with.
Robert Hyatt
: >
: >
: > I have been after ICC to adopt a new 'Cheating' policy. Now, when
: > they catch
: > someone using a computer, they simply add a (C). I'm after them to
: > *ban* that
: > player immediately, with no refund, and to make this policy known to
: > anyone
: > that joins ICC. Computer cheating is a *huge* problem...
: Does that mean that you want them to ban Lonnie?
I don't understand the question. Lonnie has a (C) after his handle. He
never pretended to be a human so far as I remember, and I think I was on ICC
before he started playing. I'm talking about humans, using a computer, but
claiming to not be using a computer. It's ruining server chess...
Rolf W. Tueschen
hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
>Komputer Korner (kor...@netcom.ca) wrote:
>: Robert Hyatt wrote:
>: >
>: >
>: > I have been after ICC to adopt a new 'Cheating' policy. Now, when
>: > they catch
>: > someone using a computer, they simply add a (C). I'm after them to
>: > *ban* that
>: > player immediately, with no refund, and to make this policy known to
>: > anyone
>: > that joins ICC. Computer cheating is a *huge* problem...
>: Does that mean that you want them to ban Lonnie?
>I don't understand the question. Lonnie has a (C) after his handle. He
>never pretended to be a human so far as I remember,
I communicated with him. He's a man of flesh and blood, sorry, ---
champagne.
>and I think I was on ICC
>before he started playing. I'm talking about humans, using a computer, but
>claiming to not be using a computer. It's ruining server chess...
Ol' blue eye' again. Bob, the world still exists for million years now.
Inspite of the cheating factor. You're dreaming of a world without
humans. ICC will survive I assure you. And when the german servers will
allow to afford a nice online standbye I'll join the party with my Deep
Blue, Bob. Knew this?
mclane
brucemo <bru...@seanet.com> wrote:
>bound to defer to the Jakarta results as being somehow more significant than
>his own impressions.
You misunderstand. We have Virtual. We bought it. We play games with
it. We have Shredder. Play games. Impressions ?? RESULTS!
Jakarta is also result. but only a few games ! Also this is not only
my impression, ask Peter Schreiner. I really don't know how you want
to judge without having Virtual.
Komputer Korner
Robert Hyatt wrote:
> : Does that mean that you want them to ban Lonnie?
>
> I don't understand the question. Lonnie has a (C) after his handle.
> He
> never pretended to be a human so far as I remember, and I think I was
> on ICC
> before he started playing. I'm talking about humans, using a
> computer, but
> claiming to not be using a computer. It's ruining server chess...
I was being facetious Bob, but there is a difference because Lonnie is
neither all computer nor all human (2Hirn) and the problem is how does
another computer tell the difference?
--
Best regards
Komputer Korner
The inkompetent komputer
Robert Hyatt
Komputer Korner (kor...@netcom.ca) wrote:
: Robert Hyatt wrote:
: > : Does that mean that you want them to ban Lonnie?
: >
: > I don't understand the question. Lonnie has a (C) after his handle.
: > He
: > never pretended to be a human so far as I remember, and I think I was
: > on ICC
: > before he started playing. I'm talking about humans, using a
: > computer, but
: > claiming to not be using a computer. It's ruining server chess...
: I was being facetious Bob, but there is a difference because Lonnie is
: neither all computer nor all human (2Hirn) and the problem is how does
: another computer tell the difference?
I dont know about computers, but I know how to tell the difference, using
a computer (or computers) and given a few games to look at... Those are
the ones I dislike...
Robert Hyatt
mclane (mcl...@prima.ruhr.de) wrote:
: brucemo <bru...@seanet.com> wrote:
I have a lot of virtual games, and a lot of shredder games, 11 of each
to be specific. IMHO, shredder played better chess. Quite simple concept,
really. Simply look at both the Jakarta games, and the result. They speak
volumes...
brucemo
Tord Kallqvist Romstad wrote:
>
> hy...@crafty.cis.uab.edu (Robert Hyatt) writes:
> >
> > However, if you make a "blitz chess" list, then Ferret has to be at the
> > very top. Check out how it did at Jakarta by winning *every* game in the
> > blitz tournament.
>
> the strongest commercial blitz programs. Of course Ferret might be even
> better, but the blitz tournament in Jakarta doesn't prove much.
Fritz finished second.
Nimzo was in Jakarta but didn't elect to play in the blitz tourney.
bruce
Tord Kallqvist Romstad
"bikejr" <bik...@serv.net.nospamme> writes:
> Why not start a ranking system for blitz chess time controls (or does such
> an animal exist?).
Do you mean a blitz list based on computer-computer games? Such an animal
does indeed exist. The SSDF produces a blitz rating list as well. The list
is published at irregular intervals, usually twice a year. The last list
I have seen was published at the end of last year. Different versions of
Chess Genius occupied the first few places. Even Lang's old Mephisto Lyon
program (on a 68030/36MHz) is close to the top of the list! Not surprisingly,
Fritz is also among the best programs on the blitz list. Rebel, The King
and Kallisto are surprisingly weak at blitz chess.
Tord
Tord Kallqvist Romstad
hy...@crafty.cis.uab.edu (Robert Hyatt) writes:
>
> However, if you make a "blitz chess" list, then Ferret has to be at the
> very top. Check out how it did at Jakarta by winning *every* game in the
> blitz tournament.
True, but Genius, Fritz, WChess and Nimzo were not there. These are, IMHO,
the strongest commercial blitz programs. Of course Ferret might be even
better, but the blitz tournament in Jakarta doesn't prove much.
Tord
Robert Hyatt
Tord Kallqvist Romstad (tor...@gyda.ifi.uio.no) wrote:
: hy...@crafty.cis.uab.edu (Robert Hyatt) writes:
: >
: > However, if you make a "blitz chess" list, then Ferret has to be at the
: > very top. Check out how it did at Jakarta by winning *every* game in the
: > blitz tournament.
: True, but Genius, Fritz, WChess and Nimzo were not there. These are, IMHO,
: the strongest commercial blitz programs. Of course Ferret might be even
: better, but the blitz tournament in Jakarta doesn't prove much.
: Tord
Fritz was there, along with nimzo. Wchess and Genius were not. As a result,
it was not a "weak" speed chess event.
: > Also, using a blitz ranking will turn the typical "program order" upside-
Harald
## Nachricht vom : 11.07.97 weitergeleitet
## Ursprung : /REC/GAMES/CHESS/COMPUTER
## Ersteller war : hy...@crafty.cis.uab.edu
RH> Shredder is the current world microcomputer chess champion. It played
RH> well in Jakarta and finished ahead of everyone there, including Virtual
RH> Chess, Crafty, Ferret, Nimzo and a host of others. Based on *that* I
RH> believe it deserves (has earned) the respect to be listed at the top of
RH> the programs I mentioned. *period*. Virtual Chess also lost in the
Hmm, so some time ago Fritz should have been leading a list of top-
programs just because he won some small worldchampionship? I can't believe
it's you who writes this.
RH> there's no guesswork. Shredder finished ahead of Virtual Chess, Shredder
RH> should be ranked ahead of Virtual Chess.
Rubbish. You know that, I supposed.
Harald Faber
Dave
Mr. Faber,
You may *feel* any way you like regarding the "small" world
championship, current ratings estimate of Mr. Hyatt on chess
software,etc.,
But when you just say **Rubbish**, with not one bit of **logic**,
**data**, or **theory** to back that up, I wonder if you know how
**absolutely stupid** you sound?
If you wish to make some claim or comment about this, please have the
intellectual courage and strength to back it up with **something**!!
It certainly can not be a shock to the NG readers, that the chess
computer that won the world championship is, naturally, considered the
best in the world at that time, under those time controls, using that
hardware, etc.
If you don't believe that's true, tell us why - you may be right, and
we'll all learn something. Just writing "Rubbish", and running away is
not too educational to the chess readers unless they want to know how
to be intellectual weenies, which I doubt is the case.
Regards,
Dave