KOMPUTER KORNER'S GOLD MEDAL AWARDS
Komputer Korner
Gold Medal Chess Software by category - Version 10
1) Chess instruction for beginner- Chess Mates
2) Chess instruction intermediate level - Maurice Ashley Teaches Chess
3) Chess instruction advanced level - Chess Mentor 1.4
4) Chess Diagram Piece Font - Hastings Alpine Electronics Font
5) Chess Diagram Piece Font- Freeware - TASCBV Chess figurines
6) Internet chess program interface - Slics 2.3j
7) Email chess recording program - Chess Recorder 6.0
8) Chess playing engine strength- Rebel 8.0
9) Chess playing engine - best learning features- M-Chess Pro 6.01
10) Chess playing engine - mass market value line- Chessmaster 5000 1.02
*
11) Chess playing engine - freeware - Crafty 11.9
12) Database program top of the line- ChessBase 6.0
13) Database program -value priced- Clubmate 2.23
14) Opening book editor - Bookup WIN 1.5.2
15) Best chesstree based database- Tascbase 2.0
16) All around best chess program - Nimzo 3.0
* This is conditional on bug in program sometimes playing instantly
being confirmed in patched version. If that is the case, then Extreme
Chess will assume the award. I HAVEN'T SEEN GENIUS 5 YET, SO IT MAY
SURPRISE ME but I was a beta tester for Hiarcs 5. So this list should be
up to date except for Genius and Chess Academy which I haven't gotten
around to look at yet. Chess Academy is an instructional program.
--
The komputer that kouldn't keep a password safe from
prying eyes, kouldn't kompute the square root of 36^n,
and kouldn't find the real motive in chessbase.
brucemo
It might be better if you follow the example of the International Olympic
Committtee, and only give out gold medals once every four years.
bruce
Robert Hyatt
brucemo (bru...@nwlink.com) wrote:
This will probably raise one heck of a firestorm, but first impressions of
Genius 5 are not good. I've been watching it play Crafty on ICC, and, at
least to this observer, it is not playing as well as genius 4. Of course,
this could be due to operator influence as well as anything else, so until
I get more data, I'll not say more. I'm hoping the new trend is not to
release a new version that beats the other "competitors" (commercial programs
on the SSDF) but, in reality, plays worse overall. I still think that CM5000
is the best of the commercial engines, from my own observations over what is
now more than 100 games... It's playing very nice chess...
Moritz Berger
On 5 Dec 1996 02:04:52 GMT, hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
>This will probably raise one heck of a firestorm,
Right !
> but first impressions of
>Genius 5 are not good. I've been watching it play Crafty on ICC, and, at
>least to this observer, it is not playing as well as genius 4.
Do you realize how much weight your comments have? I think that
you should be very careful with quick assessments like above. You should
know best that play against a single opponent (i.e. Crafty) isn’t enough
to estimate overall playing strength.
> Of course,
>this could be due to operator influence as well as anything else, so until
>I get more data, I'll not say more.
Why do you say anything at all if you don’t have enough facts?
The server ratings suffer much from inaccuracies like playing on one machine
via Alt+Tab (hi Lonnie !), since most programs can only use minimal
hash tables
this way. Also, depending on the multitasking behavior of programs,
they will give
too much CPU time to other processes, e.g. the telnet client to ICC.
Add to this
operator influence at faster time controls and you’ll see how meaningful your
statements about Genius 5 really are. Sorry for flaming you Bob, you usually
know very well what you’re talking about, but here I have to disagree
with you.
> I'm hoping the new trend is not to
>release a new version that beats the other "competitors" (commercial programs
>on the SSDF) but, in reality, plays worse overall.
> I still think that CM5000
>is the best of the commercial engines, from my own observations over what is
>now more than 100 games... It's playing very nice chess...
I always thought that CM5000 would suffer badly from it’s inadequate opening
book. - I also remember several posts here that CM5000 on ICC accepts only
certain time controls and opponents that favor the rating of the
program. Wasn’t
it you who always warned about this (together with human operators who decide
not to accept matches from players they lost too many games against?).
Playing strength at 2 12 time controls is no good measurement of a programs
strength at longer time controls. Rebel 8 is a good example, I think
that it does
much better at action chess or longer time controls. And it also needs bigger
hash tables for this purpose!
Let’s wait for some games against human opponents at reasonable time controls
and then have a look again at Genius’ playing strength.
Moritz
Robert Hyatt
Moritz Berger (Moritz...@msn.com) wrote:
: On 5 Dec 1996 02:04:52 GMT, hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
:
: >This will probably raise one heck of a firestorm,
:
: Right !
:
: > but first impressions of
: >Genius 5 are not good. I've been watching it play Crafty on ICC, and, at
: >least to this observer, it is not playing as well as genius 4.
:
: Do you realize how much weight your comments have? I think that
: you should be very careful with quick assessments like above. You should
: know best that play against a single opponent (i.e. Crafty) isn’t enough
: to estimate overall playing strength.
Correct, and I didn't do that. Look at what you quoted above. I clearly
indicated that this was based on results obtained by Crafty, and then later
also explained that I had no idea how the operator was using CG5, and how
he/she has it set up.
In reading my first paragraph, I really should have phrased it like this:
This will probably raise one heck of a firestorm, but my first
impression of genius 5.0, based only on games it has played against
Crafty, is that genius 5.0 is not playing as well as genius 3 and 4
were playing. Note that this says *nothing* about the overall skill
level of the program nor about how it will play against people. It's
only based on how it "matches up" with Crafty. Perhaps I'm doing
something in Crafty that takes advantage of some unknown weakness in
Genius 5, or the two programs simply are different enough that Crafty
does well against it, primarily based on the luck of the differences in
the evaluation or search. The (dreaded) SSDF list will be a much more
reliable estimate of its playing strength since there are more opponents
in the mix. Should it play near the top, I'll simply consider myself
lucky and not change anything for a while. :)
:
: > Of course,
: >this could be due to operator influence as well as anything else, so until
: >I get more data, I'll not say more.
:
: Why do you say anything at all if you don't have enough facts?
Because I've been asked dozens of times, both in the newsgroup and
via email. I have a lot of data for this program... and you notice
that I carefully qualified my comment that this was results against
Crafty *only*. That's the only opinion I form, because I don't watch
commercial programs play each other on the server.
The only "facts" I don't have is how the operator is running CG5. If
he does like "AARKVARK" and overrides it on many occasions, if he forces
moves when he shouldn't, etc. And I'll probably never know that.
:
: The server ratings suffer much from inaccuracies like playing on one machine
: via Alt+Tab (hi Lonnie !), since most programs can only use minimal
: hash tables
: this way. Also, depending on the multitasking behavior of programs,
: they will give
: too much CPU time to other processes, e.g. the telnet client to ICC.
: Add to this
: operator influence at faster time controls and you’ll see how meaningful your
: statements about Genius 5 really are. Sorry for flaming you Bob, you usually
: know very well what you’re talking about, but here I have to disagree
: with you.
Probably was something I should not have written quite like I did. However,
my original tenet still stands here, that against Crafty, for whatever reason,
genius 5 seems to be playing worse than what I've come to expect from genius
3 and genius 4, but *only* in games vs Crafty. Note the *only* as that is
"the frame of reference" for my opinion. It could well beat kasparov for all
I know. Any of the following points are absolutely possible, but I'll
leave their "proof" as an exercise for the reader: :)
1. genius 5 is far superior to genius 4 in normal OTB play.
2. the genius 5 I've seen is not being operated in such a way that it plays
the best it is capable of. We've had dozens of discussions on ICC channel
64 about this subject. I didn't even notice it until others started asking
me what was going on. In fact, I was not aware that CG5 was even available
yet, shows what I know. :)
3. genius could be running on a slow machine. It's purportedly on a P5/166,
which is not bad of course. Crafty's on a P6/200, which is well over 2x
faster. However, there are other CG4's on P5/166's that are definitely
stronger when playing Crafty than cg5 seems to be. By "slow" I mean that
there are P5/166's and P5/166's. some are faster than others, some have
more RAM, some have EDO RAM and others don't. Some have a big L2 cache,
others don't, etc.
4. the operator might be using a poor opening book. Lang has always had
one of the best books around, but operator selection might be causing some
problems there...
In short, all I can respond to is the queries I've gotten about my
impression of how it looks. And to each person that I've responded via
email, I've been careful to point out that the "operator" can exert
great influence on the strength of the program, in both positive and
negative ways.
:
: > I'm hoping the new trend is not to
: >release a new version that beats the other "competitors" (commercial programs
: >on the SSDF) but, in reality, plays worse overall.
: > I still think that CM5000
: >is the best of the commercial engines, from my own observations over what is
: >now more than 100 games... It's playing very nice chess...
:
: I always thought that CM5000 would suffer badly from it’s inadequate opening
: book. - I also remember several posts here that CM5000 on ICC accepts only
: certain time controls and opponents that favor the rating of the
: program. Wasn’t
: it you who always warned about this (together with human operators who decide
: not to accept matches from players they lost too many games against?).
Yes indeed. However, Lonnie plays Crafty all the time with his gaggle of
programs. On equal machines, with Lonnie operating, CM5K seems to be the
best *against Crafty* of what he has, which is pretty dang all-inclusive
when you listen to him. He's also a good operator and doesn't waste a lot
of time. I'm not basing my opinion on CM5K from *ratings* on the servers,
rather I base it on watching the games it plays against Crafty (or, on
occasion, against Ferret). Ratings on the servers are way out of line
anyway. The "mean" might be close, but boy the variance.
I can post results against most anyone you'd like to see (vs crafty) but
Lonnie is difficult, because I never know which program he is running.
Here's results against Lonnie with a variety of programs, running on
chess.net: (ego-crusher is CG5 on a P5/166, one of several that are on
ICC.) Lonnie might shed more insight into which program(s) are doing
best "against crafty" (again) but there was a time when CM5000 was
winning nearly every game. That's changed a lot as I've worked on
mistakes I saw Crafty making in those games, but the record speaks for
itself.
opponent win draw loss
Lonnie 56 24 81 (P6/200, CM5000, Rebel8, etc.)
Ego-crusher 8 6 2 (P5/166, CG 5.0)
:
: Playing strength at 2 12 time controls is no good measurement of a programs
: strength at longer time controls. Rebel 8 is a good example, I think
: that it does
: much better at action chess or longer time controls. And it also needs bigger
: hash tables for this purpose!
:
: Let’s wait for some games against human opponents at reasonable time controls
: and then have a look again at Genius’ playing strength.
:
: Moritz
As I said, it might play wonderfully against other programs and/or against
humans. If you've followed this newsgroup, I've been a proponent of Genius
for several years, and I've repeatedly posted that I think it is the best
commercial program around. However, against Crafty, CM5000 has become the
"thorn in my side" for the past few months, but I'm not convinced that CM5000
is better against people than genius. It's just that the early genius 5 results
were very surprising. I expect a month or two's worth of headaches when each
new program first comes out, and usually I get 'em. For some reason, this time,
with genius 5, I didn't...
Of course, everything I've said is unscientific rambling, based on a total of
maybe 100 games against genius 5. For all I know, using autoplay on two P6/200
machines, Crafty might well lose every game. I don't have a clue, other than
the results I have personally seen. Time will certainly tell, because you can
take 4 copies of Genius 4 on ICC, and one *always* seems stronger than the
other 3. Whether the human is actually "helping" or not is unknown, but the
difference between two "computer opponents" that are using the same program
can be astounding.
Peter W. Gillgasch
brucemo <bru...@nwlink.com> wrote:
> Komputer Korner wrote:
> >
> > Gold Medal Chess Software by category - Version 10
[ snip ]
> It might be better if you follow the example of the International Olympic
> Committtee, and only give out gold medals once every four years.
Amen to that. I was toying with a real mean followup put Bruce hit the
nail on it's head as we say in Kermany :)
-- Peter
john quill taylor
brucemo <bru...@nwlink.com> wrote:
>Komputer Korner wrote:
>>
>> Gold Medal Chess Software by category - Version 10
>>...
>It might be better if you follow the example of the International Olympic
>Committtee, and only give out gold medals once every four years.
On the other hand, a well-done list would have to go through many
revisions. Look at all the new versions of software coming out each
month!
It might be better to have this kind of information on a web site.
So, KK, do you have a web site yet? (Well, neither do I...)
I'd be happy to have you in my killer bookmarks file ;-)
(a little asymmetric friday humour meant in jest)
____________________________________________________________________
Where do I want to go today? Someplace where there are no computers.
- jqt -
Lonnie Cook
On 5 Dec 1996 02:04:52 GMT, hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
>brucemo (bru...@nwlink.com) wrote:
>: Komputer Korner wrote:
>: >
>: > Gold Medal Chess Software by category - Version 10
>: >
>: > 1) Chess instruction for beginner- Chess Mates
>: > 2) Chess instruction intermediate level - Maurice Ashley Teaches Chess
>: > 3) Chess instruction advanced level - Chess Mentor 1.4
>: > 4) Chess Diagram Piece Font - Hastings Alpine Electronics Font
>: > 5) Chess Diagram Piece Font- Freeware - TASCBV Chess figurines
>: > 6) Internet chess program interface - Slics 2.3j
>: > 7) Email chess recording program - Chess Recorder 6.0
>: > 8) Chess playing engine strength- Rebel 8.0
>: > 9) Chess playing engine - best learning features- M-Chess Pro 6.01
>: > 10) Chess playing engine - mass market value line- Chessmaster 5000 1.02
>: > *
>: > 11) Chess playing engine - freeware - Crafty 11.9
>: > 12) Database program top of the line- ChessBase 6.0
>: > 13) Database program -value priced- Clubmate 2.23
>: > 14) Opening book editor - Bookup WIN 1.5.2
>: > 15) Best chesstree based database- Tascbase 2.0
>: > 16) All around best chess program - Nimzo 3.0
>: >
>: > * This is conditional on bug in program sometimes playing instantly
>: > being confirmed in patched version. If that is the case, then Extreme
>: > Chess will assume the award. I HAVEN'T SEEN GENIUS 5 YET, SO IT MAY
>: > SURPRISE ME but I was a beta tester for Hiarcs 5. So this list should be
>: > up to date except for Genius and Chess Academy which I haven't gotten
>: > around to look at yet. Chess Academy is an instructional program.
>:
>: It might be better if you follow the example of the International Olympic
>: Committtee, and only give out gold medals once every four years.
>:
>
>This will probably raise one heck of a firestorm, but first impressions of
>Genius 5 are not good. I've been watching it play Crafty on ICC, and, at
>least to this observer, it is not playing as well as genius 4. Of course,
>this could be due to operator influence as well as anything else, so until
>I get more data, I'll not say more. I'm hoping the new trend is not to
>release a new version that beats the other "competitors" (commercial programs
>on the SSDF) but, in reality, plays worse overall. I still think that CM5000
>is the best of the commercial engines, from my own observations over what is
>now more than 100 games... It's playing very nice chess...
>
I have to agree with Bob on the cg5 level of play and M being objective. Here is
a game with cg5 being the opposition. Cm5K totally outmuscled it. Crafty has
done the same thing.
[Event "ICC 3 12 12/02/96"]
[Site "Internet Chess Club"]
[Date "1996.12.02"]
[White "Ego-crusher"]
[Black "DoctorWho"]
[WhiteElo "2669"]
[BlackElo "2623"]
[Result "0-1"]
[ECO "B10"]
[NIC "CK.01"]
[LongECO "Caro-Kann: anti-anti-Caro-Kann defense"]
1. e4 c6 2. c4 d5 3. cxd5 cxd5 4. exd5 Nf6 5. Bb5+ Nbd7
6. Nc3 g6 7. d4 Bg7 8. d6 exd6 9. Qe2+ Qe7 10. Qxe7+ Kxe7
11. Nf3 Nb6 12. O-O Be6 13. Re1 h6 14. Bd2 Rhc8 15. Rac1 a6
16. Bd3 Nbd5 17. Nxd5+ Nxd5 18. Bc2 Kd7 19. h3 f5 20. Bb3 Rxc1
21. Rxc1 b5 22. h4 Re8 23. a4 Rc8 24. Rxc8 Kxc8 25. axb5 axb5
26. Kh2 Kc7 27. Bc2 Ne7 28. Kg3 Nc6 29. Be3 Nb4 30. Bd1 Bc4
31. b3 Be6 32. Be2 Kc6 33. Bd1 Nd5 34. Bc2 Nc3 35. Ng1 Bd5
36. h5 g5 37. Bxf5 Bxb3 38. Kh2 b4 39. f4 gxf4 40. Bxf4 Bd1
41. Bd2 Bxd4 42. Bxh6 Be5+ 43. Kh3 Bxh5 44. g4 Bf7 45. Nf3 b3
46. Bc1 Nd1 47. Nxe5+ dxe5 48. Kg3 b2 49. Bxb2 Nxb2 50. g5 Bg8
51. g6 Nc4 52. g7 Kd6 53. Kg4 Ke7 54. Kg5 Kf7 55. Be4 Kxg7
56. Kg4 Be6+ 57. Kg3 Kf6 58. Bg2 Ne3 59. Be4 Bd5
{White resigns} 0-1
Lonnie
If Satan ever loses his hair,
there'll be hell toupee.
Komputer Korner
Peter W. Gillgasch wrote:
>
> brucemo <bru...@nwlink.com> wrote:
>
> > Komputer Korner wrote:
> > >
> > > Gold Medal Chess Software by category - Version 10
>
>
> > It might be better if you follow the example of the International Olympic
> > Committtee, and only give out gold medals once every four years.
>
> nail on it's head as we say in Kermany :)
>
> -- Peter
Some of us are concerned with helping the newbies and others who need
information and guidance in selecting computer software and then there
are others like Gillgasch that like to trash everything that anybody
says.
--
Komputer Korner
The komputer that kouldn't keep a password safe from
prying eyes, kouldn't kompute the square root of 36^n,
kouldn't find the real motive in ChessBase and missed
the real learning feature of Nimzo. Long live Nimzo!!!!!!
Peter W. Gillgasch
Komputer Korner <kor...@netcom.ca> wrote:
> Some of us are concerned with helping the newbies and others who need
> information and guidance in selecting computer software and then there
> are others like Gillgasch that like to trash everything that anybody
> says.
Nothing against "helping newbies" but what the hell are FAQs for ?
If you want to give "guidance" then post your stuff once a month
or post a pointer to a website or someting... [if your opinions /
"gold medals" don't change on a daily basis that should be good
enough, right ?].
And then there are others that like to see their "name" in the
"author" field of every 2nd post... Usenet is not a "when will
I become famous" machine... Maybe we should think about the
creation of rec.games.chess.computer.customer-support or something.
*I* read this group for discussion of "how to create/improve" a
program and not for endless discussion of "what button to press
in chess software XYZ" or "long live XYZ stuff" or "my 27th dogma about
computer chess".
As I said quite often some of your posts lead to valuable discussion
[and I like those] and when a new program/version comes out I like
to read your tests, but this endless postings of your personal
favourites condensed in meaningless/subjective lists is simply
*abuse* of the net for the sole purpose of giving your chipset an
ego boost 8^) Imagine how ugly the net would be if everybody who
has a vague interest in any hardware/software category posts his list of
personal favourites to the net...
"Information and guidance" ? Nah, tell that your chip designer 8^) You
think that this is rec.games.chess.computer.advocacy or something and
this is the reason why you are going onto the nerves of many people who
look for information content and discussions about ideas and design.
*That* is the reason why you get flak by some technical folks lately.
-- Peter
Chris Whittington
Komputer Korner <kor...@netcom.ca> wrote:
>
> Peter W. Gillgasch wrote:
> >
> > brucemo <bru...@nwlink.com> wrote:
> >
> > > Komputer Korner wrote:
> > > >
> > > > Gold Medal Chess Software by category - Version 10
> >
> > [ snip ]
> >
> > > It might be better if you follow the example of the International Olympic
> > > Committtee, and only give out gold medals once every four years.
> >
> > Amen to that. I was toying with a real mean followup put Bruce hit the
> > nail on it's head as we say in Kermany :)
> >
> > -- Peter
>
> information and guidance
Novel use of the words 'information' and 'guidance'.
Don't you really mean 'muddled thinking' and 'confusion' :)
Chris Whittington
> in selecting computer software and then there
> are others like Gillgasch that like to trash everything that anybody
> says.
Komputer Korner
James Garner wrote:
>
> Komputer Korner (kor...@netcom.ca) wrote:
>
> : The komputer that kouldn't keep a password safe from
> : prying eyes, kouldn't kompute the square root of 36^n,
> : kouldn't find the real motive in ChessBase and missed
> : the real learning feature of Nimzo. Long live Nimzo!!!!!!
>
Komputers have to wear their mistakes on their sleeves or actually
storage sites.