Colossus X - how strong today ?
mclane
I still wonder how strong this program could be, if it would have
hash-tables e.g.
The program should be continued. Marty Bryant had a really good
program in the old days, it is a pity that he is not programming chess
anymore.
Has anybody similar or any experience with putting colossus X on a
state-of-the-art hardware and let it play vs. top-programs ?
If not, I could start by posting some nice games ??
Interests in this ?
best wishes
mclane
mclane
hei...@ira.uka.de (Ernst A. Heinz) wrote:
>Thorsten,
>I have recently bought the "Colossus X Game Collection" on CD-ROM
>including "Colossus X Chess" for just *** 5 DM *** at a Hertie store ... :-)
!!!
Hello Ernst !
Maybe I have the same version !!
>The program runs fine on our 233MHz AMD K6 and actually reaches more
>than 100K nps in the middlegame.
Exactly !!
> So, it is quite fast indeed. *But* its
>search trees really blow up enormously such that it rarely exceeds depths
>of 8-9 plies.
Colossus 4.0 for CPC 464 was more selective and maybe stronger.
Bryant made IMO a step backwards with making version X more
brute-force-like. It is a pity. I am sure the Amstrad CPC464 version
no. 4 on a PC could smash many programs away !
When I remember it right, the cpc464 version 4 of Colossus had
separate material and positional evaluations. He showed it always
apart from each other and sacced and played really brilliant. If
somebody has still a cpc464 and colossus or wants to sell it, I am
interested in buying this old machine and colossus back.
>In a series of Nunn test games with a 2:1 time handicap for "DarkThought"
>our baby crushed "Colossus X" all the time. I sincerely doubt that any of
>today's top chess programs score less than 90% against "Colossus X".
You must be careful with the mouse. Don't use the mouse. Use the
keyboard for move-input, because the mouse kills the permanent-brain.
We must get Martin to make version 4 for the PC, than he has maybe a
top program. I guess the version for the amstrad was more selective
because otherwise it would not have searched deep enough. And the PC
version, he thought, oh - i have plenty resources, I can make more
brute-force. What a pity.
>=Ernst=
Thanks for your answer Ernst. It is good to see that other people are
also sometimes that sentimental like me. So i am not that alone with
my nostalgic feelings. On the other hand, you are the person I would
have thought at least to buy colossus. I thought: ernst, as a
materialist, he is more a guy to buy chessBase 7 than to buy colossus.
That would have fitted better into my prejudices.
I am sure the 5,- DM was in charge for your sentimental deal ?
Whatever. Welcome in the club !
best wishes
mclane
Ernst A. Heinz
> I still wonder how strong this program could be, if it would have
> hash-tables e.g.
> The program should be continued. Marty Bryant had a really good
> program in the old days, it is a pity that he is not programming chess
> anymore.
> Has anybody similar or any experience with putting colossus X on a
> state-of-the-art hardware and let it play vs. top-programs ?
Thorsten,
I have recently bought the "Colossus X Game Collection" on CD-ROM
including "Colossus X Chess" for just *** 5 DM *** at a Hertie store ... :-)
The program runs fine on our 233MHz AMD K6 and actually reaches more
than 100K nps in the middlegame. So, it is quite fast indeed. *But* its
search trees really blow up enormously such that it rarely exceeds depths
of 8-9 plies.
In a series of Nunn test games with a 2:1 time handicap for "DarkThought"
our baby crushed "Colossus X" all the time. I sincerely doubt that any of
today's top chess programs score less than 90% against "Colossus X".
=Ernst=
mclane
"Steve Maughan" <maughanD...@bigfoot.com> wrote:
>You might like to look for a Spectrum emulator for the PC and Colossus 4.
>The Spectrum had the same processor as the Amstrad (Z80) and I think the
>latest versions run a fare bit faster than the original Spectrum (3.5 MHz)
Good idea ! But I don't know if colossus4 on spectrum behaves the same
my cpc464 version behaved. Remind that cpc464 had 64 KiloByte ram. And
the spectrum only 16/32.
Maybe the cpc version is stronger ?
Has anybody out there colossus 4 ? If so - for which "machine" ?!
best wishes
mclane
Steve Maughan
Quenton Fyfe
On the subject of reminiscence and Martin Bryant programs, does anyone else
remember White Knight 11 & 12 ? (perhaps only UK people will). They ran on
the BBC Micro, a 6502 based personal computer, running at 2MHz (yes two),
with 32k of RAM (of which you only got to use about 20 if you wanted
graphics as well).
This was some time in the '80s, before IBM PCs ruled the roost. I remember
leaving the machine on overnight, seeing how long it would take to analyse a
million nodes !
No doubt it would be stuffed by everything today - but I'm sure it could
still beat me !
Cheers
Quenton Fyfe
mclane wrote in message <6n0mjc$8sg$1...@steve.prima.ruhr.de>...
>I still wonder how strong this program could be, if it would have
>hash-tables e.g.
>The program should be continued. Marty Bryant had a really good
>program in the old days, it is a pity that he is not programming chess
>anymore.
>Has anybody similar or any experience with putting colossus X on a
>state-of-the-art hardware and let it play vs. top-programs ?
Sean
Where can I buy this Colossus X program???
Where is there a review of it ???
I do not know anything about it!
Best Regards,
Sean
mclane wrote:
mclane
"Quenton Fyfe" <que...@excelsiordirect.com> wrote:
>On the subject of reminiscence and Martin Bryant programs, does anyone else
>remember White Knight 11 & 12 ? (perhaps only UK people will). They ran on
>the BBC Micro, a 6502 based personal computer, running at 2MHz (yes two),
>with 32k of RAM (of which you only got to use about 20 if you wanted
>graphics as well).
My friend had a bbc-b. And I bought an amstrad cpc464 while I was in
britain. This was before the machine was exported to germany.
I remember the cpc had copied the copy-key-feature from the bbc.
My friend had white-knight too. But it had no chance against my
colossus 4.0.
The 6502/2 is not slower than the z80A/4.
The 6502 is better for chess.
Oh man. Where is all the old time gone ?
There we sat, 2 germans, having 2 britain machines. Arround us these
crazy commodore VC64 guys with their toy-computer. And we did the
whole day doing computerchess. We made our own chess-program. I
remember we began with it on the machines we had before.
He had a commodore vc64 and I had a sinclair zx-spectrum.
But it was very difficult to write a chess program in this basic.
The vc64 had cryptic signs for setting the cursor, and the spectrum
had difficult key-shift-system, each key had 3 or even 4 types of
meanings. But than we made a big step forward with bying these
Acorn-bbc and Amstrad cpc464.
Our chess programs made progress. Cpc had a real good and fast basic,
relating to the other machines. The only thing that made me crazy was
the tape-rekorder as data-storage-system. It took ages to load or save
a program. And often azimut-problems made it difficult to reload old
stuff.
How much faster is my today's pc ?!?
The next step/jump came after the cpc464. I bought an ATARI ST. Wow -
this was really a breakthrough in ram, speed and operating-system.
And my omicron/gfa-basic compiler was faaaaaast.
>This was some time in the '80s, before IBM PCs ruled the roost. I remember
>leaving the machine on overnight, seeing how long it would take to analyse a
>million nodes !
>No doubt it would be stuffed by everything today - but I'm sure it could
>still beat me !
>Cheers
>Quenton Fyfe
>mclane wrote in message <6n0mjc$8sg$1...@steve.prima.ruhr.de>...
>>I still wonder how strong this program could be, if it would have
>>hash-tables e.g.
>>The program should be continued. Marty Bryant had a really good
>>program in the old days, it is a pity that he is not programming chess
>>anymore.
>>Has anybody similar or any experience with putting colossus X on a
>>state-of-the-art hardware and let it play vs. top-programs ?
>>If not, I could start by posting some nice games ??
>>Interests in this ?
>>
>>
>>best wishes
>>
>>mclane
>>
best wishes
mclane
mclane
f...@accountant.com wrote:
>Ahem. The Spectrum you are referring to came in two flavours, 16K and 48K
>(*not* 32K) RAM (16K ROM in each).
Right. Your brain reminds it better. I still have my 16K spectrum, and
it still works. But I lost my chess-prgs for it.
>Colossus ran on both types - the same version.
Aha.
>Suggesting that it was virtually the same code for the CPC.
Maybe.
>While of course the greater RAM on the CPC could have been used for hash etc.,
>I bet it wasn't. This stuff was written a long time ago!
>Accordingly, I guess that the only difference in strength can be estimated by
>the MHz ratio and RAM wait state data.
colossus has no hash. I guess it has null-move, but no hash.
As Ernst pointed out, the selectiveness is very ineffective.
But - we have to consider that colossus x was a product from 1988, and
colossus 4 was even years before that. This was one year before the
german-november-revolution !
>Kind regards
>fca
But I still guess that colossus 4 on a pc would be stronger than
colossus x on the same machine. What a pity...
best wishes
mclane
mclane
Sean <spe...@cyberus.ca> wrote:
>Where can I buy this Colossus X program???
>Where is there a review of it ???
>I do not know anything about it!
>Best Regards,
>Sean
Sorry sean, but the programs and the machines we talk about, are a
little rusty-. they are from 1988 and earlier (zx spectrum was 1982).
The program colossus was a famous chess-program with amazing features
in those times. It solved different kind of mates, had a
learning-book, you could edit the book, it had a next-best function,
you were able to select between brute-force and selective search, and
and and and.
Colossus was the son of White Knight from Martin , written for Acorn
BBC. I have had version 4 and later version x for the Atari ST.
So - this is only nostalgic stuff of some oldies..-..
best wishes
mclane
Ernst A. Heinz
Thorsten,
>>In a series of Nunn test games with a 2:1 time handicap for "DarkThought"
>>our baby crushed "Colossus X" all the time. I sincerely doubt that any of
>>today's top chess programs score less than 90% against "Colossus X".
>
>You must be careful with the mouse. Don't use the mouse. Use the
>keyboard for move-input, because the mouse kills the permanent-brain.
Right, I noticed that and I actually very much prefer to watch the
"info" screen over the "board" screen. Still, the 3D CGA graphics
for the chess board are superb if considering how long ago it was done. :-)
>We must get Martin to make version 4 for the PC, than he has maybe a
>top program. I guess the version for the amstrad was more selective
>because otherwise it would not have searched deep enough. And the PC
>version, he thought, oh - i have plenty resources, I can make more
>brute-force. What a pity.
Yes, it would really be nice to have a new "Colossus" for x86.
>Thanks for your answer Ernst. It is good to see that other people are
>also sometimes that sentimental like me. So i am not that alone with
>my nostalgic feelings. On the other hand, you are the person I would
>have thought at least to buy colossus. I thought: ernst, as a
>materialist, he is more a guy to buy chessBase 7 than to buy colossus.
So much for your prejudices w.r.t. "materialistic bean counters" ... :-)
> I am sure the 5,- DM was in charge for your sentimental deal ?
Hmh, yes and no -- I would not have bought it for more than 40 DM.
=Ernst=
mclane
hei...@ira.uka.de (Ernst A. Heinz) wrote:
>Thorsten,
>Right, I noticed that and I actually very much prefer to watch the
>"info" screen over the "board" screen. Still, the 3D CGA graphics
>for the chess board are superb if considering how long ago it was done. :-)
I have tried switched between infinite (method 1=selective and method
2 =brute force) and have come to the same result you came, that the
selective component does not come deep enough if we consider how many
NPS colossus does. But we have to reconsider that this program is an
old program.
>Yes, it would really be nice to have a new "Colossus" for x86.
We should blackmail him ! Maybe we find something to put pressure on
him :-))
Hm ... How can we convince him that programming draughts is shit and
programming chess is the real kings-job ?!?
Maybe we should try to corrupt him with $ ?
>> On the other hand, you are the person I would
>>have thought at least to buy colossus. I thought: ernst, as a
>>materialist, he is more a guy to buy chessBase 7 than to buy colossus.
>So much for your prejudices w.r.t. "materialistic bean counters" ... :-)
Referring your person I have 2 prejudices:
1. you behave like a typical motherson, as a result of this
2. you behave like a typical materialist, as a result of this:
you have problem to lose, you are very precise, rational, logically,
but a force is driving you, the force to be the best, to show anybody
that you are the better guy. Typical for mothersons. Mothersons have
problems to work in a team. Especially to work with other mothersons.
Mothersons can only work with fathersons (hello Markus !:-) and
mothersons prefer always to be right.
So far to my prejudices. Mothersons have also advantages. They are
precise, try to compensate their minority-complex with a heavy amount
of work-overload to get success. The disadvantages: they hate women,
especially intelligent women who could beat them in public!, they need
to tease people, they need to attack those who accept them (mothersons
are sadists, fathersons are masochists)) , they are the majority in
our society.
Motherson and fathersons work in a team, like husband and wife.
Examples: Marx + Lenin, Friedel and Steinwender or : Heinz + Gille.
In the brain of a motherson, typical patterns occur that lead to
typical behaviour as adult.
You are a very typical motherson. It was a pleasure to see you behave
almost exactly how I was in my youth Ernst.
I don't see prejudices as negative. Only prejudices that are secret
and unknown are serious IMO. Ulf Lorenz was a motherson too. But - I
have to admit, that IMO he has changed. I don't know why. But i guess
in his biography something important must have changed because he has
really developed over the years. Normally mothersons do not develop
their personality. They have no development. In opposite to fathersons
who make a development, although the collosion to the
father-ego/person cannot be named a very positive future/development.
I got a father and my first-love. I am not sure what changed Ulf.
Would be interesting to study.
Has anybody before tried to classify you (i hope you don't see this
negatively because on the other hand it also explains you much about
yourself and the reason you have developed THIS way) ?
>> I am sure the 5,- DM was in charge for your sentimental deal ?
>Hmh, yes and no -- I would not have bought it for more than 40 DM.
! :-))
>=Ernst=
best wishes
mclane
Lenspencer
Now that brings back some memories. I have an Atari 800XL (actually several of
them), and I have Colossus 3 and 4 among my collection. C3 was more flexible
(you could edit the position), but C4 could think on your time and it had a
quasi-3D display. Among my others:
Sargon II (Hayden) I bought it on cassette, but once I got a disk drive I found
a way to sector copy the tape to disk rather quickly - loads much faster.
Chess 7.0 (Odesta). I picked this one up for $20 when it was still going for
$80, and I didn't even own a disk drive yet!
Chessmaster 2000 (Software toolworks) Got this one mailorder for $10. Atari
version on one side, Commodore 64 on the other.
Atari Computer Chess (Atari) A ROM cartridge that was given to me a LONG time
ago, also the worst of the commercial programs I have on the Atari.
Chess (Parker Brothers) Yes, the people who brought us Frogger and Qbert also
did a chess program in a cartridge, and it was not that much better than
Atari's version.
Anybody know of any others to add to this grand old collection? :o)
mclane
hei...@ira.uka.de (Ernst A. Heinz) wrote:
>Thorsten,
>> Referring your person I have 2 prejudices:
>>
>> [...] (funny cartoon with amateurish psycho-analytical stuff snipped)
>Thank you very much for the *free* analysis of my "materialistic
>father-son ego"
mother-son Ernst, motherson ! Fatherson are the followers, mothersons
are the leaders. The fathersons are the pacifists, you are the
conquerer. I thought you would like it :-))
>-- I only wish your knowledge were less superficial
>because then the readers could actually profit from your diligent
>postings.
! :-)))
>It always makes me wonder why a self-pronounced "mother-son" like you
>thinks mainly in closed "black & white" categories and shows so many
>overwhelming signs of the "minority complex" which supposedly dominates
>your whole whereabouts.
I have overcome that stage with getting a (step-) father in 1987 Ernst
! This is of course very rare...
But i find it funny today, to identify those who have not made this
progress yet !!
The theory about fathersons and mothersons is from Volker Elis
pilgrim. It is NOT black and white, because there are 4 types (4
combinations of stamping due to 2 kind of sex in the parents setting).
The fathersons and mothersons are only 2 types who are very easily to
identify.
>Fortunately, reality features a lot more colors and nuances than you
>seem to be able to imagine.
In the childhood of a child, there are only few persons stamping it.
Father and mother. Or in other words: male and female persons.
These dualistic possibilities is not invented by me or by pilgrim, it
is nature having created 2 sexes. So don't blame me, blame god.
If he would have created more stamping roles, we would have more
combinations than the 4 we mainly have !
>> The disadvantages: they hate women, especially intelligent women who
>> could beat them in public!
>I cannot let this personal attack against my beloved girl-friend stand
>uncommented.
?!?! Attack ? You are joking ! I don't even know her.
Normally mothersons marry or live with motherdaughters, who are used
to behave the way the leader-motherson likes a women to be...
> She has an M.Sc. ("Diplom") in Computer Science like
>myself and I bet she has more brains in each of her finger-tips than
>most people are blessed with during their whole lives.
This is nice, but has nothing to do with the fact that especially
mothersons don't like to be beaten by women, see Kasparov or Fischer,
2 famous mothersons.
>Hopefully, the alleged process of your own personal development has not
>yet come to a final stand-still.
No - the older i grow, the more I develop into a fatherson. This has
also negative effects (one main negative result is that many thing in
life do not affect you anymore and you see in anything a positive
force and you want to help anybody. Fathersons like to be nice. And
develop a kind of fatalistic mitlaeufertum, especially when they met a
strong leader-person. But - fathersons will end in a collision with
their fathers-ego...
BTW: I think you can see what I mean with categorizing these 2 types
by watching Markus, he is the classical fatherson.
didn't you see the difference between you and him ? He is always nice,
friendly, trying to make peaceful and helpful compromises, you are the
force that is sharp, strong, accurate, and - in the end , superior and
leading. You are both very good examples of these 2 categories. It is
funny to watch, especially when you know that those "marriages"
between fatherson Gille and motherson Heinz is the same Engels and
Marx lived in for years. I guess this was one reason Peter had
problems to integrate into the trio. He seems to be a motherson too.
But 2 mothersons in ONE room ends always in "war" .... :-))
>Best wishes.
Thanks.
>=Ernst=
best wishes
mclane
mclane
Tom Kerrigan <kerr...@io.frii.com> wrote:
>Ernst, how could your analysis of Thorsten claim any sort of accuracy when
>you didn't even ask him what his astrological sign is??
The stamping by the parents has nothing to do with the stars come out
at night, Tom. :-))
Also his analysis is wrong. I have no problems to lose. Or to be
wrong. Or to be unknown. I don't need to see my name in any thread or
to be known as beeing right. :-)
2 bit of parents, gives 00, 11, 01 and 10 types of stamping. And all
degress of in between. But - stars ? Bist du abergläubisch ?
>Cheers,
>Tom
best wishes
mclane
Ernst A. Heinz
Thorsten,
> Referring your person I have 2 prejudices:
>
> [...] (funny cartoon with amateurish psycho-analytical stuff snipped)
Thank you very much for the *free* analysis of my "materialistic
father-son ego" -- I only wish your knowledge were less superficial
because then the readers could actually profit from your diligent
postings.
It always makes me wonder why a self-pronounced "mother-son" like you
thinks mainly in closed "black & white" categories and shows so many
overwhelming signs of the "minority complex" which supposedly dominates
your whole whereabouts.
Fortunately, reality features a lot more colors and nuances than you
seem to be able to imagine.
-----
> The disadvantages: they hate women, especially intelligent women who
> could beat them in public!
I cannot let this personal attack against my beloved girl-friend stand
uncommented. She has an M.Sc. ("Diplom") in Computer Science like
myself and I bet she has more brains in each of her finger-tips than
most people are blessed with during their whole lives.
-----
Hopefully, the alleged process of your own personal development has not
yet come to a final stand-still.
Best wishes.
=Ernst=
Tom Kerrigan
Ernst, how could your analysis of Thorsten claim any sort of accuracy when
you didn't even ask him what his astrological sign is??
Cheers,
Tom
Ernst A. Heinz <hei...@ira.uka.de> wrote:
: Thorsten,
purpl...@my-dejanews.com
mcl...@prima.ruhr.de (mclane) wrote:
>
> I still wonder how strong this program could be, if it would have
> hash-tables e.g.
> The program should be continued. Marty Bryant had a really good
> program in the old days, it is a pity that he is not programming chess
> anymore.
> Has anybody similar or any experience with putting colossus X on a
> state-of-the-art hardware and let it play vs. top-programs ?
> If not, I could start by posting some nice games ??
> Interests in this ?
I find that interesting, I don't have that program, but I have Colossus 4 for
c64 I would be interested if you or someone posted results
Mark
> best wishes
>
> mclane
>
>
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/rg_mkgrp.xp Create Your Own Free Member Forum