Author: Thorsten Czub
Date: 11:54:34 11/27/05
Go up one level in this thread
On November 27, 2005 at 14:03:44, Steve B wrote:
>ok i did not know this situation ..
>so there might have been many prototypes made for him
>
>but why do you think its a Rebel?
>
>you have seen my post on the Rebel test for the BT2450
>the solve times are not consistent with Eriks MMM 3.7 and the search routines
>are different
>
>are you saying not only the hardware in the prototypes Louman used was different
>but the program as well??
>
>Steve
yes. I think i have found the "solution" and evidence in the Louwmann test
article printed in CSS 4/86
page 12 says e.g.
it was difficult for schroeder to transfer his program from Z80 to 6502.
and H+G only wanted to give him 4 KByte Ram.
but he needed 8 KByte.
in september 1985 rebel did not reached the wishes playing strength.
but 1986 H+G gave him another chance.
now with 8 KByte Ram.
he says he played 500 test games !!
he said page 13:
when the program was finished, the selective chess program
had 19 KByte, 8 for the very big OS 5 Kbyte for opening book.
19+8+5=32 = 2x16 EPROM
rebel cologne was born.
in 21 month ed had invested 4000 h programming.
and jan had outplayed 1400 test games !!
he said :
ALL GAMES HAVE BEEN PLAYED ON 3.6 mhz hardware.
[you can imagine that louwmann could not have played this
with 1 module. and you can ALSO imagine that he could not have had MANY
boards in his home. it is therefore IMO very possible that he used the many
modules in MOBILE hardware]
For the end customer version 5 mhz will be used.
But i found more EVIDENCE:
on page 14 of this report he writes about the OS and features of this module,
and here he indeed says:
showing the search depth of the selective and the brute force part !!!
but we have more evidence :
again on page 14 he says:
at least i give you the results of 176 testgames rebel played
between 26.4. until 4.6.1986.
and now listen:
THESE TESTGAMES HAVE BEEN PLAYED WITH 2 OTHER DUTCH TESTERS WITH 40/120
so we can estimate that at least 3 modules must have been
circulating !
IMO more because i don't think louwmann used only 1.
but together with the 2 other testers we have maybe 5 units.
Of course THIS VERSION was not 100% identical with the Rebel5 module that
came on the market.
thats for sure.
i am sure the commercial version had a different
software.
it can also be that the prototypes were stronger, because the rebel5 module that
was sold on the market was a disapointment to the testers.
while louwmann said he had the following results with his prototypes
rebel - amsterdam 20-16
- avantgarde 15-9
- par excell 16-8
- novag expert 17-7
- mmII 17-7
- Turbostar KSO 15.5-8.5
- chess player 12.5-7.5
= 113-63
if we can copy the eproms, we can make this version available to ALL people
having MM2 and B+P modules.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.