Author: Albert Silver
Date: 04:49:38 12/19/98
Go up one level in this thread
On December 18, 1998 at 16:59:20, Ralph E. Carter wrote:
>On December 18, 1998 at 16:52:01, Komputer Korner wrote:
>
>>On December 18, 1998 at 16:39:56, Ralph E. Carter wrote:
>>
>>>On December 18, 1998 at 16:18:52, Herbert Gindl wrote:
>>>
>>>>
>>>>I compared Genius 5 under W98 with the so called Genius 6:
>>>>(K6-2 400 128MB)
>>>>Both: 32 MB hash, no permanent brain, no book, level infinite
>>>>
>>>>After
>>>>
>>>>1.a4
>>>>
>>>>Genius5 shows: 1.Sf6 (1/20) 9-21 0.27 00.38
>>>>d4d5Sf3Lf5Lg5e6e3Lb4
>>>>
>>>>1. Sf6 (1/20) 10-22 0.24 1:55 Sf6d4d5Sf3Lf5Lg5Se4e6
>>>>
>>>>
>>>>Now to Genius 6:
>>>>
>>>>1. a4
>>>>Genius6 shows: as above, absolute no difference.
>>>>
>>>>Now I know for sure why there is no Genius5 engine included.
>>>
>>>Genius 5 is an excellent program.
>>>But this is one of those facts, which IF true, must be made known to all
>>>prospective customers.
>>>This is a legitimate function of chess newsgroups.
>>>Thank you.
>>
>>This is not conclusive that the engines aren't different. A bunch of positions
>>are needed to prove that.
>>--
>>Komputer Korner
>
>Right! Let's investigate!
>
>(This is why I capitalized the word "IF", and left the question mark on the
>thread title.)
>
>mongrel
Why not try and run both on a test suite such as Louguet II and compare the
results. EVen if the times are not identical (due to optimizations) if all the
results are proportionately the same you will have more evidence. Otherwise they
are different. Remember, that an upgrade doesn't mean the program was re-written
from head to toe.
Albert Silver
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.