Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: SSDF should test Aristarch 4.50 instead of endless Crafty XX.YYs !!

Author: Albert Silver

Date: 13:21:03 10/29/04

Go up one level in this thread


On October 29, 2004 at 15:47:28, Sune Fischer wrote:

>On October 29, 2004 at 08:52:00, Albert Silver wrote:
>
>>>>> 3 Deep Fritz 8.0  256MB Athlon 1200 MHz   2790   26   -25   855   72%  2625
>>>>> 6 Deep Fritz 7.0  256MB Athlon 1200 MHz   2769   24   -24   898   66%  2655
>>>>> 7 Fritz 8.0  256MB Athlon 1200 MHz        2759   22   -22  1043   65%  2649
>>>>>10 Fritz 7.0 256MB Athlon 1200 MHz         2744   29   -28   634   63%  2653
>>>>>16 Deep Fritz 256MB Athlon 1200 MHz        2718   30   -29   571   61%  2642
>>>>>28 Deep Fritz 7.0  128MB K6-2 450 MHz      2655   31   -31   498   53%  2637
>>>>>29 Deep Fritz  128MB K6-2 450 MHz          2650   22   -21  1071   58%  2590
>>>>>35 Fritz 7.0  128MB K6-2 450 MHz           2633   38   -37   348   53%  2611
>>
>>>>> 1 Shredder 8.0 CB  256MB Athlon 1200 MHz  2818   34   -32   481   70%  2673
>>>>> 2 Shredder 7.04 UCI 256MB Athlon 1200 MHz 2809   24   -23   967   71%  2648
>>>>> 5 Shredder 7.0  256MB Athlon 1200 MHz     2772   27   -25   801   70%  2624
>>>>>11 Shredder 6.0 Pad UCI 256MB Athlon 1200  2725   22   -22  1033   62%  2642
>>>>>12 Shredder 6.0  256MB Athlon 1200 MHz     2723   31   -30   547   62%  2634
>>>>>19 Shredder 7.0 UCI 128MB K6-2 450 MHz     2699   34   -34   429   54%  2669
>>>>>26 Shredder 5.32  256MB Athlon 1200 MHz    2663   26   -26   739   50%  2661
>>>>>37 Shredder 6.0 UCI 128MB K6-2 450 MHz     2618   43   -43   264   52%  2607
>>
>>>>The comparison is relative to other amateurs.
>>>
>>>Why the comparision to Crafty?  Bob has won two world championships and has been
>>>chess programming longer than most of the other authors have been on the face of
>>>the earth.  I agree that perhaps other "amatuers" should be tested - but not at
>>>the expense of Crafty.  We can go with few less testing of the professionals
>>>(imo) as shown above.  Do we really need 4 different flavors of Frizt 7?
>>
>>I'll defend those 4 flavors actually. For one thing, it is known that Deep Fritz
>>7 is more than an SMP version of Fritz 7, it is an upgrade of the engine, even
>>on a single CPU. Also, the testing on more than one hardware configuration has
>>provided many fascinating results, allowing us to better understand how much is
>>gained by doubling or tripling the raw processing power.
>>
>>I wouldn't take away Crafty of course, which more or less paved the way by being
>>the first amateur to be tested on the list if I'm not mistaken. However, it
>>really would be nice to see some of the other very strong amateurs also tested.
>>
>>                                           Albert
>
>I think it's a waste of time to test deep fritz on single processor systems.
>
>Either you have a dual and want to see how the deep programs do on duals, or you
>don't have a dual and so you would instead buy the version optimized for single
>processor.

I'm afraid you missed the comment that Deep Fritz 7 is not merely a
multi-processor version of Fritz 7, it is an improved engine even on a single
CPU. Of course, that improvement is not worth the double price tag it comes
with.

>
>But also in light of the constant race for amateurs trying to catch up, more
>information on this would be interesting. Are amateurs catching up or do the top
>professionals increase their lead?

Yes, I agree and would like to see some of the amateurs tested as well.

>All in all the flood of fritzes is rather boring, IMO.

I understand, and agree it is overdone. The desire is to keep an impression of
improvement and innovation, but twice already (Fritz 6 and 8) the released
version wasn't really a significant improvement, and needed to wait a number of
patches to show the promised progress.

                                            Albert


>
>-S.



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.