Nice to see somebody playing there with Toga Checkov Beta 3.1.2 SE:

406. toga ii checkov_beta 3.1.2se 186 2506 n/a 446. 15.01.2008 26.01.2008 32

I wonder what the hardware being used is? The Beta 3, based on Toga II 1.4 Beta5c has not made an appearance yet on Playchess, on a Quad it should be a bit stronger I hope, because Toga Checkov Beta 3.1.2 SE can only use one processor

http://kirr.homeunix.org/chess/discussion-board/viewtopic.php?f=7&t=2992&sid=5be84b0df9c2890235200d6ae60c062f#p30403

Because I have not tested it in games much, I don't have very high expectations of the Beta 3. But I do hope the multithreading was not corrupted by my hacks! Denis Mendoza made a pretty fast compile. In the meantime there is a Toga 1.4 Checkov Beta 4 but that is still a private version here. So it is not in any lists yet either :) You wouldn't believe how heavy pruning I'm using at the moment; NullMove Reduction = 5 is the default! I tried to incorporate Adaptive Null Move pruning but for some reason that did not seem to work very well, possibly because the Verification Search is also adaptive and this works better with fixed Null Move reduction depths. Verification Reduction = 7 but that is a bit illusionary, because in most cases Beta 4 will only use a depth of 5 plies instead of 7. Reduction depths of 5 And 7 where the default is only 3 and 5 respectively!

But without a better evaluation it is not really likely to bring much and possibly could play much worse. With better evaluation it should also be possible to do better pruning, - or more efficient searching, as once was tentatively shown in an experiment with Glaurung, although it is not so easy to verify what effects are synergistic there and what independant effects. But if Larry says that Rybka 3 has a much better eval, in principle it should be possible to translate that to a better search as well. I see no reason for any diminishing returns myself, apart from more games ending in a draw, but that is also dependant on playing style and type of search I think, otherwise more than 90 % of correspondence games would be drawn.

I'm no correspondence player like you Harvey, what do you think; is correspondence chess close to being drawn only at the top?

Regards,
Eelco 

I think I got my Fritz 9 for something like $10 awhile back at discount at a software store.

I think that the reason is that we're dealing with two different types of engine room people.  On the one hand, you have the casual people who don't really care about rating so much--they just like seeing engines duel it out and such.  These types of people are not what one would call enthusiasts--they aren't constantly updating engines.  On the other hand, people who used Fritz 9 have upgraded to Fritz 10 and Fritz 11 by now.

Another reason could be that some people who are actually using far stronger engines are naming their engines as Fritz 8 to trick people, especially centaurs.  The Fritz 8 numbers are rising far faster than they were six months ago.

I don't know what you mean (and it's not April Fool's Day anymore :-) ).  I can still use Fritz 9 on my machine, and I've had it longer than a year.  I think it's just that the types of people who buy Fritz 9 are more serious computer chess players, and have moved on to Fritz 10 and 11, not to mention Rybka, Zappa Mexico, and Hiarcs, while those who have Fritz 8 are not very serious.  Also, many people probably like to use Fritz 8 to test opening books.

This is where the contention of playing favorites comes in and the bullshit that comes out of too much moderation. It get to be a joke. I see Mike is back- Good for him!!!