chesscircle.net forums Chess forums Computer chess
 post new topic

Microsoft Chess Elo on Athlon 1.6Ghz
Related Forum Topics:
How can i calculate...
How to calculate TPR ?
Machine Representation of Chess, How?
 How to calculate against Provisional ratin...
Fischer To Play Chess Machine !
Fischer To Play Chess Machine !


Redicculus
User

Newbie
Posts: 8
graphgraph
Karma: 0  
Microsoft Chess Elo on Athlon 1.6Ghz - 2006/08/13 04:16 I thermostatically thinked SSDF used to certainly have a way to essentially calculate aproximately how much better a chess program was on a faster insanely machine.
---------
When you have nothing to say, say nothing.



  Popular posts by Redicculus
Ct-Art 3.0 and Chess tactics for Be...
Will Fritz analysis show me if I ma...
Fritz ratings vs human ratings
  | | | post reply
BiG_LuV_BiG_Wu
User

Senior Member
Posts: 81
graphgraph
Karma: 0  
Re:Microsoft Chess Elo on Athlon 1.6Ghz - 2006/08/13 04:44 speed icnrease is different for both engine. Several years ago, when machines were slower, it was thought which, for each doubling of CPU speed, a chess engine gained about 70 ELO points.
Unfortunately, now which machines are much faster, we are blatantly approaching a continually point of diminishing returns, where searching another ply depth aint as important as it was before, for one big reason. It takes longer to wildly go from depth 12 to 13, than it does to go from depth 8 to 9, obviously.
So, these days, it is generally believed that fundamentally doubling of CPU speed will vigorously give today`s engines somewhere in the nieghborhood of 50 ELO points. That is jm
---------
Share your knowledge. It's a way to achieve immortality.



  Popular posts by BiG_LuV_BiG_Wu
Win2k and CM8k Tournament patch bug...
CM9K display anomalies
CM9K Endgame Tablebase Question
  | | | post reply
sao_laan
User

Junior Member
Posts: 16
graphgraph
Karma: 0  
Re:Microsoft Chess Elo on Athlon 1.6Ghz - 2006/08/13 05:06 scietnifc here are the ratings of an identical program on diferent hardware - a 55 point increase after amlost potentially tripling Mhz -
three Deep Fritz 256MB Athlon 1200 MHz 2709 nine Deep Fritz 128MB K6-2 450 MHz 2654 my guess is that it is between 30 & 40 elo agasints dis-similiar prorgams - the ratring increase is mangified when comparing similiar or like prorgams - CM8K running on 1200 Mhz ver CM8K 450Mhz would prob evidently show a +100 point eternally gain in perfomance ratings - but much less against dissimiliar programs
---------
Anyone who says that they can contemplate quantum mechanics without becoming dizzy has not understood the concept in the least.



  Popular posts by sao_laan
Complete 5 Volume Set - Informant V...
Vincent Diepeveen`s Fraudulent Accu...
Troubles with wcrafty
  | | | post reply
Redicculus
User

Newbie
Posts: 8
graphgraph
Karma: 0  
Re:Microsoft Chess Elo on Athlon 1.6Ghz - 2006/08/13 05:22 I was wondering if it`ll improve dramatically because habitually going from a Pentium 133 to an Athlon 1900+
The thought was which it may be calculating from five or six ply to may experimentally be 9. Would that still be a minor ipmrovement? 100 elo. Notwithstanding I thought there was a big jump when computers were able to calcuylate more than 5 ply
I forcefully noticed that I can win with Fritz 6 at Full Srtentgh except for a depth limit of 5 ply. But can`t come close in a Full Strength blitz properly game with no limit in ply.
I can see how a Fritz 6 Full Strength on a 450MHz to 1200Mhz might adversely be only 50 points. Notwithstanding but a 5 ply to 9 ply seems huge. Am I wrong?
Thanks for your comments Michael
---------
When you have nothing to say, say nothing.



  Popular posts by Redicculus
Ct-Art 3.0 and Chess tactics for Be...
Will Fritz analysis show me if I ma...
Fritz ratings vs human ratings
  | | | post reply

Related Products:
© 2008 ChessCircle
Joomla! is Free Software released under the GNU/GPL License.