SCCT Rating List - Rybka 4.1 NO-SSE x64 6c

[Sedat Canbaz]

Hello Friends,

Last Updated Engine:
Rybka 4.1 NO-SSE x64 6c

SCCT Rating List - All Versions:
http://www.sedatcanbaz.com/chess/scct-rating/

Some Notes:
-Rybka 4.1 NO-SSE x64 6c = Deep Rybka 4.1 is played as default (the version,which does not support SSE4.2)
-So far, Rybka's performance is quite good,let's see what will be the standings after 1000 games per player

Best Regards,
Sedat

[Sedat Canbaz]

New Update:
Rybka 4.1 NO-SSE x64 6c's performance after 618 games:
http://www.sedatcanbaz.com/chess/scct-rating/


Best,
Sedat

[Modern Times]

Hi Sedat, so it is 3 Elo behind the SSE version ?

[Sedat Canbaz]

Hi Sedat, so it is 3 Elo behind the SSE version ?

Correct !

Best,
Sedat

[Sedat Canbaz]

Dear Friends,

After 1000 games:
Rybka 4.1 NO-SSE x64 6c performed just one (1) Elo less than Deep Rybka 4.1 SSE4.2 x64 6c

SCCT Rating List - All Versions:
http://www.sedatcanbaz.com/chess/scct-rating/

Greetings,
Sedat

[Sedat Canbaz]

Rybka 4.1 NO-SSE x64 6c's test is resumed...
And 239 games will be played more
In the end,Rybka NO-SSE version will have 1500 games per player

*The current Elo difference:4

Code: Select all

Rank Name                       Elo     +    – games score oppo. draws
14 Rybka 4.1 SSE42 x64 6c       3275   11   11  1781   60%  3206   48%
18 Rybka 4.1 NO-SSE x64 6c      3271   13   13  1261   61%  3193   49%

SCCT (3m+2s) Rating List-All Versions:
http://www.sedatcanbaz.com/chess/scct-rating/

Best,
Sedat

[Sedat Canbaz]

Rybka 4.1 NO-SSE Test is over:

Code: Select all

Rank Name                          Elo    +    - games score oppo. draws
  14 Rybka 4.1 SSE42 x64 6c       3276   13   13  1800   59%  3212   49%
  18 Rybka 4.1 NO-SSE x64 6c      3275   14   14  1500   60%  3204   49%

Some Notes:
Even after 1500 games per player,
Rybka 4.1 No-SSE version performed with same Elo difference as before:just one (1) Elo less than Rybka 4.1 SS4.2
*Note also that Rybka NO-SSE version (at 1000 games per player) was scored with same Elo difference

This test (playing 500 games more) was useful also about SCCT's Elo bar error margins
As we see.... mostly of the participants Elo difference (Error bar margins) dont change a lot...
That Elo difference is going mainly around minus plus approx.3-5 Elo points and rarely 5-10 Elo points
*Note:i mean for Elo error margins especially for after 1000 games per player

For more details:
http://rybkaforum.net/cgi-bin/rybkaforu ... ?tid=25552

SCCT Rating List - All Versions:
http://www.sedatcanbaz.com/chess/scct-rating/


Best Regards,