Komodo 5 v "The Beast"- 1 game....1,000,000 games?

[geots]

Nothing really here to write home about- normally just another 50-game match. As I had been running a number of different engines ag. Strelka (aka "The Beast") and Komodo's number came up. To be honest, in the middle of these Freeware RRs, I had forgotten how strong Komodo 5 was. Hence this match.

The "oddity"- if that is a good word for it, was the reason for posting these results. As I had also forgotten until I posted these results in my files that I already had a 1000-game match between these 2- with EXACTLY the same controls. And the 1000-game match also ended in < 10 elo difference! Of course, the winner of that match is no more relevant than the winner of this one. I wouldn't touch it- call it a waste of time- but I can't help but wonder the elo difference after 10,000 games.

Maybe it is true after all that we have 3 levels of believability: 1.Lies 2.Damn Lies 3.Statistics (Please, no deep analytical study. I posted it merely for fun. Tho it is dead-on true- still just fun.)





XPS 8500 Intel i5-4 True Cores
Fritz 11 gui
1CPU/64bit
128MB hash
Bases=NONE
Ponder_Learning=OFF
Perfect 2012b.ctg w/12-move limit
4'+2"
Match=50 games

Code: Select all

 XPS-Core-i5, Blitz:4'+2"  12/5/2012  

                     
1   Strelka 5.5 x64   +7    +14/-13/=23   51.00%   25.5/50
2   Komodo 5 64-bit   -7    +13/-14/=23   49.00%   24.5/50

Imagine I have wasted enough of your good time-

gts

[Carotino]

Nothing really here to write home about- normally just another 50-game match. As I had been running a number of different engines ag. Strelka (aka "The Beast") and Komodo's number came up. To be honest, in the middle of these Freeware RRs, I had forgotten how strong Komodo 5 was. Hence this match.

The "oddity"- if that is a good word for it, was the reason for posting these results. As I had also forgotten until I posted these results in my files that I already had a 1000-game match between these 2- with EXACTLY the same controls. And the 1000-game match also ended in < 10 elo difference! Of course, the winner of that match is no more relevant than the winner of this one. I wouldn't touch it- call it a waste of time- but I can't help but wonder the elo difference after 10,000 games.

Maybe it is true after all that we have 3 levels of believability: 1.Lies 2.Damn Lies 3.Statistics (Please, no deep analytical study. I posted it merely for fun. Tho it is dead-on true- still just fun.)





XPS 8500 Intel i5-4 True Cores
Fritz 11 gui
1CPU/64bit
128MB hash
Bases=NONE
Ponder_Learning=OFF
Perfect 2012b.ctg w/12-move limit
4'+2"
Match=50 games

Code: Select all

 XPS-Core-i5, Blitz:4'+2"  12/5/2012  

                     
1   Strelka 5.5 x64   +7    +14/-13/=23   51.00%   25.5/50
2   Komodo 5 64-bit   -7    +13/-14/=23   49.00%   24.5/50

Imagine I have wasted enough of your good time-

gts

The match is on the cutting edge...