I Am Running Strelka 5.6 tests- Promise Accurate Results!

[geots]

And d/tella's run continues somewhat. Still 500 games to go, but at some point Strelka needs to get back on track.




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
5'+3"
Match=1000 games

Code: Select all

XPS-Core-i5, Blitz:5'+3"  2/27/2013


d/tella ve0 x64   +16    +128/-104/=268   52.40%   262.0/500   
Strelka 5.6 x64   -16    +104/-128/=268   47.60%   238.0/500 

Until then-

george

[Tomcass]

Wooops!

This d/tella is really strong!. Thanks for this exciting test, George.

Tom.

[geots]

Wooops!

This d/tella is really strong!. Thanks for this exciting test, George.

Tom.

Thanks for your reply and interest, Tom. I was just coming to mention that at the next posting of results, it is a good chance I will reveal the name of the author!

And I follow and read every result you post. Every single one! I apologize for not replying more often. Your work is impeccable.


Regards,

george

[geots]

Should not be long now.


gts

[jd1]

Wooops!

This d/tella is really strong!. Thanks for this exciting test, George.

Tom.

Thanks for your reply and interest, Tom. I was just coming to mention that at the next posting of results, it is a good chance I will reveal the name of the author!

And I follow and read every result you post. Every single one! I apologize for not replying more often. Your work is impeccable.


Regards,

george

You've got me interested too

Jerry

[geots]

Wooops!

This d/tella is really strong!. Thanks for this exciting test, George.

Tom.

Thanks for your reply and interest, Tom. I was just coming to mention that at the next posting of results, it is a good chance I will reveal the name of the author!

And I follow and read every result you post. Every single one! I apologize for not replying more often. Your work is impeccable.





Regards,

george

You've got me interested too

Jerry

Hang around maybe an hour or so and you will know. And maybe it will help to know you will be able to use it yourself- but what you can get will be stronger than this, I would think.



Best,

george

[geots]

Well, I think it is time to go ahead and let everyone know the engine they are seeing the results from against Strelka 5.6. Be sure you understand the author had nothing to do with deciding on the secrecy- other than okaying it. And the decision to come clean, so to speak, was mine also. After okaying it thru the author, of course. I make no decisions without clearing them thru him first. Nothing was done just to fool people, but I did it to protect the man I am testing for and his engine.

My theory was simple- keep it under wraps until I see what it can do. OK, the time is now. I think you know the author(s)- Don Dailey and his partner Larry Kaufman. And the engine is one of their latest betas. I don't have to tell you what it can do- you are looking at the results:

Game 585 mark:




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
5'+3"
Match=1000 games

Code: Select all

XPS-Core-i5, Blitz:5'+3"  2/28/2013 


Komodo 4536.00 64-bit   +19    +155/-123/=307   52.70%   308.5/585
Strelka 5.6 x64         -19    +123/-155/=307   47.30%   276.5/585

Strangely, neither of these engines is a loser in this case. Look above at all the engines Strelka 5.6 destroyed before it met k-4536.



See you in 80 or 90-

george

[Tomcass]

What a surprise, George! Both engines are very strong. Let's wait for this promising new Komodo. Strelka 5.6 is a big reference. IMHO Don and Larry should be confident with their work and release it as soon as possible. What a long time waiting for this release...

Regards,

Tom.

... and thanks for your kind comment above, my friend.

[Ajedrecista]

Hi George:

Well, I think it is time to go ahead and let everyone know the engine they are seeing the results from against Strelka 5.6. Be sure you understand the author had nothing to do with deciding on the secrecy- other than okaying it. And the decision to come clean, so to speak, was mine also. After okaying it thru the author, of course. I make no decisions without clearing them thru him first. Nothing was done just to fool people, but I did it to protect the man I am testing for and his engine.

My theory was simple- keep it under wraps until I see what it can do. OK, the time is now. I think you know the author(s)- Don Dailey and his partner Larry Kaufman. And the engine is one of their latest betas. I don't have to tell you what it can do- you are looking at the results:

Game 585 mark:




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
5'+3"
Match=1000 games

Code: Select all

XPS-Core-i5, Blitz:5'+3"  2/28/2013 


Komodo 4536.00 64-bit   +19    +155/-123/=307   52.70%   308.5/585
Strelka 5.6 x64         -19    +123/-155/=307   47.30%   276.5/585

Strangely, neither of these engines is a loser in this case. Look above at all the engines Strelka 5.6 destroyed before it met k-4536.



See you in 80 or 90-

george

No surprise at all! There are other public tests of Komodo betas in ImmortalChess Forum, as you surely know:

The Battle of the Century - Houdini vs. Komodo

I wish good luck to Komodo team. Thanks for sharing results, George!

Regards from Spain.

Ajedrecista.

[beram]

Hi George:

Well, I think it is time to go ahead and let everyone know the engine they are seeing the results from against Strelka 5.6. Be sure you understand the author had nothing to do with deciding on the secrecy- other than okaying it. And the decision to come clean, so to speak, was mine also. After okaying it thru the author, of course. I make no decisions without clearing them thru him first. Nothing was done just to fool people, but I did it to protect the man I am testing for and his engine.

My theory was simple- keep it under wraps until I see what it can do. OK, the time is now. I think you know the author(s)- Don Dailey and his partner Larry Kaufman. And the engine is one of their latest betas. I don't have to tell you what it can do- you are looking at the results:

Game 585 mark:




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
5'+3"
Match=1000 games

Code: Select all

XPS-Core-i5, Blitz:5'+3"  2/28/2013 


Komodo 4536.00 64-bit   +19    +155/-123/=307   52.70%   308.5/585
Strelka 5.6 x64         -19    +123/-155/=307   47.30%   276.5/585

Strangely, neither of these engines is a loser in this case. Look above at all the engines Strelka 5.6 destroyed before it met k-4536.



See you in 80 or 90-

george

No surprise at all! There are other public tests of Komodo betas in ImmortalChess Forum, as you surely know:

The Battle of the Century - Houdini vs. Komodo

I wish good luck to Komodo team. Thanks for sharing results, George!

Regards from Spain.

Ajedrecista.

[/size]

Thx George, for publishing these results
This in general 50-50 result between Strelka 5.6 and Komodo 4536 is very much in line with what Larry said two weeks earlier in the TCEC forum. Namely that the Komodo dev. version had almost chased down Houdini 2, by 20 ELO
So still a long way to go from here to Houdini 3
And I dont think that Robert Houdart is just sitting and waiting for Komodo to clear the gap

grts Bram