TalkChess.com

mwyoung wrote:My testing agrees with that conclusion, I guess you will wait for the next official release before testing the stronger versions of stockfish?

Nice work Stockfish team if they do take the top spot on your list.

Hi Mark,

I expect Stockfish DD 64-bit to establish a slight edge over Houdini 4 64-bit in our 40/40 1CPU and 4CPU lists.
Komodo is likely to be further back in third spot.

The Stockfish testing framework will only see it become stronger still.
It's going to be tough work for Robert and Larry/Mark to get past them in the near future.

Graham.

Graham Banks wrote:

mwyoung wrote:My testing agrees with that conclusion, I guess you will wait for the next official release before testing the stronger versions of stockfish?

Nice work Stockfish team if they do take the top spot on your list.

Hi Mark,

I expect Stockfish DD 64-bit to establish a slight edge over Houdini 4 64-bit in our 40/40 1CPU and 4CPU lists.
Komodo is likely to be further back in third spot.

The Stockfish testing framework will only see it become stronger still.
It's going to be tough work for Robert and Larry/Mark to get past them in the near future.

Graham.

I agree with that, but it is the competition that drives the innovations in engines we love to test. We have that very much in common.

Peace...

And I hope no one attacks your results....

mwyoung wrote:

Graham Banks wrote:

mwyoung wrote:My testing agrees with that conclusion, I guess you will wait for the next official release before testing the stronger versions of stockfish?

Nice work Stockfish team if they do take the top spot on your list.

Hi Mark,

I expect Stockfish DD 64-bit to establish a slight edge over Houdini 4 64-bit in our 40/40 1CPU and 4CPU lists.
Komodo is likely to be further back in third spot.

The Stockfish testing framework will only see it become stronger still.
It's going to be tough work for Robert and Larry/Mark to get past them in the near future.

Graham.

I agree with that, but it is the competition that drives the innovations in engines we love to test. We have that very much in common.

Peace...

And I hope no one attacks your results....

Don't worry. All testers get criticised at times. Not nice though, especially when you're doing things properly and trying to be impartial.

To be honest, I'd not really noticed what had happened in your thread before you drew it to my attention earlier.

Main thing is that we have fun doing what we do, as you alluded to.

Happy to put the past behind us if you're willing to do so. Many things change over time. My apologies for our harsh exchanges earlier, as I'd not intended it to turn out that way.

Regards,
Graham.

beram wrote:Stockfish DD is now best with 66,49%

I just hope the other camps can muster enough resources to keep it competitive going forward. Nice work Bram.

Graham Banks wrote:

mwyoung wrote:

Graham Banks wrote:

mwyoung wrote:My testing agrees with that conclusion, I guess you will wait for the next official release before testing the stronger versions of stockfish?

Nice work Stockfish team if they do take the top spot on your list.

Hi Mark,

I expect Stockfish DD 64-bit to establish a slight edge over Houdini 4 64-bit in our 40/40 1CPU and 4CPU lists.
Komodo is likely to be further back in third spot.

The Stockfish testing framework will only see it become stronger still.
It's going to be tough work for Robert and Larry/Mark to get past them in the near future.

Graham.

I agree with that, but it is the competition that drives the innovations in engines we love to test. We have that very much in common.

Peace...

And I hope no one attacks your results....

Don't worry. All testers get criticised at times. Not nice though, especially when you're doing things properly and trying to be impartial.

To be honest, I'd not really noticed what had happened in your thread before you drew it to my attention earlier.

Main thing is that we have fun doing what we do, as you alluded to.

Happy to put the past behind us if you're willing to do so. Many things change over time. My apologies for our harsh exchanges earlier, as I'd not intended it to turn out that way.

Regards,
Graham.

Yes I agree, and happy testing. Peace!

mwyoung wrote:

Graham Banks wrote:

mwyoung wrote:

Graham Banks wrote:

mwyoung wrote:My testing agrees with that conclusion, I guess you will wait for the next official release before testing the stronger versions of stockfish?

Nice work Stockfish team if they do take the top spot on your list.

Hi Mark,

I expect Stockfish DD 64-bit to establish a slight edge over Houdini 4 64-bit in our 40/40 1CPU and 4CPU lists.
Komodo is likely to be further back in third spot.

The Stockfish testing framework will only see it become stronger still.
It's going to be tough work for Robert and Larry/Mark to get past them in the near future.

Graham.

I agree with that, but it is the competition that drives the innovations in engines we love to test. We have that very much in common.

Peace...

And I hope no one attacks your results....

Don't worry. All testers get criticised at times. Not nice though, especially when you're doing things properly and trying to be impartial.

To be honest, I'd not really noticed what had happened in your thread before you drew it to my attention earlier.

Main thing is that we have fun doing what we do, as you alluded to.

Happy to put the past behind us if you're willing to do so. Many things change over time. My apologies for our harsh exchanges earlier, as I'd not intended it to turn out that way.

Regards,
Graham.

Yes I agree, and happy testing. Peace!

Thanks Mark.

I would tend to agree.

Actually, though, to me 'speed' is relatively unimportant. The mechanical/technological side of things is ever increasing. At some point, what might be reached in 5 min now, will be reached in 2 seconds.

To me, ultimately, the most important thing is the 'quality' of the individual moves in how they relate to given positions. This is why I have always preferred certain engines over say those that might be able to 'whip' another engine is bullet time controls.

leavenfish wrote: the 'quality' of the individual moves

agree 100%; I had just PM'd someone saying the same thing.

ouachita wrote:

leavenfish wrote: the 'quality' of the individual moves

agree 100%; I had just PM'd someone saying the same thing.

I used to write my thoughts on this (in agreement etc.) and was always ignored in this forum board.
I HAVE realized what everyone keeps saying, that it needs huge amounts of games to estimate strength comparisson to be the fact, if that's all we are doing. And as a generalization.
But I always thought there could be a different criteria in which certain aspects of the moves can be analized so that an even clearer picture can be gained regarding comparative elo, over a much smaller number of games.

e.g. I NEVER heard anyone say that Spaasky might have been a huge amount stronger than Fisher, from the world champiomship games in 1972, even though it was only the ridiculously small number of about 20 games which should have meant absolutely nothing.

PGN - http://kirill-kryukov.com/chess/discuss ... p?id=31100