Ridiculously Fast SSD's.. How good for chess?

[frosch]

Ok.
Is it possible to provide a parameter to steer the search depth
when using table base? Just like other programs do?
kind regards
Bernhard

It is easy, but pointless. How would you have any idea what works best in each position? The scheme I use has been tuned/tested over tens of thousands of games, trying various EGTB probe depth limits, to find the setting that is best overalll...

in fact it is often useful. as a chessplayer you know what positions are critical and where early tablebase access will bring better results, i.e. improves the search.

[bob]

Ok.
Is it possible to provide a parameter to steer the search depth
when using table base? Just like other programs do?
kind regards
Bernhard

It is easy, but pointless. How would you have any idea what works best in each position? The scheme I use has been tuned/tested over tens of thousands of games, trying various EGTB probe depth limits, to find the setting that is best overalll...

Are you saying the speed of the medium the TBs are stored on doesn't affect the optimal depth, or does Crafty check the access time for a tablebase probe and adjust the depth dynamically? I would certainly think (perhaps naively) that things would change if I moved the TBs from a slow network drive to a ramdisk.

Depends. On lots of earlier versions, crafty adjusted the probe depth dynamically as it observed the NPS change during the game. More recent versions have been simplified to use something that works well over a wide range of disk performance levels. From the rather slow SATA stuff on my laptop, to SCSI-raid on my office box...

Having a user-settable parameter would be worse at least as often as it would help, which would hurt more than it would help. Even tuning it for one case would not be optimal for the next one on the same machine...

[bob]

Ok.
Is it possible to provide a parameter to steer the search depth
when using table base? Just like other programs do?
kind regards
Bernhard

It is easy, but pointless. How would you have any idea what works best in each position? The scheme I use has been tuned/tested over tens of thousands of games, trying various EGTB probe depth limits, to find the setting that is best overalll...

Well, you could have a go with loading only the 4 pieces into memory.

I only wonder how much it will give in Crafty.

In XiniX, I have the "make qsearch as reliable as possible without too much cost" philosofy, where you have the "qsearch is unreliable anyway, so limit it without too much unreliability" philosofy.

This means that on average, my qsearch will be deeper, and this happens to be the place where most bitbase hits take place.

In addition I would think that if we both hit a bitbase on ply 3 in qsearch, it has more effect in an expensive qsearch (though this is more a feeling, I can't really prove it)

Tony

I would not disagree. I found that probing egtbs in the q-search was just too costly at the speeds I usually see. If I search 16M nodes per second, that could be 8-10M probes per second, and no disk can sustain that...

[bob]

Ok.
Is it possible to provide a parameter to steer the search depth
when using table base? Just like other programs do?
kind regards
Bernhard

It is easy, but pointless. How would you have any idea what works best in each position? The scheme I use has been tuned/tested over tens of thousands of games, trying various EGTB probe depth limits, to find the setting that is best overalll...

in fact it is often useful. as a chessplayer you know what positions are critical and where early tablebase access will bring better results, i.e. improves the search.

That misses the point. Which would you rather do:

(1) start an analysis run with a "good" (typical) setting that will get as deep as possible under those settings...

(2) start an analysis and re-run it several times to find the setting that produces the deepest analysis for that particular position/game, but which will likely not be optimal for other games?

You seem to be saying (2) is better. When in reality, (1) is far better because you could use all the time you spent making multiple "tuning runs" to better analyze the game with just one run. And since the tuning will be specific to that particular game, you will have to repeat this every time you analyze a different game...