TalkChess.com

This is Stockfish 1.7.1

You can download sources + binaries + Salvo Spitaleri's new
official book from:

http://homepages.tesco.net/henry.ablett/jims.html

Direct links:
http://dl.dropbox.com/u/5047625/stockfish-171-ja.zip
http://dl.dropbox.com/u/5047625/Stockfi ... 20Book.zip

This version is 100% functionally and speed equivalent with 1.7 so NO
ELO CHANGE and DON'T NEED TO RE-TEST if you have already tested 1.7


This versions removes the new and broken Hyper Threading detection
code and falls back on the well known 1.6.3 behaviour regarding
counting the CPU you have in your machine.We want to stress that you
can _always_ override what SF found by setting the correct value of
"Threads" UCI parameter to the number of physical cores of your
computer.

This version works around an issue in zugzwang positions where 1.7
fails to detect zugzwang. To avoid to break compatibility with 1.7 we
have NOT fixed the issue by default, but we have added an UCI
parameter called "zugzwang detection" that is OFF by default and when
set to on makes SF to correctly detect zugzwang, this could alter ELO,
although not more then 1-2 points, but because we want 1.7.1 to be
100% identical to 1.7 we have disabled this by default.


In the next future version this option will be removed and "zugzwang
detection" will always be enabled."


Marco, Joona, Tord

Thanks to all Stockfish team. I like computer chess for 31 years ago, as I recall, is the first time I see a free program between the two best in the world.

I have HT off on my pc as this option is better for other engines that I use. Fot The latest Stockfish Version, would 8 cores be better than 4?

Thanks

Thanks

Maria

I wonder how do you know that
"zugzwang detection" cannot alter ELO by more then 1-2 points

I am sure that you did not play enough games to test it at long time control.

Thanks!

This works without any problems on my netbook. (The universal version wasn't necessary). With this version the "Threads" parameter shows in Shredder Classic.

Uri Blass wrote:I wonder how do you know that
"zugzwang detection" cannot alter ELO by more then 1-2 points

I am sure that you did not play enough games to test it at long time control.

You are right Uri, actually I have started a regression test but after 500 games results are a bit strange and we will need to redo the test.

I said few points because the change impact is very low, I have measured the impacted nodes and is less then 1% of cases....but real games results seem to show another story and we have to better analyze it, anyhow 1.7.1 is setup to be used with default parameters by testers so that there is no difference with 1.7

I would like to emphasize that although this version is supposed to be functionally equivalent to 1.7 with the default settings, there is one little issue to be aware of:

As explained above, this version reverts to the 1.6.3 code for detecting the number of CPU cores. This means that it doesn't choose the right number of threads by default on CPUs with hyperthreading, like my Core i7 iMac. If Stockfish doesn't detect the correct number of CPU cores on your machine, it is important not only to adjust the number of search threads, but also the "Minimum Split Depth" parameter. On a CPU with 4 or fewer cores, this parameter should be set to 4. On an eight-core machine, it should be set to 7.

mhalstern wrote:I have HT off on my pc as this option is better for other engines that I use. Fot The latest Stockfish Version, would 8 cores be better than 4?

I am not sure whether I understand your question. Did you mean to ask "...would 8 threads be better than 4"?

For optimal strength, the number of threads should always equal the number of physical CPU cores (for Stockfish, and probably also for all other chess engines). If you have a quad-core machine (with or without hyperthreading), use four threads. Also remember to set "Minimum Split Depth" to 4 if you have a quad-core.

PauloSoare wrote:Thanks to all Stockfish team. I like computer chess for 31 years ago, as I recall, is the first time I see a free program between the two best in the world.

Actually it is the first time that the _officially_ best in the world is between two free (and open sourced, that is far more sensible then being free) programs