Winboard Forum

Scidlet 3.61 by Shane Hudson
http://scid.sourceforge.net/

Windows x64 & win32 Intel compiler 10 p.g.o builds
+ parallelized test builds.

Big speed gains over standard build included with Scid.
x64 version almost twice as fast & 32 bit not far behind.

Available from my homepage.

Jim.

Jim Ablett wrote:Scidlet 3.61 by Shane Hudson
http://scid.sourceforge.net/

Windows x64 & win32 Intel compiler 10 p.g.o builds
+ parallelized test builds.

Big speed gains over standard build included with Scid.
x64 version almost twice as fast & 32 bit not far behind.

Available from my homepage.

Jim.

Yes Jim. I also had x64 success with this one. But I'm not sure if my source was for 3.6.1 or 3.6. i diddn't paid much attention before. I used the old source by Dann Corbit from his past archives.

Mizar is also a great project. I learned much on it's website publications.

Jim Ablett wrote:Scidlet 3.61 by Shane Hudson
http://scid.sourceforge.net/

Windows x64 & win32 Intel compiler 10 p.g.o builds
+ parallelized test builds.

Big speed gains over standard build included with Scid.
x64 version almost twice as fast & 32 bit not far behind.

Available from my homepage.

Jim.

Hi Jim, could you do the same to Scidlet 3.6?
I am asking because(if my memory doesn't fail?) Scidlet 3.61
was quite weaker than 3.6. I don't remember exactly, if the reason
was the implementation of IID by Shane or Dann himself, thus I am
not sure anymore if 3.61 ever was an official release by Shane.


Best wishes,
Guenther

Hi Guenther,

Scidlet 3.61
was quite weaker than 3.6. I don't remember exactly, if the reason
was the implementation of IID by Shane or Dann himself, thus I am
not sure anymore if 3.61 ever was an official release by Shane.

If IID was the only change from 3.6 to 3.61, you
can switch it off by adding

Code: Select all

useIID = false

to the 'scidlet.ini' file.

I'll have a look at the code differences. If there are any other changes,
I'll re-compile 3.6.

regards,
Jim.

Any updates on the strength issue Jim?

Tony Thomas wrote:Any updates on the strength issue Jim?

Hi Tony,

I haven't noticed any drop in strength in my blitz testing, but it could
be that speed increase in my compiles over the previous ones is enough
to cancel out the negative IID performance or the problem only affects
longer time controls. Actually my IID build is performing better against earlier
version for me.

Jim.