DiscoCheck 5.0

Discussion about chess-playing software (engines, hosts, opening books, platforms, etc...)
lucasart
Posts: 201
Joined: Mon Dec 17, 2012 1:09 pm
Contact:

Re: DiscoCheck 5.0

Post by lucasart » Sun Oct 27, 2013 1:40 pm

User923005 wrote:I used your script to build it, of course.
Attached are the results before I renamed the header file and after.
I just updated to GCC 4.8, and managed to reproduce your problem. You forgot the $1 argument:

Code: Select all

$ ./make.sh ./discocheck
which produces an executable called './discocheck'.

I also fixed a bug which GCC 4.8 unveiled. It was an access out of bounds of an array, which happenned to give the correct result with earlier version of GCC. By pure chance there was some extra padding with zero value there, and that happenned to be the value I needed (but not with GCC 4.8 which showed a different bench signature).

If you pull the latest code, and compile as above, it should produce no warning or error messages (I've just done it with GCC 4.8.1 right now). And it should have this signature (node count)

Code: Select all

$ ./discocheck bench
...
signature = 5357949
time = ...
"Talk is cheap. Show me the code." -- Linus Torvalds.

Jeremy Bernstein
Site Admin
Posts: 1226
Joined: Wed Jun 09, 2010 7:49 am
Real Name: Jeremy Bernstein
Location: Berlin, Germany
Contact:

Re: DiscoCheck 5.0

Post by Jeremy Bernstein » Mon Oct 28, 2013 12:41 pm

Here's an OSX 32/64-bit fat binary, should run on all systems 10.5+.

Jeremy
Attachments
discocheck-osx-fat.zip
(683 KiB) Downloaded 240 times

User923005
Posts: 616
Joined: Thu May 19, 2011 1:35 am

Re: DiscoCheck 5.0

Post by User923005 » Mon Oct 28, 2013 8:37 pm

Yes, it works fine now.
...
signature = 5357949
time = 7.46296

Post Reply