Dmitri, I suggest that if you want to see the magnitude of the problem (and help fix them) that you try to compile Firenzina (there are IDEs for it, such as Code::Blocks IIRC).
Hint: the error messages are over 10,000 lines long.
Under Linux, that is? Wow! I did not realize that, because I didn't have a chance to try. Will help fix, once I figure out how.
As I said, use Code::Blocks - it supports MSVC++'s compiler, as well as GCC (from MinGW), and since the intel compiler is compatible with MSVC++, probably it too.
It supports concurrebt builds - which will make compiling much faster.
It will even take your Visual C++ project file.
I suggest you give it a try - I'm tempted to try it myself.
Dmitri, I suggest that if you want to see the magnitude of the problem (and help fix them) that you try to compile Firenzina (there are IDEs for it, such as Code::Blocks IIRC).
Hint: the error messages are over 10,000 lines long.
Under Linux, that is? Wow! I did not realize that, because I didn't have a chance to try. Will help fix, once I figure out how.
In actual fact, there are 22,005 lines of error message. THAT is how bad the situation is. I can send it to you if you like.
http://dgusev.cs.edinboro.edu/GameDev/C ... LP_x64.zip This is a quick fix / slight improvement. While working on the 32-bit compile, I noticed that LargePages were commented out in the original code. The change increases kNodes/sec somewhat.
Gusev wrote:http://dgusev.cs.edinboro.edu/GameDev/C ... LP_x64.zip This is a quick fix / slight improvement. While working on the 32-bit compile, I noticed that LargePages were commented out in the original code. The change increases kNodes/sec somewhat.
I have now released two generic no-popcnt 64-bit Windows compiles of Firenzina 2.2.2 xTreme at http://dgusev.cs.edinboro.edu/GameDev/C ... nzina.html (Archive No. 3 on the list.) One of the compiles (LP) supports Large Pages, the other one doesn't. The strength of the generics is ~+7 Elo over Fire 2.2 xTreme GH 64bit.
32-bit compiles are still being tested. Once I am done with them, I plan to update the code archive (small changes had to be made on the 32-bit side). Then I will download Matthews's version with Linux stuff and try to compile it.
Gusev wrote:I have now released two generic no-popcnt 64-bit Windows compiles of Firenzina 2.2.2 xTreme at http://dgusev.cs.edinboro.edu/GameDev/C ... nzina.html (Archive No. 3 on the list.) One of the compiles (LP) supports Large Pages, the other one doesn't. The strength of the generics is ~+7 Elo over Fire 2.2 xTreme GH 64bit.
32-bit compiles are still being tested. Once I am done with them, I plan to update the code archive (small changes had to be made on the 32-bit side). Then I will download Matthews's version with Linux stuff and try to compile it.
I wish you the best of luck - I was tired and swapped the vales of true and false (true should be 1 and false should be 0 in fire.h).
Gusev wrote:I have now released two generic no-popcnt 64-bit Windows compiles of Firenzina 2.2.2 xTreme at http://dgusev.cs.edinboro.edu/GameDev/C ... nzina.html (Archive No. 3 on the list.) One of the compiles (LP) supports Large Pages, the other one doesn't. The strength of the generics is ~+7 Elo over Fire 2.2 xTreme GH 64bit.
32-bit compiles are still being tested. Once I am done with them, I plan to update the code archive (small changes had to be made on the 32-bit side). Then I will download Matthews's version with Linux stuff and try to compile it.
I wish you the best of luck - I was tired and swapped the vales of true and false (true should be 1 and false should be 0 in fire.h).
Matthew:out
Maybe it'll give us a strong competitor of Moron 1.0.
