New engine releases 2019

[Dann Corbit]

So the last 2 months I've been caught up with the chess programming virus again. Nothing too serious as I don't have much of a test environment (just a quad core laptop). It came to a point where I couldn't do much improvement anymore without doing intensive testing. So here goes the latest version.

Invictus revision 305 now available:

https://github.com/ed-apostol/InvictusC ... s/tag/r305

Last CCRL 40/4 rating is:
Invictus r228 64-bit 4CPU 2376 (+120 vs 1 CPU)!
Invictus r228 64-bit 2256

Latest version should be much stronger!

I had to change getPiece in the include file to get it to build with MSVC++ Visual Studio 17 (I moved the code inline):
inline int position_t::getPiece(int sq) { return pieces[sq]; }
Now I realize I probably could have just decorated the implementation file signature with the word inline.

My non-pext build is a million NPS higher than yours with 11 threads. I have no idea why that should be. I was unable to build with g++, so it might be nice to get a makefile or a cmake text file if you use that

[Edsel Apostol]

So the last 2 months I've been caught up with the chess programming virus again. Nothing too serious as I don't have much of a test environment (just a quad core laptop). It came to a point where I couldn't do much improvement anymore without doing intensive testing. So here goes the latest version.

Invictus revision 305 now available:

https://github.com/ed-apostol/InvictusC ... s/tag/r305

Last CCRL 40/4 rating is:
Invictus r228 64-bit 4CPU 2376 (+120 vs 1 CPU)!
Invictus r228 64-bit 2256

Latest version should be much stronger!

I had to change getPiece in the include file to get it to build with MSVC++ Visual Studio 17 (I moved the code inline):
inline int position_t::getPiece(int sq) { return pieces[sq]; }
Now I realize I probably could have just decorated the implementation file signature with the word inline.

My non-pext build is a million NPS higher than yours with 11 threads. I have no idea why that should be. I was unable to build with g++, so it might be nice to get a makefile or a cmake text file if you use that

Hi Dan, I used your null move formula. Thanks for that. I also use VS 2017. It's weird that you have that problem. I didn't attempt any PGOs, just the default optimization /O2 so that may explain the speed difference.

[Guenther]

So the last 2 months I've been caught up with the chess programming virus again. Nothing too serious as I don't have much of a test environment (just a quad core laptop). It came to a point where I couldn't do much improvement anymore without doing intensive testing. So here goes the latest version.

Invictus revision 305 now available:

https://github.com/ed-apostol/InvictusC ... s/tag/r305

Last CCRL 40/4 rating is:
Invictus r228 64-bit 4CPU 2376 (+120 vs 1 CPU)!
Invictus r228 64-bit 2256

Latest version should be much stronger! :)

Thanks Edsel.

I need to create my own no/popcnt version though, as non-pext is already too advanced for my old quadcore.

[Edsel Apostol]

So the last 2 months I've been caught up with the chess programming virus again. Nothing too serious as I don't have much of a test environment (just a quad core laptop). It came to a point where I couldn't do much improvement anymore without doing intensive testing. So here goes the latest version.

Invictus revision 305 now available:

https://github.com/ed-apostol/InvictusC ... s/tag/r305

Last CCRL 40/4 rating is:
Invictus r228 64-bit 4CPU 2376 (+120 vs 1 CPU)!
Invictus r228 64-bit 2256

Latest version should be much stronger!

Thanks Edsel.

I need to create my own no/popcnt version though, as non-pext is already too advanced for my old quadcore.

There is a NOPOPCNT define on top of bitutils.h. Maybe I'll upload a non-pext no-popcnt version.

[Scally]

So the last 2 months I've been caught up with the chess programming virus again. Nothing too serious as I don't have much of a test environment (just a quad core laptop). It came to a point where I couldn't do much improvement anymore without doing intensive testing. So here goes the latest version.

Invictus revision 305 now available:

https://github.com/ed-apostol/InvictusC ... s/tag/r305

Last CCRL 40/4 rating is:
Invictus r228 64-bit 4CPU 2376 (+120 vs 1 CPU)!
Invictus r228 64-bit 2256

Latest version should be much stronger!

Hi Edsel,

There’s no Makefile or any hint in the README, so how do I compile your engine?

Thanks,

Al.

[Damir]

So the last 2 months I've been caught up with the chess programming virus again. Nothing too serious as I don't have much of a test environment (just a quad core laptop). It came to a point where I couldn't do much improvement anymore without doing intensive testing. So here goes the latest version.

Invictus revision 305 now available:

https://github.com/ed-apostol/InvictusC ... s/tag/r305

Last CCRL 40/4 rating is:
Invictus r228 64-bit 4CPU 2376 (+120 vs 1 CPU)!
Invictus r228 64-bit 2256

Latest version should be much stronger!

Thanks for the new update Edsel.

[Gabor Szots]

- compilation of Sapeli 1.64 added for older hardware (my compile)

Thanks for helping in this, Günther. Sapeli 1.64 64-bit is already on our list with 32 games but it will be almost 300 in this week's update and I'm going to add more games after I have learned exactly how much improvement it shows (huge, that's sure).

[Gabor Szots]

- TomThumb 3.0

It is the opposite, 0.3. I'm not sure I like it as .msi though.

[Dann Corbit]

Here is invictus without PEXT as a PGO build:
https://drive.google.com/open?id=1Zvik9 ... KMQPJyg62R

I changed the default thread count to 11, and the default hash to 16GB, so don't forget to change those in the UCI options if you use this version.

[Guenther]

- TomThumb 3.0

It is the opposite, 0.3. I'm not sure I like it as .msi though.

I was sure it was announced as 3.0 at the time I posted it here, but I cannot
prove it, wayback hasn't saved much snapshots.

Anyhow, you could extract the installiation with e.g. uniextract (I use an old version, which still works for this msi installation).
In the directory Tom Thumb Chess\lib\ there is a jar file named tom-thumb-0.3.jar, but it seems it is not compiled
for working as a standalone program, unlike the 0.2 version I have here.
It always complains about a missing main class tomthumb.app.App, thus there are some dependencies on other files...

The GUI itself would need very long to load here and immediately eats 1GB of ram or more.