New release: rofChade 2.3

[Wolfgang]

CEGT 40/4 test here (RofChade 2.3): https://cegt.forumieren.com/t1286-testing-rofchade-2-3
and here (PeSTO 2.210): https://cegt.forumieren.com/t1287-testing-pesto-2-210

[jdart]

I don't like that dynamically linked stuff at all, it's a remnant of the past, at that time storage and memory were so small that it sounded like a good idea.

I agree. I always link statically on Windows. A lot of my users are not technical and don't know what a .dll is or how to download the MSVC redistributable. So I try to make things bulletproof for them.

[Gary Internet]

Question for Ronald. I'm watching Rofchade 2.217 play at TCEC in the Cup. How close is that version to the newly released 2.3?

[Chessqueen]

Question for Ronald. I'm watching Rofchade 2.217 play at TCEC in the Cup. How close is that version to the newly released 2.3?

I believe they are very close in rating since it was release within a few days. but what makes the huge difference is that super hardware used in TCEC, compare to the one used here CCRL

[Ronald]

Question for Ronald. I'm watching Rofchade 2.217 play at TCEC in the Cup. How close is that version to the newly released 2.3?

Very close. There is only one difference between the two versions. The TCEC version contains an Opposite Colored Bishop correction, which seemed to work against Xiphos, but I was not sure in general, so I removed it from 2.3. I've been trying OCB before but I still can't seem to get it to work positively for rofChade. I think will try a more general imbalance approach in the future. It looks like rofChade it benefitting a little more from the TCEC hardware than Xiphos, single threaded Xiphos is stll stronger in my tests.