Terra - a last test and a last bug

[Gábor Szots]

Hi Peter,

Just to make the picture complete, I have done the LCT2 on Terra 3.3. It scored 2535 (P: 43%, C: 78%, E: 65%) on a PIII 933 MHz machine. Hash was set to 48 MB, and 4-piece Nalimov EGTB's were in use.
However, positions 30 and 35 posed a problem running Terra as a UCI engine: after a couple of seconds it crashed. As WB it went OK.

G?bor

[Peter Fendrich]

Hi Peter,

Just to make the picture complete, I have done the LCT2 on Terra 3.3. It scored 2535 (P: 43%, C: 78%, E: 65%) on a PIII 933 MHz machine. Hash was set to 48 MB, and 4-piece Nalimov EGTB's were in use.
However, positions 30 and 35 posed a problem running Terra as a UCI engine: after a couple of seconds it crashed. As WB it went OK.

G?bor

Thank you G?bor,
Yeah I know it is the same old UCi bug...
I just hate to end the projet with a big ugly bug!
I will give the whole story in another post her in this forum.
Regards
/Peter

[Patrick Buchmann]

Hi Peter,

Patrick Beucler write interresting thing in my forum http://f50.parsimony.net/forum200321/messages/9446.htm

Terra 3.3 lost on time with tablebases in Shredder Classic 1.1 but doesn't lost on time if not using tablebases.

Regards,
Patrick

[Peter Fendrich]

Hi Peter,

Patrick Beucler write interresting thing in my forum http://f50.parsimony.net/forum200321/messages/9446.htm

Terra 3.3 lost on time with tablebases in Shredder Classic 1.1 but doesn't lost on time if not using tablebases.

Regards,
Patrick

Hi Patrick,
Yes, Patrick Beucler and I have email contact. I've also sent him a new version that doesn't seem to have any UCI problems. Maybe he found another problem however. I will have a closer look on that, this evening.
/Peter