Author: Chuck
Date: 05:24:56 08/16/02
Go up one level in this thread
On August 16, 2002 at 07:27:47, Gian-Carlo Pascutto wrote: >On August 16, 2002 at 07:00:40, Chuck wrote: > >>It seems to me the tree "explosion" is still a big, big problem and the author's >>claim was false. My evidence suggests Hiarcs is crippled at deep analysis. > >Before jumping to this conclusion, I assume you *did* check whether it's >analysis speed actually dropped, i.e. it takes an extraordinary *factor* of >time more to complete another ply than otherwhise? > >I agree with the other poster that is is likely just a display problem. > >-- >GCP What I have observed is not only a nps problem but the programs' output halts as well. In the example I quoted above, the program generated a 15 ply evaulation at 8 hours and 8 minutes. At 46 hours and 49 minutes, the program has yet to generate anything further and is still searching 16 ply. This is acceptable for any one case but has been my constant experience on at least 3 computers. So the nps is irrelevant, as the key point that Hiarcs is not usable at very deep iterations is based more on the fact that no output is generated. Additionally, on the same computers all the other programs continue to generate analysis at more regular intervals.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.