Author: Tord Romstad
Date: 00:21:09 02/07/06
Go up one level in this thread
On February 06, 2006 at 21:47:09, Ryan B. wrote: >Rather it is to hide extra hidden search or for marketing propaganda Marketing propaganda? What does the nodes/second ratio have to do with marketing? >I do not know but I do know Rybka manipulates its node count. "Manipulates" is a too negatively loaded word. It implies something dishonest, when it is perfectly possible (and IMHO much more likely) doesn't do quite the same kind of work at each node as most other programs. The exact definition of a "node" differs between different programs, and even if Rybka uses the most common definition (one node for each call to search() or qsearch()) the behavior you observe (high tactical strength compared to the N/s, very irregular N/s depending on the position) isn't really that strange. It could be, for instance, that Rybka is moving the pieces around a lot on the internal board at each node in order to decide just how static the node is, and use this to make search decisions. Such internal, "horizontal" searches at each node could easily reduce the N/s count a lot, while preserving tactical strength. It also makes a lot of sense that the N/s is much lower in tactically messy positions. This is just one of many possible explanations, of course. In any case, I think "manipulate" is an unfortunate word to use. Tord
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.