Author: Enrique Irazoqui
Date: 05:08:12 12/28/99
Go up one level in this thread
On December 28, 1999 at 06:55:43, Albert Silver wrote: >On December 27, 1999 at 17:01:58, Enrique Irazoqui wrote: > >>On December 27, 1999 at 16:36:55, Albert Silver wrote: >> >>>On December 27, 1999 at 08:54:48, blass uri wrote: >>> >>>>On December 27, 1999 at 08:08:55, Albert Silver wrote: >>>> >>>>>On December 27, 1999 at 07:42:33, blass uri wrote: >>>>> >>>>>>I am interested to know if there are results of new programs like Junior6 in >>>>>>enrique's test suite >>>>>> >>>>>>Uri >>>>> >>>>>I seem to remember Enrique saying it was 42 points stronger than Junior 5 >>>>>according to his test suite. >>>>> >>>>> Albert Silver >>>> >>>>I remember Enrique saying that Junior5 was 42 points stronger than Junior4.6 >>>>according to his test suite but nothing about Junior6 >>>> >>>>Uri >>> >>>Hmm... Well, I remember a post where Amir challenged the accuracy of a test >>>based on tactics, for such a test would never show the improvements made from >>>Junior 5 to Junior 6, at which point Enrique said that his tests had shown >>>Junior 6 to be 42 points stronger, and smiled. Still, if Enrique shows up, he'll >>>tell us. >>> >>> Albert Silver >> >>The 42 points was the difference between J5 and J4.6. I didn't finish yet >>running the test with J6. So far, J6 is doing better than J5, but just a bit >>better. > >Ach so! I stand corrected. (mumble-grumble...plot....grumble....conspiracy...) :) > It will be interesting then to see if what Amir said about your test is >true. That it will not reveal J6's significant improvements due to their >positional nature, Amir said this, but also referring to the difference between J5 and J4.6. Th e42 points in my test did reveal this difference though. I am also very interested to see how well my test relates to the SSDF list regarding J6. If it doesn't, it will be the first time it fails. Then we will have to figure out why it failed in this particular case and not in the others. If it fails. Enrique > and then compare this with SSDF's results. Actually, even >more interesting will be to see J6's results in the SSDF compared with its >results against human players, as it will presumably have enough games after the >league events next year. > > Albert Silver >> >>Enrique
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.