Author: Djordje Vidanovic
Date: 09:25:41 10/30/03
Go up one level in this thread
On October 30, 2003 at 09:48:09, Robert Hyatt wrote: >On October 29, 2003 at 23:24:50, Keith Evans wrote: > >>On October 29, 2003 at 13:07:02, Robert Hyatt wrote: >> >>>On October 29, 2003 at 12:29:29, Gian-Carlo Pascutto wrote: >>> >>>>On October 29, 2003 at 10:04:27, Robert Hyatt wrote: >>>> >>>>>On October 29, 2003 at 01:31:06, Gian-Carlo Pascutto wrote: >>>>> >>>>>>On October 28, 2003 at 23:07:54, Robert Hyatt wrote: >>>>>> >>>>>>>So the books are 100% different? >>>>>>> >>>>>>>I don't believe it. 10% I might buy, but based on Jeroen's comments, they >>>>>>>are obviously not completely mutually exclusive lines... >>>>>> >>>>>>Of course not, they all have the Sicilian obviously, just as I'm sure >>>>>>your book does. >>>>>> >>>>>>You can't have three good books and have them not have a single variation >>>>>>in common. But if you think the books are largely identical, you are completely >>>>>>mistaken. >>>>>> >>>>>>-- >>>>>>GCP >>>>> >>>>> >>>>>I'll say it again. If they have less than 90% common lines, I would be >>>>>amazed. Just think about why... >>>> >>>>Yes, as I already said, and you obviously didn't read, there are of >>>>course common mainlines. But the programs style is substantially different >>>>and so are the books. >>>> >>>>-- >>>>GCP >>> >>> >>>And I'll say it again also. If they have 90% in common they are not >>>_that_ different. Or can I change 10% of Crafty and enter two copies in >>>the next WMCCC? Or three copies? Or N copies? 10% is not much. >> >>Any idea how much a Noomen book and a Kure book would have in common? Is it >>anywhere near 90%? >> >>-K > > >I have no idea. But it would likely be serendipity. > >IE if I do a lot of study to decide to include a particular line, and you >do a lot of study to include the same line, that is ok. That happens in engine >development (hashing, evaluation ideas, etc.) But if I do a lot of study to >decide to include a particular line, and then give that to you, my effort is >now spread across two engines, through the book. that is the point... As far as I am concerned, the Kure and Noomen books cannot be very different in the sense that both try to follow solid lines with the latest refutations of some unclear lines and not to solid moves. But the bulk of the work is actually done on what the program that you are making the book for does when it leaves the book... This is from my very recent book making experience with Ruffian. I think that both Noomen and Kure do a lot of analysis using the programs they compile books for. Therefore one could talk about overlapping books, but one surely cannot say that the books of the same author, or different authors, can be or are dentical or too similar (irrespective of the overlapping). Djordje
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.