Author: Robert Hyatt
Date: 19:12:23 10/29/99
Go up one level in this thread
On October 29, 1999 at 18:33:14, Amir Ban wrote: >On October 29, 1999 at 16:59:20, Robert Hyatt wrote: > > >>9. Singular extensions (I don't know what you do, but genius, wchess, and >>others have/do use these) came from deep thought developers Hsu and Campbell. >> > >I talked with Kittinger in 1995. He didn't have SE then, and was skeptic if the >concept is right. > He was on ICC regularly about 3 years ago and used to talk with Bruce and myself there all the time. He was using it. I don't know whether he kept it, but he asked a lot of questions about it as he knew I had done it in CB from our discussions at the last two ACM events. He started with PV singular extensions, and seemed to be happy. I think he did it because Lang claimed that genius 2 used them too. > >>the list goes on and on... >> >> >>>There is nothing special I have seen in the Crafty source code. Just the >>>basic things, well tuned and documented, but nothing special. >> >>No "special parallel search?" Non-trivial to do. Non-trivial to get right. >>etc. no unusual evaluation terms? Seems that _everybody_ suddenly decided >>that it was 'right' to probe in the search, not just at the root. I've been >>doing it about as long as Bruce (he wrote his own tablebase code, while I used >>the Edwards stuff that was public. Edwards was doing it before I was, >>obviously, as he wrote the probe code for Crafty. Whether he probed exactly as >>I do today is another question. But I notice that more and more commercial >>programs are doing that. Where'd it come from? >> > >It's obvious, but you need tablebases that are fast enough. That's the real >trick. they have been around for a long time. I've been using them since 1995 on PC machines, as has Bruce. Steven Edwards was using them way before that on PC-class machines. The point is, everyone _assumed_ that you couldn't do the I/O in the search, because it just feels wrong. No one thought a lot about the idea of being more careful, probing only after captures that take you down to the right number of pieces, etc. Steven Edwards is the author of that idea, btw. not me. It just so happens that my source code, and my results probing in the search (plus Ferret's results too, no doubt) convinced everyone that the impossible was doable. > >> >> >>> >>>And what do you expect people to do with your source code anyway then >>>to have a look at it? Isn't that the purpose? >> >>Sure... but you guys don't get it. Intel spends a year of secrecy to >>develop a new processor. They spring it on the world, _and_ they publish >>papers describing _exactly_ what they did. IE they get the lead-time to take >>advantage, but then the publish details that takes the entire industry forward. >> > >It so happens I'm quite familiar with Intel, and there's no truth in what you >say. Intel will publish with a product anything that is needed to make you >comfortable using it and buying it. That's quite a lot, usually, but they won't >tell you anything beyond that. Do you ever read IEEE Micro? Or any of the other journals on VLSI and the like? Check them out. I have. IE I have a video tape released by Digital right after the first alpha was released. It had the original design team do presentations on each internal feature of the alpha and why it was done the way it was. I have an entire library of such tapes that I use in my architecture course. IBM explaining their first internally-designed microprocessor. HP submitted a tape on the HP PA architecture. They are far more open than you might realize, you just have to look. When intel released my quad xeon box, I downloaded the PDF that gives the _complete_ specifications for the box. Connectors, clock timings, chips supported, power supply specifications, hot-swap specifications... _everything_ in a publicly downloadable (big) file. > >One of the things Intel currently does is a strategic effort to reinvent PC >architectire from an open standard into something Intel-proprietary. The >so-called "firmware hub", e.g., will replace the old BIOS, and the LPC bus >replaces the ISA bus. The specifications are secret or restricted to Intel >partners. If Intel succeeds in this, competitors like AMD will have a real >problem. > Different issue. I don't care if there is competition or not. I only want to know what they are doing, and that they provide in minute detail after a chip hits the market. Because any good engineering house can figure it out anyway with a microscope and logic analyzer... >Microsoft is doing something similar in the past few years. They made DOS into >the most successful OS ever by making it totally open and attracting third-party >developers, who really made DOS successful. Microsoft now thinks that those >third-party developers are a nuisance and they are closing many specifications. >For example, NTFS (the NT file-system) is not documented. It is better documented than you think. There are linux drivers around to mount ntfs file systems under development. I don't know whether it was reverse-engineered or microsoft has technical reports to explain ntfs, but someone has access. > >Amir And I am not saying that playing the 'profit-first' game is horrible. Only that it is not the way things were when I 'grew up'. And I definitely remember that things were 'fun' when I was growing up.
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.