Author: Robert Hyatt
Date: 08:15:13 09/18/05
Go up one level in this thread
On September 18, 2005 at 01:19:17, Mridul Muralidharan wrote: >On September 18, 2005 at 00:51:17, Robert Hyatt wrote: > >>On September 18, 2005 at 00:36:15, Mridul Muralidharan wrote: >> >>>On September 17, 2005 at 23:53:05, Robert Hyatt wrote: >>> >>>>On September 17, 2005 at 22:33:58, Vincent Diepeveen wrote: >>>> >>>>>On September 17, 2005 at 17:02:09, ALI MIRAFZALI wrote: >>>>> >>>>>Take a time control of 40 in 2 with pocket fritz against crafty without nullmove >>>>>and give crafty factor 1000 in nps extra. >>>>> >>>>>Deep Blue didn't use nullmove either. >>>> >>>>What does null move have to do with it? >>>> >>>>In 1996 there wasn't a computer on the planet that could beat deep blue. This >>>>is almost 10 years later. >>>> >>>>What is the point of this discussion??? >>>> >>>>Just a very lame attempt at starting a flame war, about a statement I supposedly >>>>made? (a statement I did _not_ make by the way)... >>>> >>> >>>I dont know about the flamewar part (I thought Vincent did not start the thread) >>>, but the 'without null move part' might be referring to the fact of deep blue >>>not using it. >>>Actually a good test would be : >>> >>>1) 1000x nps advantage >>>2) No null move >>>3) Use full singular extension as 'explained' by them. >> >>This is a completely worthless experiment. Take out my null-move search. >>Attempt to graft their singular extensions onto my program. What about my >>evaluation? My search extensions? How could one possibly add and remove bits >>and pieces of Crafty, to make something into the approximate skill of deep blue? >> >>Next, why is this important? My 1000x statement had nothing to do with >>null-move vs no null-move... >> >>> >>>I suspect (3) _will_ kill your search and keep the searchdepths much below what >>>the pocket fritz will get :) >> >>I don't think so. I had their full SE implemented in Cray Blitz. Its cost was >>almost 2 plies. But then tactically it was reaching very deep stuff to offset >>that. I've never found a workable SE that impressed me as "this is really good" >>when it comes to Crafty... >> >>On the quad opteron, my search depth would then probably drop to 13 plies. But >>then I get a factor of 1000X faster. My branching factor would be closer to 6 >>with no null-move, which would ramp me up by 4 plies without null move, or 10 >>plies with normal null move and a branching factor of around 2.0... >> >> >> >>> >>>So , even though Deep blue might have been invicible from programs of that age's >>>standards - it will get royally kicked around by even weak amateur programs of >>>today (bugfree ones ofcourse) running on modern hardware ! >>> >>>Regards, >>>Mridul >> >>again, based on what? Null-move is not generally credited with making a program >>200 Elo stronger. I might one day run some decent-length games with null vs >>non-null to see what the actual rating difference would be for Crafty, then one >>might actually extrapolate what 1000x faster hardware would do by actually >>playing that time-odds match as well. Then we don't need to guess, speculate, >>or anything else... >> >>I'll remind you that a few years ago (I don't remember exactly when although you >>can find precise mention of the experiment here) I had the chance to play Crafty >>(I think on some quad box, which one I really don't remember) against Cray Blitz >>(less singular extensions, the version on the machine I had access to did not >>have that version) on a T932. Thing was searching about 7M nodes per second. >>It gave crafty a pretty good drubbing. And Cray Blitz on that machine could not >>touch deep thought, much less deep blue, based on actual OTB games against them >>at ACM events... > >I think you are misreading my statements - and I am _not_ continuing with what >was initially posted. ><rambling> >There is this usual perception that 'deep blue solved chess' - which is a load >of shit : and if nothing, this will disprove it. Of course that perception exists. And, unfortunately, since they no longer compete, it will never go away either... Not much we can do about it... >Yes , it might have been strong for programs of its time : thats all. >Compared to programs of today , it may get a sound thrashing. >Crafty was getting what - 15-17 ply at WCCC2005 ? And deep blue was getting like >12 or so ? What about chessmaster? 6-7 plies? Do you think it gets run off the board by everyone else today? Of course not. There are "plies" and there are "plies". Comparing plies from A vs plies from B is pointless... ></rambling> > >We cannot really approximate Deep blue since we dont know what was put inside it >- what eval , what extension parameters (other than the published ones) , how >bad was its branching factor , how bad was its move ordering , how it was doing >qsearch , pruning , how stupid was its speedup (not approximate extrapolated >values : actual time to depth reduction) etc - and in most of these , naw - all >of these , it will be very inferior to what most of us do today - simply because >7 years is a lot of time and much progress has been made. In 1995-6, I pointed out that "we" were about 10 years behind them in speed. I was perhaps a little optimistic. We are not quite there yet, and probably need another couple of years at least to hit anywhere near 100M-200M. Yes other progress has been made. But if you extrapolate the ELo improvement for a factor of 1000X, which is about their speed advantage in 1995, you will see that the progress from 1995 to 2005, factoring out hardware improvements, is a _long_ way from what a speed factor of 1000 will add to the program. With a horrible branching factor, 1000X could give 5+ plies. That's a lot to overcome. When their evaluation didn't have any serious holes (at least none any GM could ever point out as a major weakness) that left one strong box... > >So how best can we approximate it ? I strongly suspect if IBM will ever bring it >out of cold storage and allow us to practically test a match ;) Which is obviously not going to happen... > >The best we can do is : >Take a bugfree program like crafty , add SE to it (very very expensive) , remove >null move from it , remove egtb deep blue used EGTBs. They used them before almost everyone else except for Belle, in fact... >, essentally try to dumb it down to what might >have been available at deep blue's times (you might be having crafty versions >from that time right ?) No. Most old versions were lost years ago in a major disk crash on my machine at UAB. But even then, Cray Blitz used null-move search, for reference. Did not use R=2, but it used null-move. In the 1980's after Beal's paper... > and give it a 1000x speed advantage. >This is _not_ going to be a deep blue versus crafty match - but an approximation >vs crafty match. >Note : the singular extension shortcuts that Bruce (he did it right ?) posted >were an approximation and not close to being as expensive as what deep blue did >- and even Bruce's methods tend to be very expensive without nullmove pruning as >depths increase (it will not be a static 2 ply advantage). > >I am not sure how many takers will be there for this 'match' , but it will >definitely be an interesting contest. > >And , you are up really late :) > I always am, but remember that I am GMT-5, so it wasn't that late. :) Like 12am or so worst case... >Regards, >Mridul
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.