Author: Vincent Diepeveen
Date: 09:57:37 02/03/04
Go up one level in this thread
On February 03, 2004 at 12:27:04, Robert Hyatt wrote: >On February 03, 2004 at 05:34:59, Vincent Diepeveen wrote: > > >>> >>>Then why are you so interested in it? Didn't you say that after the WCCC >> >>They claim to have worlds largest chesscomputer, let them proof their point that >>they also managed to get to work some processors, because i will be amazed if >>their speedup is much above 1.0 :) > > I wouldn't be surprised or amazed at anything. I've seen distributed chess >programs work just fine. Sun Phoenix is but one. It is doable. Whether it is if you refer to aphid: a) sun phoenix was just 16 processors b) very clearly Jonathan stated it would not scale at algorithms which are not fullwidth and he expects it to not scale above 16 cpu's c) it was a fullwidth program d) it was using a fast latency network when compared to the nps latency like 50 microseconds or so, whereas chessbrain is facing way worse than 200 microsecond latency problems So again another proof of your poor state of analytical insight into parallel computer chess. The 2 things are *not* comparable. >doable on the scale they want to use is another issue, but I'm way old enough >and wise enough to not discount such a project just because I believe it is >hard. I'm sure that at some point in time, people scoffed at the idea of taking >a heart from a dead person and transplanting it into a live person with a >defective heart. Yet it is done where I work multiple times per day, every day. > > >> >>>_you_ were going to post all your logs and speedup data for everyone to see? >>>Have you posted it? And you ask/bug others to post theirs??? >> >>Everyone who emailed me, i emailed all the logfiles. in fact some runs i even >>posted onto CCC :) >> >>If you want to put my logfiles at your ftp site, no problems. I don't have a >>homepage currently, that's all. >> >>Note that during the games i also was kibitzing mainlines, i'm sure you saw that >>yourself too being a regular observer :) >> >>>> Important is to have >>>>that logfile. I'm not even asking that you make things like collecting the total >>>>number of nodes searched (without losing system time in diep i collect >>>>statistics in DIEP at a central point this gets logged to the logfile). >>>> >>>>If you can save the total number of nodes searched for a team, you sure can save >>>>for that single game you play against a GM the logfile with the search depth in >>>>it that the central point had. >>>> >>>>How selective it was or wasn't is not important now. Trivially you must be >>>>creative the first few plies or you won't be able to get all nodes to work. In >>>>diep i'm forced at 460 processors to sometimes already split before nullmove >>>>gets made, otherwise i don't get all nodes to work simply. This where it is well >>>>known by everybody that splitting before or during nullmove is horrible for your >>>>speedup. >>>> >>> >>>It isn't known by me. You can split _in_ a null-move search just fine. I do it >>>now, I did it in Cray Blitz. Everyone else I know of also does it with no >>>problems... >>>of course don't let small facts get in the way of big nonsense... >> >>Cray Blitz had just 16 processors at a shared memory machine with fast >>interconnects and very slow processors, just 100Mhz or so. > >No, 1ghz. Again, look at the T90. The C90 was 500mhz. Tye YMP was about >200mhz. Although that does not tell the true picture. Compare the T90 memory >bandwidth to any machine you want, then gasp. And it could keep that bandwidth >busy all day long because of "vector operations". > > > > >> >>That's beginners stuff to get to work compared to > 100 cpu's. >> >>Also your thing was fullwidth. Non-recursive nullmove R=1 will reduce the entire >>search at most 1 ply, that's not counting as a real nullmove search :) >> >>You were for 99.99999% fullwidth minus at most 1 ply :) >> >>Your search depths of course proof that. By the way where are the logfiles from >>these games? > > >Which games? CB? Long lost along with all old crafty versions, my Ph.D. >dissertation electronic copy, all the old log files for many ACM events, old >papers I had written, etc. This is old news. > >I do have a few printed logs, say from the game Cray Blitz vs Joe Sentef, the >first time a computer beat a master in a long game. Of course the log was >published in Chess Life and Review as well... > > > > >> >>Problems splitting really start getting major league above 64 cpu's. Thanks to >>nullmove the trees are very instable, fullwidth they aren't. A search that's >>fullwidth is not so easy to split at 500 processors, but with nullmove you get >>real nightmare problems because you continuesly must abort very remote >>processors and different subsets. A very tiny search tree can abort some huge >>search tree. Fullwidth this happens less. >> >>Additionally a distributed project has dying processor problems and different >>latencies to different nodes, which make getting a speedup a lot harder. > >No doubt it is harder. But "harder" != "impossible"... > > >> >>I'm sure Colin Frayn & co know more there about the problems at such numbers of >>cpu's than you'll ever know :) > >I'd doubt it. Since I was doing this stuff before many were born. My first >parallel chess program played in 1978 at the ACM event in Washington, DC, on a >Univac 1100-series with two processors. When did yours first play? > > >> >>Everyone reports at 2 cpu's already major speedup problems when splitting before >>nullmove. >> > >I split before, after, etc. I don't see "major speedup problems"... No >literature I have read supports such a claim either.. > > > >>So if you claim now that you have never had problems here i demand you to >>implement it in crafty and play with that next amateur tourney :) >> > > >I split everywhere. Inside a null-move search. Outside a null-move search. >Doesn't matter to me at all.
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.