IWB wrote:
But 99.x% of all engine users uses this 0%!
So your statements pertain only to commercial engines, and have no general relevance. That is all I wanted to point out.
When was the last tournament where a programmer took part which was held in "ponder off"? WC, ICC tourneys, Leiden, Paderborn ... whatever, as soon as a tourney is played with programmers they use "Ponder ON" if the engine can do that!
Playing touneys ponder off is a makeshift as tere where/are just single core CPUs and some guys wanted to make rating lists with a lot of games.
I repaet and stand for it: "A real tournament should be ponder on"!
Again your statement is heavily colored by you very limited focus on toy computers and the commercial programs written for them.
If a program written for a 1024-core supercomputer (i.e. not a cluster, but the real thing) would participate in such a tournament, and they would have to hire time on such a facility, they would almost certainly prefer to spend the available budget of CPU-seconds by running as many CPUs as they can afford when their own clock is running, rather than wasting CPU seconds on pondering. Better to run 512 CPUs during half the time, than 256 CPUs all the time.
From the viewpoint of efficient use of hardware, pondering is simply a bad idea.
Very interesting but a bit nerdy! If you go back and have a look how this posting started and anticipate what the original poster might be interested in I think your 512 CPU cluster comparision is a bit over the top. I think your point of view might be interesting for a bunch of programmers, but not for the majority of users.
You might be right out of your programmer point of view, but that was not the question I answered in the beginning nor is it of more "general" relevance than my humble "vast majority user opinion"!
The same trade-off applies to the one asking the original question. The only thing that is relevant for this is really if there would be useful work for the computer to do outside your own thinking time, or that it otherwise would be idle anyway. If you are on an OTB event, only running your own engine, having someone else pay for the electricity, you can occupy the computer with pondering in stead of being idle, no matter how inefficient a time usage that is. Because leaving it idle is always even more inefficient.
But if there is something useful to do in stead, in becomes a different matter. And when someone is doing engine-engine tournaments, there is always an opponent, and usually more than a single game to play.
So I think indeed that the original asker falls in the category of users that would benefit most from playing ponder off. Arguments like "You should not play on one computer, but buy a second one, because then you can leave each of them idle half the time, to better simulate a REAL tournament" would not be good advice for such a user. Why would he want to simulate a REAL tournament? Most likely he just wants to produce good Chess games or interesting engine-engine results, in as short a time as possible. Having him carry his computer to a distant location, or use two different computers and log them in at FICS, because that simulates better what participants to REAL tournaments have to do, does not help him in any way to achieve these purposes...
hgm wrote:Ponder on will not hurt on a dual. But engines do not benifit as much from pondering as they would from giving them twice longer TC with ponder off, and then playing two games simultaneously.
IWB wrote:
But 99.x% of all engine users uses this 0%!
So your statements pertain only to commercial engines, and have no general relevance. That is all I wanted to point out.
When was the last tournament where a programmer took part which was held in "ponder off"? WC, ICC tourneys, Leiden, Paderborn ... whatever, as soon as a tourney is played with programmers they use "Ponder ON" if the engine can do that!
Playing touneys ponder off is a makeshift as tere where/are just single core CPUs and some guys wanted to make rating lists with a lot of games.
I repaet and stand for it: "A real tournament should be ponder on"!
Again your statement is heavily colored by you very limited focus on toy computers and the commercial programs written for them.
If a program written for a 1024-core supercomputer (i.e. not a cluster, but the real thing) would participate in such a tournament, and they would have to hire time on such a facility, they would almost certainly prefer to spend the available budget of CPU-seconds by running as many CPUs as they can afford when their own clock is running, rather than wasting CPU seconds on pondering. Better to run 512 CPUs during half the time, than 256 CPUs all the time.
From the viewpoint of efficient use of hardware, pondering is simply a bad idea.
Very interesting but a bit nerdy! If you go back and have a look how this posting started and anticipate what the original poster might be interested in I think your 512 CPU cluster comparision is a bit over the top. I think your point of view might be interesting for a bunch of programmers, but not for the majority of users.
You might be right out of your programmer point of view, but that was not the question I answered in the beginning nor is it of more "general" relevance than my humble "vast majority user opinion"!
Bye
Ingo
The vast majority of commercial GUI/engine users don't even know
what ponder is and if they have ponder on.
