More low quality games or few better quality games?

[swami]

It seems to me that people are only interested in stats, and stats alone.I'd better make it a quick reliable game rather than bother with long time control quality games, to get stats.

I enjoy watching games as I'm able, so I prefer slower time controls. Quality of play is therefore important to me.

However, people have different preferences, and as you say, the relative ratings of engines don't differ markedly in most cases regardless of the time control used.

Quality is important to me too in other things

but with computer chess, I feel it is more of a quantity than a quality, but that's just my opinion. since majority are interested in stats/elo rather than games.

[Ovyron]

If you have a very fast hardware compared to the average user then you can use 1' + 1" but if you are just about level with the average, maybe you can use 2' + 2" and if you have slower hardware you can use 4' + 4".

I have a 1CPU 2.1Ghz Athlon. It's ancient but at least faster than a Pentium 4

So far, Grahams' 40 moves in 5 minutes repeating sounds like a good compromise. Thanks all for your comments.

[Graham Banks]

but with computer chess, I feel it is more of a quantity than a quality, but that's just my opinion. since majority are interested in stats/elo rather than games.

Possibly, but is there any evidence for your claim?

[swami]

but with computer chess, I feel it is more of a quantity than a quality, but that's just my opinion. since majority are interested in stats/elo rather than games.

Possibly, but is there any evidence for your claim?

evidence? simple: just wait for the new updated version of *any* engine and see what the talk of town or what the most common question would be

or take a forum, and see the number of discussions relating to engine strength and compare it to the analysis of a single played game posted here.

[Graham Banks]

but with computer chess, I feel it is more of a quantity than a quality, but that's just my opinion. since majority are interested in stats/elo rather than games.

Possibly, but is there any evidence for your claim?

evidence? simple: just wait for the new updated version of *any* engine and see what the talk of town or what the most common question would be

or take a forum, and see the number of discussions relating to engine strength and compare it to the analysis of a single played game posted here.

My question was with regards to quality rather than quantity.
I would suspect that there would be many more enthusiasts than you think who would prefer to look over some good quality games and see a rating based on that over a period of time, rather than seeing somebody produce a hundred or more games in a day for the sake of a quick rating.
However I could well be wrong. My point was that there's no evidence either way on which to base a statement.

[swami]

but with computer chess, I feel it is more of a quantity than a quality, but that's just my opinion. since majority are interested in stats/elo rather than games.

Possibly, but is there any evidence for your claim?

evidence? simple: just wait for the new updated version of *any* engine and see what the talk of town or what the most common question would be

or take a forum, and see the number of discussions relating to engine strength and compare it to the analysis of a single played game posted here.

My question was with regards to quality rather than quantity.
I would suspect that there would be many more enthusiasts than you think who would prefer to look over some good quality games and see a rating based on that over a period of time, rather than seeing somebody produce a hundred or more games in a day for the sake of a quick rating.
However I could well be wrong. My point was that there's no evidence either way on which to base a statement.

Ofcourse, Fair Enough. But with the noise level on the forum soon after any engine gets released, and their curiosity to know the ratings, I thought they are interested in stats rather than games, i know there are some who check games as well as stats.

[Graham Banks]

with the noise level on the forum soon after any engine gets released, and their curiosity to know the ratings, I thought they are interested in stats rather than games, i know there are some who check games as well as stats.

Yes, I agree that many are generally keen to get as much information as possible as soon as possible.
That's only natural of course. But I'm just not sure that it's at the expense of quality as opposed to quantity. It's a fine line probably between the two anyway at the highest level.

Is that your girlfriend in your avatar?

[Spock]

Ofcourse, Fair Enough. But with the noise level on the forum soon after any engine gets released, and their curiosity to know the ratings, I thought they are interested in stats rather than games, i know there are some who check games as well as stats.

People certainly want a rating for a new engine as fast as possible, that is where blitz is perfect. CEGT's medium 40/20 on an X2 4200+ is a very good compromise, reasonable quality games and they can be churned out quite quickly. Our 40/40 on X2 4600+ is really difficult for quick results, and indeed we are normally a little slow to get statistically valid results for new engines. But we get there in the end (usually...)

[Dann Corbit]

I've been thinking about getting serious about chess engine testing, but I still have to decide on the time control, since I don't have many CPU time for this, I have to make a compromise, and these are the two time controls that I'd want to use:

1' +3" Games: This is the fastest that I would go, and I would get a decent amount of games much faster.

1' +14" Games: The idea is to have a good increment time so the game's quality remains constant and it depends on the length of the game. For example, a 40 moves game would have the quality of a 10' + 0" game, but a 80 moves game would have the quality of a 20' + 0" game, and so on. A longer game would get a better quality.

On 1' +3" long games, the quality gets severely bad because the engines have to play too fast at the end and the result is basically random. But 1' +14" games take too long to finish and it would take me a long time to get a good amount of games.

What do you suggest?

When you run fast games, I think it adds a bit more randomness, and requires that you check the results more carefully (you will see lots of losses for time, for instance).

When the chess is slow, the games are more beautiful. It would be nice if, during a thousand hours of chess contests, besides and answer "A is stronger than B." we would also have a beautiful byproduct of nice to follow games.

Now, the big problem here is that slow games take a long time, and nobody has the patience for it. So the way that this sort of thing can work is to join a consortium like SSDF, CEGT, or CCRL and add your games to the pile.

Now, if you are against doing joint projects, then probably some sort of blitz will be necessary unless you don't mind contests that take a very long time to complete, and give results for chess engines that are obsolete by the time that the contest is done.

I think for side projects if you are going to do something by yourself, a themeatic tournament might be intersting. An oddball opening that looks interesting could yield something interesting.

But for the bottom line, I would say:
"Do exactly what _you_ want to do."

[swami]

with the noise level on the forum soon after any engine gets released, and their curiosity to know the ratings, I thought they are interested in stats rather than games, i know there are some who check games as well as stats.

Yes, I agree that many are generally keen to get as much information as possible as soon as possible.
That's only natural of course. But I'm just not sure that it's at the expense of quality as opposed to quantity. It's a fine line probably between the two anyway at the highest level.

Is that your girlfriend in your avatar?

Well, I just had that opinion based on everyday happenings in the forum, and what their talks usually were and frequency of it etc

I'm sure people who are interested in high level engines games, they would go for it, as for lower rated engines, only programmers would download them.

I have downloaded CCRL games , quick wins games, Games from Rybka/Fruit etc

and Yep, since school days