Komodo 4 on long time control

[jpqy]

I remember long time ago that people here where taking about quality games.We had then a system like a Intel 486 DX..to have nice games we played 2hours games for each side,like a normal human game.
Everything played faster on a computer was not good?! The games loose quality!?
Then suddenly we get the 486 DX-2 upgrade ..and yes my test position i used before who takes 1hour to find a Mat in 12..was finded in 30min.
What a Wow feeling was that..and say ok ,now we can reduce the time by half for playing games!
I continued with my position till i had a AMD 3400+ and find the solution with same program ofcourse in 1sec.
I think you begin to follow my meaning..i'am now running my system number 14..a i7 970 @4,1Ghz..if i play 5min. games then my old system will need days to even reach the same quality games from before..that everybody wanted!
So,in all my years testings it doesn't matter what time you use..the strongest engine will be the strongest in any time control.
If a engine can't win in 5min.,10min.even 30min. it's not a number 1 engine.
We know all which engines have taken the number one position these last 20years..time changed alot with a big jump from Rybka ,and long time we say,chess have reach his upper limit.Personal i don't believe that,we are still far away from it.
The fun comes again with Robbolito 0.09 ..a 1core engine running on my Q6600 @3,6Ghz 4cores..and Robbolito destroyed all my 4core engines,unbelieveble strong!
Then a big jump again with Houdini..i started many lists..my last is now +21.700games..and nothing till today can't reach Houdini ,even with 4cores it stays above other 6cores engines.
Critter come closed..so i really look forward what the next Critter will do!

As for Komodo..Komodo 3 is a nice engine,in my other list it stays just under Robbolito 0.09..these 1core engines are the bottom from my list.
Komodo mp will need alot to reach Houdini..and i read here that Komodo team even want to pass Houdini..that would be very amazing!

As you know that most people testing on faster time controls..why not make a special compile for the people who using these fast TC.
Maybe 5% plays long TC and a other 45% use it for analyses.
So two compiles: Komodo mp game play & Komodo mp analyse (just a idea)
Then you can say,ok use for fast TC Komodo game..and long/analyse Komodo analyse..very simple!
No discuss anymore,about TC control to short to long,and bla bla bla..

When Rybka was on top..it doesn't matter what TC you use..it was the strongest engine.
When comes a engine out who can be stronger..suddenly there are TC problems with the engine?? you can only check on long time control to know sure if this engine is stronger?? i call it excusses,and discusses for nothing!

Houdini is here..Houdini is today the strongest engine no matter what TC
for long TC we have to less games,because almost nobody run these games anymore! But our system is so many times faster then in the old days where we run 2hours games.
Utra fast games i find not so good,is my personal meaning,i check always what my system can handle..and i want my cpu use 100% during the games.And for today is 5min. the min. for my system! where i reach depth 18-20 in 2sec. after opening!

20 years ago i was happy to see depth 3-6 in my 2hour games who where from high quality then

Jean-Paul.

[Frank Quisinsky]

Hi Larry,

look here:
40 moves in 150 minutes, 40 moves in 150 minutes, rest of the moves in 60 minutes and each game can be replay ... over 700 games!

http://www.amateurschach.de/swcr/_swcr- ... 2011-2.htm

one game over 10 hours with ponder = on
An other tourney with so long time controls and so many games you will not find on fast hardware.

And what is the results:
I do in all the years so many different things around computerchess but never I played such a boring tourney. No game I can visiting live. To long the games needed.

So many remis games, so many boring games I never saw. So what, for which reason such tourneys should we play.

I think I should have fun if the games are still running. One SWCR games need 40 minutes. This is fun, I can visiting the games LIVE and the time is very good to try to find the combinations during the games. Perhaps 40 in 20 is good too but without resign mode a game is still running with 40 in 20 around 1 1/2 hours. Again, this is too long if you will playing more as 1.000 games.

In SWCR 8 games are still running to the same time. Via HDMI to my TV and I have a lot of channels more If I have interest to visiting games I search a good channel on my TV and have fun because all 4 Q9550 have a HDMI connection to my TV.

And to Komodo again:
Komodo is stronger in positional chess, yes. And this is to see in longer games, yes. The advantage is perhaps 30 ELO I think to the others. But you should never forget that Komodo have some tactical holes. The advantage the other strong engines have is perhaps here 30 ELO too. I think we should not forget the weaknesses of an engine if we compare it with others.

Houdini for an example is clearly better in tactics as Komodo. Komodo is a bit better in positional chess. The different at the moment is around 50-60 ELO. In my opinion no chances here if we compare it with longer time controls.

Example:
Same different between SWCR and IPON and IPON is playing with the half time as I in SWCR. Only 3-4 engines benefit from longer time controls. Junior is here the best example. I don't believe that Komodo is one of this typ of engines and the reasons are the tactical holes.

Best
Frank

[MM]

[
Don and Larry needs to meet one of their deadlines and release Komodo 4 for testing.

All they are waiting for is a version of Komodo 4 that can overtake Houdini in the rating lists. With all the trash talk from them about Houdini's problems. Why the delay?

Basically, our problem is that we have two simultaneous projects going on. One is to come out with an MP version of Komodo as of the date code was frozen for it, and the other is the continued improvement of the SP version, which will eventually migrate to the MP.

I can propose an experiment you or anyone else out there with some good hardware can do. Run a long match of Komodo 3 64-bit SSE against Houdini (1.5 or 2.0, take your pick) 32-bit at some fairly long time control, maybe something like 25 minutes plus 15 seconds increment (i.e. five times what IPON uses), preferably with Ponder off so you can run twice as many at once. The point is that our latest Komodo version is roughly improved over Komodo 3 by the same amount that Houdini 64-bit outrates Houdini 32-bit. So whatever the result of such a match might be (I haven't run it so I don't know), it should be a good predictor for how our latest version will do against Houdini 64 bit at whatever time control you use. I know that Houdini 32-bit kills Komodo 3 64 bit at bullet chess (1' per game or anything comparably fast), so my expectation of a close result at long time controls rests squarely on my belief that Komodo scales much better than Houdini. I hope someone runs this test and posts the results here. But please, we need a few hundred games, not 50 or so. Of course if several people each run 50, that works fine!

Running a match now. With 5/6 TC. Perfect 12 abk. both engines.
Houdini 1.5 w32. vs. Komodo 3 -32. result so far. 9-0-9.
Houdini is killing Komodo after 18 games. Komodo has not won one game yet.

Best,
Gerold.

I understood Komodo 3 64 bit SSE - Houdini 1.5a 32 bit.

Of course and also long time control.
You are free not to do it but if you respond to Larry who asked to do testing of Komodo3 64 bit SSE against Houdini 1.5a 32 bit at long time control then I do not see the point of giving blitz results with komodo 32 bits.

I was going to run 100 plus games but since no one wants them run at blitz i stopped at 30 games. Komodo 3 won one game out of 30 played. Looks like Komodo needs a little more work.

Best,
Gerold.

You missed the whole point. We know that Komodo 3 32 bit is relatively very weak, just look at the rating lists. It is the product of a few minutes' work, we gave it zero priority (that may change in the future). The point was to use Komodo 64 bit as running Houdini 32 bit handicaps it by about the amount we have improved Komodo by since Komodo 3. That will tell you roughly where we stand right now. We'll still lose at blitz I think, though narrowly, but at something like the suggested 25' + 15" we might well come out ahead in a long match. Anyway I'd like to know.

I can't run a rapid/long match Komodo 64 bit SSE - Houdini 1.5 32bit.
Can someone else run it?

Thank you

Best Regards

[gerold]

[
Don and Larry needs to meet one of their deadlines and release Komodo 4 for testing.

All they are waiting for is a version of Komodo 4 that can overtake Houdini in the rating lists. With all the trash talk from them about Houdini's problems. Why the delay?

Basically, our problem is that we have two simultaneous projects going on. One is to come out with an MP version of Komodo as of the date code was frozen for it, and the other is the continued improvement of the SP version, which will eventually migrate to the MP.

I can propose an experiment you or anyone else out there with some good hardware can do. Run a long match of Komodo 3 64-bit SSE against Houdini (1.5 or 2.0, take your pick) 32-bit at some fairly long time control, maybe something like 25 minutes plus 15 seconds increment (i.e. five times what IPON uses), preferably with Ponder off so you can run twice as many at once. The point is that our latest Komodo version is roughly improved over Komodo 3 by the same amount that Houdini 64-bit outrates Houdini 32-bit. So whatever the result of such a match might be (I haven't run it so I don't know), it should be a good predictor for how our latest version will do against Houdini 64 bit at whatever time control you use. I know that Houdini 32-bit kills Komodo 3 64 bit at bullet chess (1' per game or anything comparably fast), so my expectation of a close result at long time controls rests squarely on my belief that Komodo scales much better than Houdini. I hope someone runs this test and posts the results here. But please, we need a few hundred games, not 50 or so. Of course if several people each run 50, that works fine!

Running a match now. With 5/6 TC. Perfect 12 abk. both engines.
Houdini 1.5 w32. vs. Komodo 3 -32. result so far. 9-0-9.
Houdini is killing Komodo after 18 games. Komodo has not won one game yet.

Best,
Gerold.

I understood Komodo 3 64 bit SSE - Houdini 1.5a 32 bit.

Of course and also long time control.
You are free not to do it but if you respond to Larry who asked to do testing of Komodo3 64 bit SSE against Houdini 1.5a 32 bit at long time control then I do not see the point of giving blitz results with komodo 32 bits.

I was going to run 100 plus games but since no one wants them run at blitz i stopped at 30 games. Komodo 3 won one game out of 30 played. Looks like Komodo needs a little more work.

Best,
Gerold.

You missed the whole point. We know that Komodo 3 32 bit is relatively very weak, just look at the rating lists. It is the product of a few minutes' work, we gave it zero priority (that may change in the future). The point was to use Komodo 64 bit as running Houdini 32 bit handicaps it by about the amount we have improved Komodo by since Komodo 3. That will tell you roughly where we stand right now. We'll still lose at blitz I think, though narrowly, but at something like the suggested 25' + 15" we might well come out ahead in a long match. Anyway I'd like to know.

Ok. First time i fine someone don't want their engine tested with equal conditions for both engines to find out how to improve their engine. I have deleted the games and Komodo from my machine. No problem.

Best,
Gerold.

[Rebel]

[quote="tomgdrumsDon and Larry do indeed talk a lot of smack in the forum.

They seemed to have become obsessed with overtaking Houdini!! It is like watching the computer chess version of "Moby Dick"!!

I like that analogy! Of course, we have a strong commercial incentive to overtake Houdini, but you are right, it goes beyond that. My main gripe about Houdart is that he has refused to acknowledge the well-established fact that the early Houdini versions were just some Ippolit version with minor modifications. Apparently, he could admit this and still sell legally if the program he cloned didn't have any conditions listed on using it, but for some reason he won't admit the obvious. He deserves credit as someone who improved a strong program considerably, but he is not a program author in any meaningfull sense.
I don't have the same bad feelings about Ippo (or other Ippo-relatives) anymore. While Ippo clearly was modelled after Rybka, it was written as an independent program and is sufficiently different to be (probably) legally ok, even if it could not enter ICGA events. If for any reason we fail to pass Houdini soon, I'll still be happy to see Critter or Rybka or even Ivanhoe do so. At least Richard Vida wrote his own program, even if it was based on Ippo, and he has been reasonably forthcoming about it.

+1

[MM]

[quote="tomgdrumsDon and Larry do indeed talk a lot of smack in the forum.

They seemed to have become obsessed with overtaking Houdini!! It is like watching the computer chess version of "Moby Dick"!!

I like that analogy! Of course, we have a strong commercial incentive to overtake Houdini, but you are right, it goes beyond that. My main gripe about Houdart is that he has refused to acknowledge the well-established fact that the early Houdini versions were just some Ippolit version with minor modifications. Apparently, he could admit this and still sell legally if the program he cloned didn't have any conditions listed on using it, but for some reason he won't admit the obvious. He deserves credit as someone who improved a strong program considerably, but he is not a program author in any meaningfull sense.
I don't have the same bad feelings about Ippo (or other Ippo-relatives) anymore. While Ippo clearly was modelled after Rybka, it was written as an independent program and is sufficiently different to be (probably) legally ok, even if it could not enter ICGA events. If for any reason we fail to pass Houdini soon, I'll still be happy to see Critter or Rybka or even Ivanhoe do so. At least Richard Vida wrote his own program, even if it was based on Ippo, and he has been reasonably forthcoming about it.

+1

+2

[Rebel]

I like that analogy! Of course, we have a strong commercial incentive to overtake Houdini, but you are right, it goes beyond that. My main gripe about Houdart is that he has refused to acknowledge the well-established fact that the early Houdini versions were just some Ippolit version with minor modifications.

You are right of course but living in a post ICGA-Rybka world I can imagine he is careful. You won't get a heartwarming welcome if you admit you have taken the sources of someone else as a base. The opposite does not work also as proven. Typical -/- situation, Houdart whatever he does always loses. I think that's not fair and something must change.

[lkaufman]

I remember long time ago that people here where taking about quality .We had then a system like a Intel 486 DX..to have nice games we played 2hours games for each side,like a normal human game.
Everything played faster on a computer was not good?! The games loose quality!?
Then suddenly we get the 486 DX-2 upgrade ..and yes my test position i used before who takes 1hour to find a Mat in 12..was finded in 30min.
What a Wow feeling was that..and say ok ,now we can reduce the time by half for playing games!
I continued with my position till i had a AMD 3400+ and find the solution with same program ofcourse in 1sec.
I think you begin to follow my meaning..i'am now running my system number 14..a i7 970 @4,1Ghz..if i play 5min. games then my old system will need days to even reach the same quality games from before..that everybody wanted!
So,in all my years testings it doesn't matter what time you use..the strongest engine will be the strongest in any time control.
If a engine can't win in 5min.,10min.even 30min. it's not a number 1 engine.
We know all which engines have taken the number one position these last 20years..time changed alot with a big jump from Rybka ,and long time we say,chess have reach his upper limit.Personal i don't believe that,we are still far away from it.
The fun comes again with Robbolito 0.09 ..a 1core engine running on my Q6600 @3,6Ghz 4cores..and Robbolito destroyed all my 4core engines,unbelieveble strong!
Then a big jump again with Houdini..i started many lists..my last is now +21.700games..and nothing till today can't reach Houdini ,even with 4cores it stays above other 6cores engines.
Critter come closed..so i really look forward what the next Critter will do!

As for Komodo..Komodo 3 is a nice engine,in my other list it stays just under Robbolito 0.09..these 1core engines are the bottom from my list.
Komodo mp will need alot to reach Houdini..and i read here that Komodo team even want to pass Houdini..that would be very amazing!

As you know that most people testing on faster time controls..why not make a special compile for the people who using these fast TC.
Maybe 5% plays long TC and a other 45% use it for analyses.
So two compiles: Komodo mp game play & Komodo mp analyse (just a idea)
Then you can say,ok use for fast TC Komodo game..and long/analyse Komodo analyse..very simple!
No discuss anymore,about TC control to short to long,and bla bla bla..

When Rybka was on top..it doesn't matter what TC you use..it was the strongest engine.
When comes a engine out who can be stronger..suddenly there are TC problems with the engine?? you can only check on long time control to know sure if this engine is stronger?? i call it excusses,and discusses for nothing!

Houdini is here..Houdini is today the strongest engine no matter what TC
for long TC we have to less games,because almost nobody run these games anymore! But our system is so many times faster then in the old days where we run 2hours games.
Utra fast games i find not so good,is my personal meaning,i check always what my system can handle..and i want my cpu use 100% during the games.And for today is 5min. the min. for my system! where i reach depth 18-20 in 2sec. after opening!

20 years ago i was happy to see depth 3-6 in my 2hour games who where from high quality then

Jean-Paul.

At the moment we don't know how to make a version of Komodo that is significantly stronger than normal at blitz although weaker at slow time controls. If we do discover how to do this, we might very well adopt your idea of separate versions for blitz and slow play. As for the merits of slow vs. blitz testing, it comes down to the former being relevant for those who use the engine for analysis, while blitz testing is relevant for those who normally just give the engine a few seconds per move when reviewing games.

[IWB]

Hi Don,

... That's why it's important to us that Komodo or for that matter Critter or Stockfish or some other legitimate program ends up on the rating lists on top....

Just as a side note I am less confident that this is such a good thing! Just keep in mind what happened to Houdini 2.0 with Strelka 5.0 just weeks after the release ... It is a sad thought, but I am less confident that being No.1 is that good if another unique code might be exposed some time later ... The consequences are spoiling the whole hobby to me!

Bye
Ingo

[Werewolf]

Larry,
What is your & Don's release cycle plans for Komodo?

Do you plan to follow the once-per-year model or the Rybka 2 model of a release every few months?

Personally, I think for serious chess players the Rybka 2 model was the best of all.