GPL discussion, sense and nonsense

[hyatt]

You speak of Crafty while I had Rybka, Stockfish and Ippo.* in mind. Apparently they seem to profit a lot more.

Not from null-move and LMR. I tested all of 'em, with and without. Although it is easy enough for anyone to run the tests. The two together are nowhere near +200 Elo...

Ed is right here I'm afraid. It's going to be 200+ elo for LMR/null move combined in SF >=1.7.
To save the effort he could just ask Marco .
In Ippo it's also close to 200, and I bet the same for Rybka .

I believe ed said "+200 +300 +300 for LMR alone." But I'll report some numbers for stockfish, fruit and Crafty. Can't deal with ip* as it is too unreliable. Get too many crashes on cluster testing.

[hyatt]

Quite simply, in that I know a number of competitors at the time that went to those events. Bruce Moreland was one, and we had talked about a "reduction scheme" during the 1996-1997 (both) ICCA WMCCC events. Bruce talked quite a bit with SMK and he didn't reveal a thing about "reductions" to Bruce, as one example. And in the above window, both Bruce and I were experimenting with this idea as previously mentioned, we just never reached a "happy point".

Your bias is unbelievable

Have you (yourself) ever spoken with a commercial? I am asking because I never met you on any tournament. You just were never present. I am speaking of 1986 (Munich) and on. Programmers talk on tournaments, the commercial ones included. From several commercial programmers I got: Alpha/Beta, killer heuristic, iterative search, aspiration search, Q-search. I returned as well. They are not much different than you, if you have a passion it's hard to keep your mouth shut especially when you are on an event with your kindred spirits. Just enter a topic and words starts to roll. You should have tried but were not there, so how can you judge so mean?

Ed

An academic is just a commercial unable to find a way to get the masses to pay him, settled for persuading the university bean counters to pay him. The former requires at a bare minimum the ability to fool enough people for enough of the time, the latter the ability to fool one person once and get tenure.

Perhaps one day you might actually do a little research into what "tenure" is all about. It certainly is not about fooling "one person". You would have to fool a couple of dozen on-campus people, some in your department, some not. Plus the majority of your peers at other universities that are working in your field since we require 5-6 _outside_ reviewers as well.

A commercial programmer only has to fool the uninformed public that might buy his product. By a little creative manipulation of tournament results, testing agencies, and the like. I believe the tenure process is _far_ more difficult.

What if there's a giant mutually reinforcing conspiracy of people trying to hype computer chess into academia as the future of artificial intelligence blah-blah and all getting spin off financial and career advantage as a result? It's all a question of who are the suckers - taxpayers?

I don't follow. The holy grail of academia is "funded research". That is, grants from organizations like NSF, DOD, DOE, DARPA, and other alphabet-soup agencies. Neither I, nor either university I have worked ever received one cent of funding for computer chess. It has always been one of those "unfunded" things that get done as time is available. In the early 70's, this was a lot about AI, but that cat was out of the bag by 1980 or so when everybody was back to Shannon-A. Therefore, I do not quite follow what you are trying to say. My _funded_ research has been in other areas, primarily parallel programming issues unrelated to chess at all...

[hyatt]

How do you know he did not discuss it with others at maybe an ICGA event?

This is just ridiculous.
He might have also discussed it with his wife, or brother or whoever, but that is completely irrelevant.
I would even go as far to claim that if Bob doesn't know for a certain "contribution", then real contribution does not even exist.
What does a chess community has out of private discussion of couple of ICGA participants (if that really happened and is not just another Harvey's manipulation) which nobody can actually confirm???
Nothing at all, nada, zero, zip, zilch is the answer.

In the "good old days" we always had a panel discussion at each ACM computer chess tournament, and at some (at least in North America) ICCA events as well. Those were specifically to discuss ideas and such. We often had paper sessions where various more in-depth presentations were made. And the ICCA Journal was _formed_ to provide a written forum for dissemination of computer chess papers. Add 'em all up (ICCA/ICGA papers) and then count the commercial authors. Donninger is the only commercial author that comes to mind. In the late 80's and thru the 90's we had r.g.c.c. Go look at old posts there and see what was discussing actual algorithms and ideas, and who just discussed general topics with no specific ideas of any kind revealed. For all I know, computer chess might have reached the same point we are at today, with nothing but commercial authors. But the key is "for all I know" because they were certainly not heavy contributors to the body of knowledge related to computer chess.

I strongly suspect computer chess would not exist as we know it if not for the early pioneers that did share ideas freely.

Let's re-phrase this shall we? In the pre-1980's when there were no available microprocessors the only people with access to computers were the various university nerds and socioopaths sharing basement space with the cockroaches and whirring discs nobody knew what to do with. They thought "Hey, wow, awesome, now there's meaning to my life of pizza, late nights with the cockroaches and no girls and maybe I can make some headway in the university departmental building game that has so far eluded me!" The goal was a commercial career at the university, nobody having thought what miniaturisation would bring, so publishing 'papers' and 'sharing' nonsense became all the rage. Then those bastard commercials with their microprocessors and low cost of entry came along, surprisingly enough occupying slightly more of the real world space where people and ideas compete for limited wealth and then anyone could do it (beard, body odour, stammering, never washing and total clamminess with the girls being the only requirement).

Is the CC topic about computer chess or computer comedy? Even though you don't realize it, we _did_ have microcomputers in computer chess tournaments in the "pre-1980's". In 1978 I used a z80 to run my electronic chess board that many have seen pictures of. Kathe Spracklen remarked "You have more computer power just sensing squares on your chess board than I have in my program" (I believe she was 6502-based at the time. The z80 was introduced in 1976... I know that in 1978 "Sargon" was playing chess in tournaments, including the ACM event. As always, your history lesson has been "tried and found wanting".

Quibbling again, surely not
Under 100 sterling GB pounds mass market homecomputers didn't exist until ZX80 and the colour version Sinclair Spectrum and there was no mass market at all for games until 1980-1, that market first getting going in the UK, hence the 'commercials' you love to hate (and incorrectly define) never got going until 1980 onwards. Before 1980 you either had to build your own computer or else pay way too much money to get one (way too much = no mass market, and certainly not for software).

Maybe over there. Not over here. We had students buying altair 8800's right and left. There was a 8800 club located near Jackson MS in 1976. We had an undergrad that did a small chess program (in asm) for the 8800 we bought. Sargon existed and played in 1978, I have no idea how much money they made. But they clearly were playing in ACM computer chess events. Even lowly Tandy (Radio Shack) had the trs-80 (trash-80) for sale in 1977, my brother bought one. Perhaps you poor sods in GB didn't have anything available, but we sure did.

And for the record, I'd suspect the _first_ commercial came from the Spracklens, which was not "from the UK"

[hyatt]

How do you know he did not discuss it with others at maybe an ICGA event?

This is just ridiculous.
He might have also discussed it with his wife, or brother or whoever, but that is completely irrelevant.
I would even go as far to claim that if Bob doesn't know for a certain "contribution", then real contribution does not even exist.
What does a chess community has out of private discussion of couple of ICGA participants (if that really happened and is not just another Harvey's manipulation) which nobody can actually confirm???
Nothing at all, nada, zero, zip, zilch is the answer.

In the "good old days" we always had a panel discussion at each ACM computer chess tournament, and at some (at least in North America) ICCA events as well. Those were specifically to discuss ideas and such. We often had paper sessions where various more in-depth presentations were made. And the ICCA Journal was _formed_ to provide a written forum for dissemination of computer chess papers. Add 'em all up (ICCA/ICGA papers) and then count the commercial authors. Donninger is the only commercial author that comes to mind. In the late 80's and thru the 90's we had r.g.c.c. Go look at old posts there and see what was discussing actual algorithms and ideas, and who just discussed general topics with no specific ideas of any kind revealed. For all I know, computer chess might have reached the same point we are at today, with nothing but commercial authors. But the key is "for all I know" because they were certainly not heavy contributors to the body of knowledge related to computer chess.

I strongly suspect computer chess would not exist as we know it if not for the early pioneers that did share ideas freely.

The usual totally distorted idea of 'what is a commercial?' Narrow definition appears to be anyone who wrote and published and sold a program.

Better definition would be anyone who made some money or career out of computer chess. You could even add to that the academics who would be commercial if they could find someone to publish their stuff, and you can certainly add all the academics anyway who made a department or a career out of computer chess.

Bob is commercial, Hsu is commercial (sold himself to IBM) etc.

Nope. My definition of "commercial" is a programmer that developed a chess program, sold it for profit, and thereby had a vested interest in learning all he/she could learn about others, while revealing nothing about their own program. I doubt you could say I fit that definition. I have _never_ sold a copy of a chess program, and I think you'd find it hard to convince anyone I had any secrets I have failed to reveal since my source is available for all to examine.

Hsu is an interesting case. Did not make money by selling chess programs, but did get paid explicitly to design and build a machine to beat Kasparov.

If you are talking about commercial vs amateur as in the ICGA, forgetaboutit. Their definitions are brain-dead and those distinctions should have been dropped 20 years ago as the old WMCCC and WCCC eventsstarted to fade away and were slowly combined.

[thorstenczub]

richard langs first machine was on Z80 A 4 mhz :

http://www.schach-computer.info/wiki/in ... Chess_2001

the machine came out 1983 for ~300 Euro.

in it was a variation of Cyrus Chess program.

I have it run in one of my dedicated chess computer tournaments.
its really a strong program for that time:

Motor                                          Punkte HaNoChExScRFMeMeFiKaChInKrRFEnCCPoAdChMaBoLeMGPr    S-B
01: Hanimex HCG1900 {Brettcomputer}                9,5/11 · 1   0 1 = 1       1     1 1 1     1     1       18,50
02: Novag Constellation {Brettcomputer}            6,0/7  0 ·     1 1 1             1 1             1       12,00
03: Chess Champion Mark V {Brettcomputer}          6,0/9      · 1 0   0   0 1         1   1   1     1        7,00
04: Excel Display {Brettcomputer}                  4,0/5  1   0 ·             1             1     1         10,50
05: SciSys Turbo 16K {Brettcomputer}               4,0/6  0 0 1   ·       1               1         1        8,00
06: RFT Chess Master Diamond {Brettcomputer}       3,0/6  = 0       · 1 0                 =         1        8,00
07: Mephisto Mirage {Brettcomputer}                3,0/6  0 0 1     0 ·                 1           1        7,00
08: Mephisto Supermini {Brettcomputer}             2,0/2            1   ·   1                                4,50
09: Fidelity Sensor Voice Champion {Brettcomputer} 1,5/3      1   0       ·     =                            6,50
10: Karpov Chess+ {Brettcomputer}                  1,5/5      0         0   · 0 1 =                          1,50
11: Chessman FX Elite {Brettcomputer}              1,0/3  0     0           1 ·                              1,50
12: Intelligent Chess {Brettcomputer}              1,0/3                  = 0   · =                          1,25
12: Krypton Pioneer {Brettcomputer}                1,0/2                    =   = ·                          1,25
14: RFT SC2 {Brettcomputer}                        1,0/3  0 0                       ·   1                    1,00
14: Enterprise S {Brettcomputer}                   1,0/4  0 0 0                       · 1                    1,00
16: CC SuperSystem III {Brettcomputer}             1,0/5  0           0             0 0 ·         1          0,00
17: Pocket Chess {Brettcomputer}                   0,5/3      0   0 =                     ·                  1,50
18: Advanced Star Chess {Brettcomputer}            0,0/1        0                           ·                0,00
18: Chess Challenger Voice {Brettcomputer}         0,0/2  0   0                               ·              0,00
18: Maestro Travel Chess {Brettcomputer}           0,0/0                                        ·            0,00
18: Boris Diplomat {Brettcomputer}                 0,0/2        0                       0         ·          0,00
18: Lexibook 425 XLight {Brettcomputer}            0,0/6  0 0 0   0 0 0                             ·        0,00
18: MGS Morphy {Brettcomputer}                     0,0/0                                              ·      0,00
18: Prodigy Destiny {Brettcomputer}                0,0/0                                                ·    0,00

47 Partien von 276 gespielt
Name des Turniers: U1500
Ort/ Land: ORION8, Deutschland
Spielstufe: Turnier 40/120 20/60

this hanimex machine in my tournament is similar to the CXG2001 and has similar hardware/software like it. only different company selling it in UK.

Sargon (2.5) programs were first sold via Applied Concept company,
and later in Fidelity company.

[Chris Whittington]

Quite simply, in that I know a number of competitors at the time that went to those events. Bruce Moreland was one, and we had talked about a "reduction scheme" during the 1996-1997 (both) ICCA WMCCC events. Bruce talked quite a bit with SMK and he didn't reveal a thing about "reductions" to Bruce, as one example. And in the above window, both Bruce and I were experimenting with this idea as previously mentioned, we just never reached a "happy point".

Your bias is unbelievable

Have you (yourself) ever spoken with a commercial? I am asking because I never met you on any tournament. You just were never present. I am speaking of 1986 (Munich) and on. Programmers talk on tournaments, the commercial ones included. From several commercial programmers I got: Alpha/Beta, killer heuristic, iterative search, aspiration search, Q-search. I returned as well. They are not much different than you, if you have a passion it's hard to keep your mouth shut especially when you are on an event with your kindred spirits. Just enter a topic and words starts to roll. You should have tried but were not there, so how can you judge so mean?

Ed

An academic is just a commercial unable to find a way to get the masses to pay him, settled for persuading the university bean counters to pay him. The former requires at a bare minimum the ability to fool enough people for enough of the time, the latter the ability to fool one person once and get tenure.

Perhaps one day you might actually do a little research into what "tenure" is all about. It certainly is not about fooling "one person". You would have to fool a couple of dozen on-campus people, some in your department, some not. Plus the majority of your peers at other universities that are working in your field since we require 5-6 _outside_ reviewers as well.

A commercial programmer only has to fool the uninformed public that might buy his product. By a little creative manipulation of tournament results, testing agencies, and the like. I believe the tenure process is _far_ more difficult.

What if there's a giant mutually reinforcing conspiracy of people trying to hype computer chess into academia as the future of artificial intelligence blah-blah and all getting spin off financial and career advantage as a result? It's all a question of who are the suckers - taxpayers?

I don't follow. The holy grail of academia is "funded research". That is, grants from organizations like NSF, DOD, DOE, DARPA, and other alphabet-soup agencies. Neither I, nor either university I have worked ever received one cent of funding for computer chess. It has always been one of those "unfunded" things that get done as time is available. In the early 70's, this was a lot about AI, but that cat was out of the bag by 1980 or so when everybody was back to Shannon-A. Therefore, I do not quite follow what you are trying to say. My _funded_ research has been in other areas, primarily parallel programming issues unrelated to chess at all...

Well, I would still say you are commercial in every sense. You sell yourself to the university and the university buys you. You get salary, tenure, pension, so what do they get for the deal? Instead of listening to you, ask the university itself ..... they point lots to your skills and knowledge and experience in computer chess and to Cray Blitz and to Crafty. You've capitalised yourself like any sensible commercial and that capital forms part of the reason the university give you salary, pension and tenure.

As a commercial operating in academia, you have different demands on you than commercials operating ion the real world. For example, publishing material and papers. Commercials making products do quite the opposite usually.

[hyatt]

Quite simply, in that I know a number of competitors at the time that went to those events. Bruce Moreland was one, and we had talked about a "reduction scheme" during the 1996-1997 (both) ICCA WMCCC events. Bruce talked quite a bit with SMK and he didn't reveal a thing about "reductions" to Bruce, as one example. And in the above window, both Bruce and I were experimenting with this idea as previously mentioned, we just never reached a "happy point".

Your bias is unbelievable

Have you (yourself) ever spoken with a commercial? I am asking because I never met you on any tournament. You just were never present. I am speaking of 1986 (Munich) and on. Programmers talk on tournaments, the commercial ones included. From several commercial programmers I got: Alpha/Beta, killer heuristic, iterative search, aspiration search, Q-search. I returned as well. They are not much different than you, if you have a passion it's hard to keep your mouth shut especially when you are on an event with your kindred spirits. Just enter a topic and words starts to roll. You should have tried but were not there, so how can you judge so mean?

Ed

An academic is just a commercial unable to find a way to get the masses to pay him, settled for persuading the university bean counters to pay him. The former requires at a bare minimum the ability to fool enough people for enough of the time, the latter the ability to fool one person once and get tenure.

Perhaps one day you might actually do a little research into what "tenure" is all about. It certainly is not about fooling "one person". You would have to fool a couple of dozen on-campus people, some in your department, some not. Plus the majority of your peers at other universities that are working in your field since we require 5-6 _outside_ reviewers as well.

A commercial programmer only has to fool the uninformed public that might buy his product. By a little creative manipulation of tournament results, testing agencies, and the like. I believe the tenure process is _far_ more difficult.

What if there's a giant mutually reinforcing conspiracy of people trying to hype computer chess into academia as the future of artificial intelligence blah-blah and all getting spin off financial and career advantage as a result? It's all a question of who are the suckers - taxpayers?

I don't follow. The holy grail of academia is "funded research". That is, grants from organizations like NSF, DOD, DOE, DARPA, and other alphabet-soup agencies. Neither I, nor either university I have worked ever received one cent of funding for computer chess. It has always been one of those "unfunded" things that get done as time is available. In the early 70's, this was a lot about AI, but that cat was out of the bag by 1980 or so when everybody was back to Shannon-A. Therefore, I do not quite follow what you are trying to say. My _funded_ research has been in other areas, primarily parallel programming issues unrelated to chess at all...

Well, I would still say you are commercial in every sense. You sell yourself to the university and the university buys you. You get salary, tenure, pension, so what do they get for the deal? Instead of listening to you, ask the university itself ..... they point lots to your skills and knowledge and experience in computer chess and to Cray Blitz and to Crafty. You've capitalised yourself like any sensible commercial and that capital forms part of the reason the university give you salary, pension and tenure.

As a commercial operating in academia, you have different demands on you than commercials operating ion the real world. For example, publishing material and papers. Commercials making products do quite the opposite usually.

Couple of points. First, what do they get? Someone that teaches courses, educates students, and the students flock to my classes rather than away from them, in spite of the work these classes require. They get research effort in various ways, from being a CO-PI on the NSF grant that bought our last cluster, to collaborations across the campus. They get publicity from computer chess for sure, but no $. In my department we have 11 other tenured faculty, none of which is involved in any sort of game theory activities, yet they are still here. Computer chess is far from "all" that I do, but I certainly continue to enjoy doing it. Even if it has to be back-burnered at times due to other things going on here.

I can think of plenty of "commercials" that publish. Intel, for example. AMD. Cray Research. Etc.

[hyatt]

Got some preliminary data, will have complete data tomorrow (for Crafty) if our A/C stays alive. I am running 3 versions of Crafty, one with null-move disabled, one with LMR disabled and one with both disabled. I already have test runs for the current 23.3 version which has them both on. 23.3 is more aggressive with LMR as I mentioned a few months back, so I'm interested to see if this increased aggressiveness makes a significant difference.

Code: Select all

   Crafty-23.3        2686    3    3 30000   64%  2574   22% 
   Crafty-23.3R01     2626    3    3 30000   57%  2574   23% 
   Crafty-23.3R02     2571    4    4 19072   50%  2574   23% 

23.3 is the normal (current) version. 23.3R01 has null-move disabled, LMR active. 23.3R02 has LMR disabled, null-move active. 23.3R02 is about 11,000 games away from finishing, and then R03 will run with both disabled.

So far, removing null-move costs 60 Elo. Removing LMR costs 115, if that rating holds up (still has a little room to move). That is a bit different from the last time I ran this when removing either null or LMR cost us 80, removing both cost us 120. Once I get this done, I will report on Stockfish 1.7 (normal) and with LMR disabled to see what it does to that program. Not going to waste the time to test null-move as well...

More as test progresses...

[hyatt]

More partial results before I go to bed. Again, 23.3 is current version with everything enabled. 23.3R01 disables null-move only, 23.3R02 disables LMR only, 23.3R03 disables both.

Code: Select all

Crafty-23.3          2697    3    3 30000   64%  2585   22% 
Crafty-23.3R01       2636    3    3 30000   57%  2585   23%
Crafty-23.3R02       2580    3    3 30000   49%  2585   23%
Crafty-23.3R03       2449    6    6  8926   33%  2584   20% 

Turning this around to see what each adds, base Elo is 2449, adding null-move takes it to +131 (R02 is null on, LMR off). Just adding LMR takes it to +187, and adding both is +248. The whole is less than the sum of the parts. I'll post the complete results tomorrow, and then run the stockfish with LMR on and off to see how it behaves...

[BB+]

Donninger is the only commercial author that comes to mind.

I'd have to search the computer-go archives (which seem to have disappeared), but he also commented (mid-2000s?) on how worthless (or maybe incentive-less) it was from the standpoint of a commercial guy to write such papers.