Re: [computer-go] Hall of fame for CGOS
Hideki, It was not my intention to disrespect you.I think the word you are looking for is condescending which is when you talk down to someone as a child. If it came across this way I'm sorry.I tried to make it easy to understand because it seems to me that English is not your first language.I see that instead I embarrassed you and I certainly did not mean to do that.My fault. The web page in question is NOT Hall of Fame, I regret the use of that word in my posts to the group and it doesn't fit what it does in the least. Some of my C programs use variable names like this too, they would confuse anyone reading my code but they are there due to some historical accident, their use has gradually changed over time or even right away, starting out as one thing then becoming another - or simply because the name has some kind of internalized association in my own mind. I am going to change even the name of the html file to reflect the more accurate meaning later today. The web page was mostly designed to track ALL the players on CGOS. After about 30 days of inactivity you lose your place on the normal page as it is intended to only show active players. I have even wanted to look at historical ratings for my own bot and they are not there and so this addresses that need. But I also thought it would be interesting to use the (more accurate) bayeselo program.I don't show programs that have only played a few games because their ratings are almost meaningless and they make the page really big (and constantly growing.) I got this idea because I was doing sql queries to get information and I realized that the rest of the participants do not have that luxury. I am upset because I put this up as a kind of gift and instead it has upset people. Have you heard the expression, no good deed goes unpunished? That is a little of how I feel about this exchange. - Don Hideki Kato wrote: Chris, I consider to post my previous mail or not. Can you imagine my mind? I know it's so disrespective but I, still, respect Don. Yes, I was not angry, rather, very sad. I've posted because to, 1) to stop this discussion, 2) Don's last post was too disirespective (at least I though so) as, he never consider my questions seriously, that is, his answers were, say, not truely ones, rather like humoring children. The first sentense of his last post, Don't worry Hideki was too disrespective, wasn't it? I've worried very seriously, really. I wanted to just postpone the use of the name HoF until we confirm the overall ratings are correct enough to use the name. Don, please remove my bots, ggmc-xxx, from your HoF if possible. I'd like to apologize to all. -Hideki Chris Fant: [EMAIL PROTECTED]: On Dec 15, 2007 1:49 AM, Hideki Kato [EMAIL PROTECTED] wrote: Has Hall of fame with incorrect ratings any sense? Rather, it may wrongly leads pepole, isn't it? No disrespect intended, but just to let you know... this is incomprehensible. You may wish to rephrase the questions. I won't discuss farther as Don seems never change his mind with any discussion but I'd like to point out that Don should have many prior things to do than builiding such meaningless HoF. This is disrespectful. If you don't like his HoF, ignore it. Also, it is quite rare for someone to change their mind on a list like this. Only hard facts have any chance of changing a mind. Many of Don's messages deal with facts about his bot and many others deal with opinions and theory. What do you expect to happen? ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ -- [EMAIL PROTECTED] (Kato) ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] Lisp time
Stefan Nobis wrote: http://shootout.alioth.debian.org/ To clarify: I don't really like these non-scientific benchmarks (in many cases I assume no one or only really few people (not including me) really understand what each micro-benchmark is really measuring). [Off Topic] What do you mean by non-scientific - are you thinking of specific methodological objections? - are you simply being dismissive? If you have specific methodological objections then you are welcome to post those on the discussion forum: http://shootout.alioth.debian.org/gp4/faq.php#aliothid If you are simply being dismissive then someone should ask - have you looked at that website this year? - where are the alternative scientific benchmarks? Whether or not only really few people ... really understand what each micro-benchmark is really measuring seems to depend entirely on what /really/ understand and /really/ measuring mean - which you don't explain. And if we just take that as given and somehow understood, it's still no more than a vague suggestion that this is somehow a bad thing - you don't explain what significant problems will result. We must make do with the imperfect evidence that we can find, not merely lament its deficiencies. Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Lisp time
--- Stefan Nobis [EMAIL PROTECTED] wrote: Isaac Gouy [EMAIL PROTECTED] writes: - are you thinking of specific methodological objections? Yes, quite some methodological objections. To be fair: I'm no expert in this field and I haven't looked on the site for quite some time. Thank you for being honest. But benchmarks are really hairy, it's really easy to measure much more hardware than compiler aspects, people are very different in proficiency (in one case I heard even about rejection of a faster version of a benchmark in one language -- but I don't remember if this was on alioth or another language comparision compilation of benchmarks). Now you're spreading gossip :-) If there aren't really tight rules and quite some top experts in benchmarking involved, I wouldn't trust any single number... but that's just me and YMMV. I don't think Darren Cook was suggesting that you trust any single number - he linked to pages that compare 18 different measurements. We must make do with the imperfect evidence that we can find, not merely lament its deficiencies. Yes, that's true. But language shootout are much overrated und I doubt that there are any deep insights in there. So yes, I think I'm dismissive. Again, you seem happy to say they are overrated and dismiss them without actually having looked - that is not scientific! I'm sorry but what you've said about the benchmarks game seems to be no more than guesses and assumptions. At least Darren Cook looked. I've said enough. Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Lisp time
http://shootout.alioth.debian.org/ To clarify: I don't really like these non-scientific benchmarks (in many cases I assume no one or only really few people (not including me) really understand what each micro-benchmark is really measuring). I posted the link to the Shoot-out site as it seemed like a fair attempt at being scientific and objective; the OS is constant, they seem to offer benchmarks on 4 different hardware platforms, and the code being used is available for viewing and (as far as I can tell) anyone can post a better version if they think they can do it better. It'd be nice if someone added an alpha-beta tree searcher algorithm, and a zobrist hashing algorithm; then we'd have a much better idea of how language speed affects typical game programming tasks. Any students here with time for that? Darren ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] RE: several million nodes
When MF evaluates a position it does local tactical search to see if blocks of stones can be captured. It does this for every block with 3 or fewer liberties, and for points at the diagonals of eyes, and to see if connections are solid by trying to cut and doing a search to see if the cutting stone can be captured. It often does more than one local search for the same block, varying the side to move first and the ko status. This local search is why the evaluation function is so slow. MF evaluates about 40 to 200 full board positions per second during the global alpha-beta search. The local search examines over 100K positions per second, so a 30 second search examines 3 or 4 million positions in local search. Here is search output for a typical early endgame position (move 175). This is on one processor of a 2.3 GHz Core Duo. First it spends 0.7 seconds in the life/death search, evaluating 101 positions. Then it does 5 iterations of alpha-beta search, on 20 candidate moves, using about 34 seconds. It does a total of 5548 full board evaluations (164 per second). The local tactical search examines 3845717 positions, 24% for eyes, 10% for connections, 19% for blocks. Search 24-48 seconds (1 max evals), 0.7 secs in life reading level 5, life reading 101 life() New best 1870 (strat 1090)m9 m9 New best 1943 (strat 988)j9 j9 Iteration 1 complete 20 moves in 1.11 secs. Total evals 154, search life() 53, Final value is 1943 for j9 : j9 New best 1818 (strat 988)j9 j9 j8 o11 New best 1885 (strat 1125)m9 m9 l6 Iteration 2 complete 20 moves in 2.57 secs. Total evals 412, search life() 311, Final value is 1885 for m9 : m9 l6 New best 1820 (strat 1125)m9 m9 l6 e13 f13 Iteration 3 complete 20 moves in 5.10 secs. Total evals 879, search life() 778, Final value is 1820 for m9 : m9 l6 e13 f13 New best 1820 (strat 1125)m9 m9 l6 e13 f13 Iteration 4 complete 20 moves in 11.14 secs. Total evals 1932, search life() 1831, Final value is 1820 for m9 : m9 l6 e13 f13 New best 1870 (strat 1125)m9 m9 l6 e13 f13 k6 Iteration 5 complete 20 moves in 33.74 secs. Total evals 5548, search life() 5447, Final value is 1870 for m9 : m9 l6 e13 f13 k6 Pass val -5.6, best val 37.4 33.7 secs. life() 5548 (164/s), searchevals 1613, Nodes 2056 ( 61/s), full gen 319, Q gen 1483, Moves 3246( 96/s), tree nodes 207/207, list 10211/10211, tac nodes 3845717 (113971/s), evalopj 107868 ( 3%), life 3171808 ( 82%) [ fixgralive 3508037 ( 91%) miai 817026 ( 21%) tvpot 603326 ( 16%), group 742975 ( 19%), conn 376456 ( 10%), eye 922580 ( 24%) ] David -Original Message- From: Peter Christopher [mailto:[EMAIL PROTECTED] Sent: Saturday, December 15, 2007 3:18 PM To: [EMAIL PROTECTED] Subject: several million nodes Hi David, Thanks for you generosity in sharing your experience and perspective on the computer-go list. I noticed your statement that MFG does several million nodes of search. I assume this is per move. I see what you mean that this would be uncachable. It seems really hard for me to believe, is that really correct? Maybe I don't understand what is meant by a node - is it the creation of a representation of a global or local situation with some form of valuation metric applied? Do several million nodes really get some valuation metric applied to all of them? Sorry to send off-list; my email address almost always gets bounced from the list for some reason (but I can read the response there). Peter ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Lisp time
the OS is constant, they seem to offer benchmarks on 4 different hardware platforms Sorry for the noise. I mis-read the top page: they use two platforms, Gentoo on Pentium 4, and Debian on AMD Sempron. (They are very roughly the same, but java6-server stands out as being considerably quicker on Gentoo/Pentium4). Darren ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Hall of fame for CGOS
For what it's worth, Hall of Fame can mean anything from the Baseball Hall of Fame (serious business) to the Mullet Hall of Fame (a total joke). To me this looks like a pretty clear misunderstanding. Don and Hideki have both contributed usefully to the mailing list, and it would be too bad if this incident spoiled that from either end. I think most people who even know what CGOS is, probably know the context in which it is used, so they're not going to take poor results too seriously. Nobody would deny that applying static ratings to a program that changes may not be representative, but if that is an issue, then people can create new IDs for their 100% debugged bots so that static ratings will give them all the credit they deserve. The new bot would not get established immediately, but I doubt there are that many people outside this mailing list who are even aware of the CGOS all-time 9x9 rating list yet. As for any lower-rated versions of the same bot, I think that as long as there exists a highly rated version of a given program that has an established rating, people will have enough sense to ignore lower-rated versions that may represent experiments, buggy prototypes, and the like. ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Re: Lisp time
On Sat, Dec 15, 2007 at 03:00:02AM +1800, Nick Apperson wrote: In theory, and ONLY in theory, assembly is the fastest programming language. I do agree with most of what you said, but I have to squeeze in a comment that assembly is not necessarily the most optimized way to write code, when it really comes to it. If you *really* have to go for it, you need to be aware of the binary expression of every instruction... I have (once!) - many many years ago - optimized some assembly code where I reused the same bytes as instructions, jump offsets, and data, carefully placing the code on the right address so that this trick would work. This might have been possible with a good assembler, but at that time I did no have one for that CPU, so I was working directly in hexadecimal. With modern processors, assemblers, and compilers, things are different. And today, nobody will take the time to optimize on that level - which is good enough. But in theory... - Heikki -- Heikki Levanto In Murphy We Turst heikki (at) lsd (dot) dk ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Hall of fame for CGOS
Thank you very much Don. I was very surprised as I couldn't expect such kindful message. Don Dailey: [EMAIL PROTECTED]: Hideki, It was not my intention to disrespect you.I think the word you are looking for is condescending which is when you talk down to someone as a child. If it came across this way I'm sorry.I tried to make it easy to understand because it seems to me that English is not your first language.I see that instead I embarrassed you and I certainly did not mean to do that.My fault. I see. I know it was my misunderstanding or too-much-thought now. The web page in question is NOT Hall of Fame, I regret the use of that word in my posts to the group and it doesn't fit what it does in the least. Some of my C programs use variable names like this too, they would confuse anyone reading my code but they are there due to some historical accident, their use has gradually changed over time or even right away, starting out as one thing then becoming another - or simply because the name has some kind of internalized association in my own mind. I am going to change even the name of the html file to reflect the more accurate meaning later today. Thank you very much, Don. As I wrote earlier, it, itself is an excellent idea and Xmas gift to all of us, I strongly believe. The web page was mostly designed to track ALL the players on CGOS. After about 30 days of inactivity you lose your place on the normal page as it is intended to only show active players. I have even wanted to look at historical ratings for my own bot and they are not there and so this addresses that need. But I also thought it would be interesting to use the (more accurate) bayeselo program.I don't show programs that have only played a few games because their ratings are almost meaningless and they make the page really big (and constantly growing.) I got this idea because I was doing sql queries to get information and I realized that the rest of the participants do not have that luxury. I am upset because I put this up as a kind of gift and instead it has upset people. Have you heard the expression, no good deed goes unpunished? That is a little of how I feel about this exchange. It's my first look at and online/offline dictionaries help little nor there seems no corresponding saying in Japan but I guess I punished you? :( - Don Eric Boesch: [EMAIL PROTECTED]: Thank you Eric for encouraging me. For what it's worth, Hall of Fame can mean anything from the Baseball Hall of Fame (serious business) to the Mullet Hall of Fame (a total joke). To me this looks like a pretty clear misunderstanding. Don and Hideki have both contributed usefully to the mailing list, and it would be too bad if this incident spoiled that from either end. I'd like to say don't worry :). I won't stop posting to this list. Nothing will change. I still respect Don. I'm very glad to hear I've contributed but it cannot be compared to Don's. BTW, HoF of cgos cannot be a joke, in any sense. I think most people who even know what CGOS is, probably know the context in which it is used, so they're not going to take poor results too seriously. Nobody would deny that applying static ratings to a program that changes may not be representative, but if that is an issue, then people can create new IDs for their 100% debugged bots so that static ratings will give them all the credit they deserve. The new bot would not get established immediately, but I doubt there are that many people outside this mailing list who are even aware of the CGOS all-time 9x9 rating list yet. As for any lower-rated versions of the same bot, I think that as long as there exists a highly rated version of a given program that has an established rating, people will have enough sense to ignore lower-rated versions that may represent experiments, buggy prototypes, and the like. I was afraid also that outsiders may evaluate the ranking at HoF too much by its name. Finally, I'd like to thank Don, Eric, Chris and other subscribers of this list and apologize to for my posting of such emotional and off-topic articles. -Hideki Hideki Kato wrote: Chris, I consider to post my previous mail or not. Can you imagine my mind? I know it's so disrespective but I, still, respect Don. Yes, I was not angry, rather, very sad. I've posted because to, 1) to stop this discussion, 2) Don's last post was too disirespective (at least I though so) as, he never consider my questions seriously, that is, his answers were, say, not truely ones, rather like humoring children. The first sentense of his last post, Don't worry Hideki was too disrespective, wasn't it? I've worried very seriously, really. I wanted to just postpone the use of the name HoF until we confirm the overall ratings are correct enough to use the name. Don, please remove my bots, ggmc-xxx, from your HoF if possible. I'd like to apologize to all. -Hideki
