Author: Uri Blass
Date: 15:19:45 12/28/03
Go up one level in this thread
On December 28, 2003 at 17:48:40, Vincent Diepeveen wrote: >On December 28, 2003 at 17:11:02, Uri Blass wrote: > >>I do not believe it. >[..} >>I do not believe it. >[..] >>I do not believe it. >[..] >>I believe just me, myself and i > >You will remain always the same person! > > >>Can you prove it in a match against Crafty on ICC? > >A match for money at icc would be fun yes. If we play for a lot of money i will >consider it. Use a tournament book and a well tested diep version. by now many >bugs are coming out of diep thanks to many testers so i feel very confident. I see Opinion of people is not important for you. The question if people believe that you can beat Crafty is not important for you. A lot of people will refuse to bet about this things as a principle so it does not prove that they believe to you. > >Look you could even get back your money by betting that diep will lose. Perhaps >Hyatt will accept it... ...how he pays you out is your problem. > >Your payout is easy. After i have won i will ask for my money. If you don't pay >you will go through the window at that 10th story we play at in Tel Aviv 2004... >...note that some already told me to do that anyway :) > >Note that crafty has censored DIEP, but this is a minor detail, which is a >choice bob takes or doesn't take. > >by the way, my hardware will be faster than crafty's hardware, i bet bob has not >missed that the first few tries i had at that machine, but there is even bigger >hardware available. Yesterday i tested 16 processor altix3000 for a while... >...didn't dare to take 64 processors but when it is about money i'm shameless >like Bob and will show up prepared. > >O yes in such a match i will of course not use the tournament book of diep. >Waste of careful prepared lines! I'll do it like Kasparov and show up with other >book and other lines than normally get played. To simply PROTECT the tournament >book. > >Note: Ever seen junior play at icc with their tournament book? > >I do not see them play there. > >Ever see Kure play with the tournament book against crafty at icc? > >Why not? > >The answer is clear: ICC is not important. > >At most you use it to play a few games there. Note that i nearly never improve >diep based upon icc games. Waste of my time... ...easier is matches you play at >home or which get played by testers. > >>> >>>Level played is usually 2 hours + 30 seconds increment a move and in other cases >>>40 in 2 + 1 all >>> >>>>I do not see Diep in the ssdf list. >>> >>>I need to pay them 2 machines to garantuee that i get on the list. >>>I refuse to pay 2 machines to garantuee that diep gets quick on the list. >> >>I do not know about a rule that you need to pay 2 machines. > >Of course not, i just posted it 10 times but you never read what i write, this >is trivial. > >You would see 100 postings from SSDF if it wasn't true of course. > >It was Karlsson who asked me for that. > >Here is my answer now to him about this: You are a surgeon, you can pay for your >own hardware! > >>If it is only to be in the list in few months and not in 1 year then it is still >>be better for you to have diep of today in the list so at least you can prove >>that it is better than the programs of today. > >Don't write nonsense here. You are not very realistic. The first spot counts at >the list, even when it is 0.0000000000000001 better than spot 2. The rest >doesn't count. The list in itself doesn't count anyway, it's too tricked by now >and that isn't SSDF's fault at all. > >Your opinion is not interesting here by the way, it is trivial that you do not >realize all this. That's why you are Uri and have time to react onto every >message at CCC. > >But from commercial viewpoint seen: you do not prepare for a full year in order >to then get delayed a few months after which new programs release which have >books trained to win against you. > >Then knowing the human interpretation error is the biggest and 100% of all >testers i know is just looking at the screen for the result, so they will read >the result from the display and phone that to you. > >I bet SSDF is doing the same. In fact that's what some SSDF representative >posted here one day. They just hope then later to get the games. > >Knowing the huge errors reading that display (aborted games to just ring a >bell!!) it is trivial that you need all games. > >At such huge number of games like 400 games for 1 program, if you take 200 games >where there is not a pgn from then that represents 1000 rating points error >margin. *and nothing else*. > >>>Further if you play under fritz8 it will eat 80% system time and your engine >>>getting 20%, additionally give commands like putting your hashtable to 1 MB. > >>I doubt if you are right but you do not need to play under fritz8. > >So you *doubt* that fritz after the first game in UCI gives the second game the >command to put your hashtable back to 1 MB? > >You *doubt* it? > >Where the *hell* were you when there were extensive threads at ICC here? > >Too busy replying on it i bet! > >GCP posted it by the way. Not Diepeveen. So you don't even have the excuse that >you do not read postings by Diepeveen. I read GCP posts but that problem is easy to solve by ignoring the hash commands of Fritz and setting the hash by some ini file. I also read that it does not happen in every game. > >The 80% system time is easy to check too. diep's nps when i played a few games >was just 20k at a fast K7. It can get 100k nps there, so it got 20% system time. >Easy right? I do not know. Maybe there is another explanation for the 20% maybe another application was running on the same time. > >>I think that you can play under arena. > >So you will get 20% system time then and Fritz 80% when running at 1 processor >machines. No reason to get 20%. Thorsten was using movei under arena and I believe it got 50% > >That's nice way to test. > >There is another 100 tricks you know... > >>> >>>So basically you are busy 6 months just avoiding to lose because of details they >>>have set favourable for them. >>> >>>SSDF is a paid list and even paying $10 i would not mind, but for a paid list >>>you expect that *all* games they play get posted, not 50% of the games. >>> >>>Further is the human interpretation in testing always the most important, i >>>remember a tester mine a few years ago who concluded diep got butchered by Rebel >>>at the time and only a few draws. >>> >>>However those 'draws' were wins for diep. When a rook up in endgame it arbitered >>>it still as a draw the rebel interface. >>> >>>So you see all companies have made endless tricks for SSDF and you need to pay a >>>shitload of money too and for what? >>> >>>The only thing that counts is world champs anyway. Arguably IPCCC tournament is >>>important too because of a serious level. >> >>I do not see why you think it is the only tournament that is important. > >What is more relevant, winning olympic silver for 60 meters sprint, or the >highschool sprint 60 meters where i got 2nd when i was 18 (and 20 kilo's >lighter... ...and when i had a dutch national title chess till 20... ...oh those >good days!)? Diep never won turnament that was as important as WBEC. I do not remember it winning a single tournament. > >What is more important: a world title or some small unknown tournament in the >local townhall? > >Why does junior just join 1 tournament a year and it is the world championship >to be precise? > >PLEASE ANSWER THAT QUESTION! > >Note that Leo and Thorsten are very nice guys. This is Amir's decision but there are more and more commercial who join WBEC. King and Ruffian are the new commercial programs but there are also Deep Sjeng,Gandalf,Rebel. Uri
This page took 0.01 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.