Author: Uri Blass
Date: 16:56:29 04/07/03
Go up one level in this thread
On April 07, 2003 at 18:13:13, Vincent Diepeveen wrote: >On April 07, 2003 at 17:50:35, Uri Blass wrote: > >>On April 07, 2003 at 17:22:00, Vincent Diepeveen wrote: >> >>>On April 07, 2003 at 17:08:50, Uri Blass wrote: >>> >>>>On April 07, 2003 at 15:57:27, Vincent Diepeveen wrote: >>>> >>>>>On April 07, 2003 at 10:15:22, Steve Maughan wrote: >>>>> >>>>>>Uri, >>>>>> >>>>>>I'll have a bash, >>>>>> >>>>>>>How much money (s)he can get from it in the following cases: >>>>>>> >>>>>>>1)The program is at similiar level to Crafty >>>>>> >>>>>>Not much. As a winboard engine (i.e. no GUI) then I'd expect you'd sell at most >>>>>>50 copies. The price is not that important as long as it is less than $20 per >>>>>>copy. Basically on 'collectors' would buy it. >>>>>> >>>>>>>2)The program is at similiar level to Ruffian >>>>>> >>>>>>This is probably the Gandalf 4 scenario. I'd expect you'd sell about 200 copies >>>>>>(in total) as long as the price is $15 - $30 >>>>>> >>>>>>>3)The program is at similiar level to Fritz8 >>>>>> >>>>>>You will not have the brand equity of Fritz but I expect you'd sell about 500 to >>>>>>1000 copies depending upon marketing. I doubt the price point can be any more >>>>>>than $25 - $35 >>>>> >>>>>but how are you going to pay for those 10 hours you need for each person that >>>>>buys your 'winboard' engine if it is without interface of its own? >>>>> >>>>>that's 10000 hours of helpdesk. no one can give helpdesk of 10k hours for just >>>>>$30k a year. >>>>> >>>>>>>4)The program is 100 elo better than Fritz8 >>>>> >>>>>>You'd need good marketing but I think you could sell 3000 copies at this >>>>>>strength - I recon you could also get $50 a copy. Of course if it was really >>>>>>this strong your best bet would be to get someone like ChessBase to sell it. >>>>>>Then I think you could sell 8000 copies at $45 i.e. quite serious money. >>>>> >>>>>isn't ruffian 100 points better than fritz8 when using same openingsbook? >>>>> >>>>>Please turn off learning in fritz (after each game delete book simply) and let >>>>>it play at random. do same thing with ruffian. take care they use same PGN book. >>>> >>>>No need for pgn >>>>You can play the nunn match under Fritz and use different positions than the >>>>nunn positions. >>>> >>>>The difference between Ruffian and Fritz8 will be smaller than the difference in >>>>the ssdf but I will be surprised if Ruffian discovered to be 100 elo better. >>>> >>>>I guess that Ruffian will be something like 50 elo weaker that means losing >>>>450-350. >>>> >>>>Maybe somebody can do it because I do not have the time to do a fair match at >>>>long time control between Ruffian and Fritz8.0.0.8 and I know your opinion about >>>>blitz. >>>> >>>>opening positions can be chosen as all the possible 400 openings of 2 plies. >>>>I hope that you do not claim that I chose these positions to help Fritz. >>>> >>>>If you do not like these positions I agree to different positions like random >>>>800 positions from the games that programs got after 20 plies in the am/com >>>>in WBEC. >>>>> >>>>>See who wins. my money on ruffian. >>>>> >>>>>ruffian is free engine. >>>>> >>>>>so that basically answers uri's question. >>>>> >>>>>A solution without interface never sells of course. >>>> >>>>The author of Ruffian never tried to sell it. >>> >>>No. >>> >>>I did do some serious matches ruffian versus others and it is not your -50 >>>points but +100 points, so you have no right to guess here wildly. >> >>I read that the author of Ruffian claimed that he believes that Ruffian does not >>earn more from time relative to other programs. >> >>I tend to believe more the author of Ruffian and not you. >>You already said in the past a lot of things that were proved to be wrong. >> >>The last thing that I remember is about a drawn position when Diep >>evaluated it as a win for black and yace on a slow hardware was right. >> >>Uri > >the only one who is continuesly disproving himself seems to be you though. I do not see what you are talking about. Ruffian is 100 elo better than Fritz8.0.0.8 when you use same openings? If there is an engine that is really 100 elo better than you will probably be able to prove superiority also at fast time control and if there is a free engine that is better than fritz8 at blitz then people will prove it. free Ruffian may be good enough to win one tournament but if you play hundreds of games with it then I do not believe that it can get rating better than fritz8.0.0.8(otherwise people who test it could prove it). I simply do not believe that you played hundreds of games with Ruffian to compare with Fritz8.0.0.8 Uri
This page took 0 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.