Author: Eran
Date: 23:49:43 09/08/00
Go up one level in this thread
On September 08, 2000 at 15:58:04, Sune Larsson wrote: >On September 08, 2000 at 15:46:48, Eran wrote: > >>On September 08, 2000 at 15:11:56, Sune Larsson wrote: >> >>>On September 08, 2000 at 14:36:38, Eran wrote: >>> >>>>On September 08, 2000 at 12:59:49, Sune Larsson wrote: >>>> >>>>>On September 08, 2000 at 00:06:18, Michel Langeveld wrote: >>>>> >>>>>>* Because it doesn't catch all of them a large database can be a great mess in >>>>>>the end. >>>>>> >>>>>>* It's also strange that add to database without doubles find more doubles than >>>>>>add to the database (Without double check) and then kill doubles. >>>>>> >>>>>>* Kill doubles kills also a lot of games which aren't doubles at all. Something >>>>>>total different players where killed >>>>>> >>>>>>* It's also strange that kill doubles kills games where both sides didn't move. >>>>>>This results that a lot of valuable information will be deleted from the >>>>>>database. Complete crosstables will be ruined then. >>>>>> >>>>>>* I've seen A LOT of database corruptions, and I lost a lot of times my database >>>>>>due to errors with CB7. >>>>>> >>>>>>Did you use CB7 a lot of times?? For example did you also receive messagebox's >>>>>>after your 500th or 1000th startup of the software? I did receive :-) I use CB7 >>>>>>almost every day. (sometimes 5 times a day or so). If you don't know what's the >>>>>>matter with kill doubles then I don't expect you are a heavy user. >>>>>> >>>>>>However I really love database in comparison with CB6. Because It's MUCH faster. >>>>>> >>>>>>Kind regards, >>>>>> >>>>>>Michel Langeveld >>>>> >>>>> Agree. Futhermore I have asked this question to ChessBase and the answer >>>>> is no. Meaning no improvements in "Killing Doubles" whatsoever. Meaning >>>>> the same , very few parameters, available. Meaning fuzzy duplicates cannot >>>>> be spotted. Meaning uphere we're going for C Assistent! Meaning conclusions >>>>> from consumers/users in my neighbourhood. >>>>> >>>>> Thanks >>>>>> >>>>>>On September 07, 2000 at 20:20:07, Marc Plum wrote: >>>>>> >>>>>>>I'm very satisfied with CB 7, and don't think that I have any reason to upgrade. >>>>>>> For that matter, version 6.01 worked very well for me, although I hated that >>>>>>>dongle. >>>>>>> >>>>>>>Just out of curiosity though, what are the problems you've had with the "kill >>>>>>>doubles" function? I hadn't noticed serious problems, admittedly it may not >>>>>>>catch all of them, but it seems to do pretty well. >>>>>>> >>>>>>>Marc >>>>>>> >>>>>>>On September 07, 2000 at 17:41:09, Anatoli wrote: >>>>>>> >>>>>>>>Hello everybody ! >>>>>>>> >>>>>>>>We are waiting for the new CB8. There will be some make up changes, but I didn't >>>>>>>>read anywhere about any improvments in Kill Doubles. Did I miss this information >>>>>>>>? From my point of view this function in CB7.01 works not very good. But from my >>>>>>>>point of view it should be one of the most important possibilities in CB. Can >>>>>>>>anybody here from ChessBase answer this question ? I have CB7 and now I have to >>>>>>>>pay around US$90 to upgrade to CB8. I am not interested in new pictures of the >>>>>>>>players, in their ratings, in new Fritz interface, or in the games' statistic. >>>>>>>>I read all previews of CB8 and I am not impressed. I always bought ChessBase >>>>>>>>products, but now for the first time I have to think. What about you, my >>>>>>>>friends, are you happy with everything you read about it ? >>>>>>>> >>>>>>>>Best wishes >>>>>>>>Tolya >>>> >>>>1) Make one or two backups of your database before you let CB7 kill doubles. >>>> >>>>2) Maybe some time later, after the release of CB8 in October, some downloadable >>>>new patches from ChessBase would fix "Killing Doubles." I hope so. >>>> >>>>3) ChessBase 7 acts like Windows 3.x, while ChessBase 8 acts like Windows 9x. >>>>That means, in CB7 you have to change from windows to another windows so often. >>>>in CB8, it acts like explorer meaning you would go from windows to another >>>>windows much less often, because some of the windows are already tiled for >>>>conveniency. >>>> >>>>4) The appearance of the CB8 is more attrative and pleasing than Assistant. >>>> >>>>5) I still prefer CB over Assistant, >>> >>> As far as killing fuzzy duplicates??!! Then please post your tricks! >>> Beacause I'm really interested in your findings... BTW are you in >>> someway connected to ChessBase? Beta-tester or... >>> >>> Thanks >>> >>> >>> >>> and CB is the number one choice among top >>>>chess players in the world. >>>> >>>>Eran >> >> >>Nope. I am just an enduser of CB7. You can try Steve Lopez's T-Notes about >>"Killing Doubles" at www.chessbase.com. > > I know that. Killing *fuzzy* duplicates it was... Any ideas? > > Thanks > > >> >>Eran I never tried killing fuzzy duplicates myself before. I think that the programmers at ChessBase need to work hard more time to create better killing programs that kill fuzzy duplicates. Those programs need to be much more complicated in order to be able to kill them effectively. However, as customers of ChessBase, we can ask ChessBase programmers to create much better killing programs for ChessBase products. I believe they will answer positively, but it may take time and patience. Analogically, a good killing program is like a good program that parses fuzzy PGN files effectively. Due to their fuzziness, creating good programs like that is extremely difficult and will never be perfect ones. That's part of realiy. Besides good killing programs, we need good discipline of typing as fewer errors as possible in our databases's headers. Anyway, you can experiment with a created small database called *experiment.cbh*. That database can contain any fuzzy games and headers, and you can experiment with different parameters of the doubles killing programs and see what happen. Your experimenting will not hurt, rather you learn from it. Thus, it is worth a try. ;) Eran Eran
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.