Author: Christopher R. Dorr
Date: 13:42:44 06/16/00
Go up one level in this thread
I do understand your frustration, but I think it's important to remember that CM7000 is a mass-market program; 99.99% of their customers don't care about the hash table issue. On a modern machine, this is perhaps the difference between a USCF 2550 level and a USCF 2535 level. For the vast majority of users, it makes no difference. At blitz time controls, which is what the vast majority plays most of their games at, this would make even less difference. To fix this would probably require several dozen hours of coding, testing, and distributing. Let's say 50 hours of developer/tester time. At $40 an hour, this is $2000 to fix a bug that makes virtually no difference in strength, and affets a *very* few customers. Additionally, it pulls a developer away from his/her work on the next version. I wouldn't fix it either. The 'professional' programs sell 90% of their programs to afficionados who *do* care about such things. Thus they are much more likely to fix these problems, as they directly impact their bottom line. I think the response of the CM7000 team is both reasonable and to be expected. Chris
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.