ptw (03/24/83)
We run a 4.1/System III homebrew on a 750 as a software development tool and are at a complete loss for cpu-cycles (according to 'vmstat') from running many compilations of our own, admittedly not very efficient, compiler. We are seeking the nirvana of the correct expansion path and have come up with some hair-brained schemes that range from hoping that 4.2 will come to our rescue, to upgrading to a larger \&'miniframe', to connecting many \&'microframes' with a hypothetical substance that conducts electromagnetic radiation, to plugging micros into our Unibus; and from attached batch (botch) processing, to 'uux'-ing to remote processors, to personal Unices. Any answer would have to solve the problem of maintaining a correct, central copy of Source Code in a Controlled System. Does anyone else have similar problems, comments, or a solution? P. Tucker Withington Automatix Incorporated ...decvax!genradbolton!linus!vaxine!ptw (617) 667-7900 x2044
crc (03/24/83)
The correct upgrade path is to make your compilier more efficient. Bad code can consume any hardware. If you management won't face up to the fact that your dog is eating them out of house and cpu, "We've got too much effort(money) invested in this code to rewrite it!", then your problem isn't hardware or software... Yours for better software and management that can think, Charles Colbert