mfinegan@uceng.UC.EDU (michael k finegan) (02/09/90)
We just installed SCO Unix on our 386, and I noticed a dramatic improvement for 'rcc' versus MSC 5.1 (~2/3 execution time). Anyone compare rcc to other 32 bit compilers ? Also: the SCO development package uses MSC/Xenix C compiler as the default, and mentions switches for 386 code. MSC 5.1 under DOS has no such features. Anyone familiar with this? Is it possible to develop 32 bit code with MSC (it can be cross-developed to run under vanilla DOS on a 386 ...) ? Any advice on the least expensive (but not unreliable) 386 box/motherboard accomodating 4 Meg main memory, and 70+ Meg hard disk, >= 20 MHz CPU ? I have seen some DTK systems (20MHz, 40 Meg/28ms, mono, 1 Meg ram - $2k), but don't know if they can have 4 Meg of memory. (you can reply via E-mail about recommendations ...) Thanks, Mike Finegan mfinegan@uceng.UC.EDU
davidsen@sixhub.UUCP (Wm E. Davidsen Jr) (02/12/90)
In article <3576@uceng.UC.EDU> mfinegan@uceng.UC.EDU (michael k finegan) writes: | | We just installed SCO Unix on our 386, and I noticed a dramatic improvement | for 'rcc' versus MSC 5.1 (~2/3 execution time). Anyone compare rcc to other 32 | bit compilers ? Interesting... I have timed things compiled with MSC against things compiled with cc under ix/386 (the xenix executables run there) and the MSC was almost always somewhat faster. -- bill davidsen - sysop *IX BBS and Public Access UNIX davidsen@sixhub.uucp ...!uunet!crdgw1!sixhub!davidsen "Getting old is bad, but it beats the hell out of the alternative" -anon
david@metapyr.UUCP (David Relson) (02/13/90)
Xenix 386 does have a switch so that code can be generated for a 286 or a 386. It's been a year since I used it so some of the following will be a bit indefinite. Xenix actually had two different compilers on it - both Microsoft, one for the 286 and one for the 386. The 386 compiler was more modern (in terms on ANSI compatibility) so code that would compile for the 386 might not compile when the286 option was specified. (It drove me nuts until I realized that it was two different parsers and code generators, not a single parser with two code generators (or one code generator working off of a 286/386 switch).)