home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!rpi!crdgw1!rdsunx.crd.ge.com!ariel!davidsen
- From: davidsen@ariel.crd.GE.COM (william E Davidsen)
- Newsgroups: comp.unix.xenix.sco
- Subject: Re: Xenix considered harmful (was Re: SCO support - a success story)
- Message-ID: <1992Sep10.175625.7788@crd.ge.com>
- Date: 10 Sep 92 17:56:25 GMT
- References: <9209010926.AA25624@dynamix.com> <437@comix.UUCP>
- Sender: usenet@crd.ge.com (Required for NNTP)
- Reply-To: davidsen@crd.ge.com (bill davidsen)
- Organization: GE Corporate R&D Center, Schenectady NY
- Lines: 51
- Nntp-Posting-Host: ariel.crd.ge.com
-
- In article <437@comix.UUCP>, jeffl@comix.UUCP (Jeff Liebermann) writes:
-
- | >right, and the fact that Xenix CONSISTENTLY OUTPERFORMS the bulky, slow,
- | >overgrown SVR4 kernel is meaningless ... and if the client is NEVER
- | Yep. The "real" problem with Unix is that it does too much
- | for most of my customers. I guess the price of success is
- | bloat and pollution.
-
- I'm not sure how you measure outperform, so I can't say for sure if I
- disagree, but the UNIX compilers produce code which runs faster (for me,
- 20-30 apps measured) and disk i/o is somewhat faster with the same disk
- hardware. Obviously UNIX uses more memory and takes up more disk, but
- UNIX runs stuff faster than Xenix, and I'm comparing Xenix to both SCO
- UNIX and Dell V.4.
-
- | comix runs SCO Unix 3.2v4.1 with every last patch, sls, app, fix,
- | and kludge I can find. I want to know what it will do BEFORE
- | it gets delivered. The name of the game is uptime and reliability.
- | I rotate the latest vendor supplied cards and drivers through
- | it constantly. The suprises are endless. The latest 3.2v4 development
- | system is now 4 times slower than the 3.2v2 dev sys. I guess it's
- | more "standard" now. When my one and only developer/customer
- | found out, he went into hysterics.
-
- Now I have to ask, slower by what measure? I haven't found a compiled
- program yet that runs more than a percent slower, and have many which
- run up to 50% faster using the new SCO compilers. If you mean compile
- speed, I'll beg off that issue, since I haven't measured the timings in
- that area. But 4X? I think I would notice 4X slower on long compiles, so
- I'll state that I don't see that. I would like to see how you got the 4X
- number before I agree or not, but I certainly don't see much difference
- in a 10K line package with multiple compiles and links.
- |
- | Some day, some vendor will produce "Unix Lite" (not Destiny) that's
- | designed as a business platform. I'll be the first in line.
- | This is what I hoped SCO would position Xenix, but didn't.
-
- I suspect that SCO has mixed feelings about Xenix. If they didn't have
- to support it they could do lots of other things. If they didn't sell it
- they wouldn't be nearly as profitable, I suspect, based on both
- estimated sales shown in various magazines and some things SCO people
- said at various times, that Xenix brings in at least 1/3 of the profits,
- since development costs are almost nil. SCO probably won't confirm this,
- and I don't blame them.
-
- I like Xenix for some apps because it runs nicely on a 2MB machine
- (say "notebook").
-
- --
- bill davidsen, GE Corp. R&D Center; Box 8; Schenectady NY 12345
- I admit that when I was in school I wrote COBOL. But I didn't compile.
-