home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.os2.misc
- Path: sparky!uunet!ornl!sunova!linac!att!cbnews!shurr
- From: shurr@cbnews.cb.att.com (larry.a.shurr)
- Subject: Re: Kernel Roulette
- Organization: AT&T Bell Laboratories, Columbus, OH
- Date: Thu, 19 Nov 1992 18:15:57 GMT
- Message-ID: <1992Nov19.181557.27605@cbnews.cb.att.com>
- References: <1992Nov16.181114.7476@usage.csd.unsw.OZ.AU>
- Lines: 75
-
- In article <1992Nov16.181114.7476@usage.csd.unsw.OZ.AU> pendrith@spectrum.cs.unsw.oz.au (Mark Pendrith) writes:
- >
- >[Mark has used 4 kernels starting with the GA and has had numerous
- >problems. Furthermore, the problems appear to be getting worse with
- >each new kernel].
-
- >Is it too much to ask for a reliable kernel after having to suffer these four
- >attempts? I'm sure I can't be alone when I say my confidence in the medium-term
- >viability of OS/2 has diminished as a result of this recent "Golden" CSD farce.
- >I believe IBM is losing both time and prestige and is in danger of squandering
- >an ever decreasing window of opportunity with this sort of blundering. It's
- >strange that everyone is asking "Can Microsoft kill off OS/2?" The irony is
- >that only IBM can kill OS/2; and continuing to release unstable version after
- >unstable version is the way to do it. Let's hope it doesn't happen.
-
- I am thankful that for some reason, I have been spared the hassles
- you have endured (knock on wood). Sure, I've had few problems, but
- they've been minor and the fixes were already available by the time
- I needed them (knock on wood again). My most intractable problem
- was periodic Trap 2's (memory errors) which have gone away since I
- changed my XCMOS memory interleave setting (C&T chipset) from 2-way
- to 4-way (better knock on wood again). I've also switched to the
- CSD, which may or may not be as sensitive to memory problems. My
- biggest remaining complaint is the slowness of VIO & VDM windowed
- sessions.
-
- I suspect that the numerous variations on a theme -- clone hardware
- from dozens (hundreds?) of vendors must create a plethora of systems
- with slightly different characteristics, creating subtle compatibility
- problems. Never has there been a "plug compatible" market (to use an
- old term) of such size and diversity, so nobody's ever tried to offer
- OS/2-like software for it before. 'Course this doesn't help you,
- since you seem to have a bad match between hardware and OS/2, and if
- you're like me, you can't just run out and buy another circuit board,
- even if you could be certain what to buy. At most, it only begins to
- explain (assuming I and all the rest who have speculated along these
- lines are correct). I think IBM has created a nearly impossible
- testing situation, perhaps hoping that the variations are not as
- great as they seem, in reality, to be.
-
- It will be interesting to see if Microsoft can cope with the same
- problems. This may explain Microsoft's efforts to position Windows
- NT as a high-end system (as evidenced by the brochure they recently
- sent me extolling the virtues of Windows for Workgroups), hoping
- thereby to avoid problems due to "excessive" diversity. Still, they
- can "position" all they want, but the market will do what it wants.
-
- Unless OS/2 implodes in the marketplace, we can hope that it will
- keep downward pressure on the strongest "positioning" mechanism
- available: price. Microsoft will be less free to jack the price up
- in order to achieve their desired "position." The coexistence of
- OS/2 and NT may make their vendors uncomfortable, but its good for us.
-
- I understand that when Microsoft demonstrated NT recently, they were
- using an Alpha system from Digital rather than an Intel/PC system.
- We have a loaner Alpha here right now, and it looks good thusfar
- and Alpha will be sold in EISA-based PC-type packing at "competitive
- pricing," but I'd be surprised to see clone-like pricing for it.
-
- Implementing on the Alpha was probably easier than on Intel: the
- chip is simple, fast, and NOT Intel compatible, eliminating the
- need to support Intel, MSDOS, Windows 3.x, and the entire range of
- 386-and-above PC hardware. That would make life easier, I'm sure.
- OS/2 3.0 for the RS6000 -- which I presume without really knowing
- is in the works -- will benefit similarly.
-
- I certainly hope that IBM will get it together and produce software
- capable of running reliably on the wide base of systems out there.
- And I still want faster video - here's hoping for favorable pricing
- on the ISA XGA-2.
-
- Larry
- --
- Larry A. Shurr (las@cbnmva.att.com or att!cbnmva!las) speaking only for myself.
- EOR (end-of-ramble)
-