Glen Goodwin skrev:
>> Glen Goodwin skrev:
>>
>> >Hmm -- my personal experience is that, unless the machine is already
>> >junked-up with a random assortment of TV and radio tuner cards, video
>> >accelerators, SoundBlasters and DVD decoders, SCSI is easily added to a
>> >motherboard with onboard IDE ports. I've personally built a few dozen
>of
>> >them, using both IDE and PCI SCSI controllers.
>>
>> >What sorts of problems are you encountering?
>Iggy Drougge replied:
>
>> You're welcome to try to sort out our pile of SCSI cards which so far
>haven't
>> worked in our OpenBSD machine.
>One machine, or many? If you have a pile of cards, one machine, and none
>of the cards work in that specific box, I'd suggest that there may be a
>fault in the box, or another device in there which conflicts with your SCSI
>cards.
Who cares what the cause is? The point is that it won't work.
>> We don't use PCI cards, though (we sold that
>> one, since we wouldn't use the SCSI for anything else than tapestreamers
>and
>> so forth). Besides, SCSI integration into PC systems is really clumsy.
>Wow - these days the BIOS setup utility on the cards allows a lot of
>flexibility and also provides information about the installed SCSI devices,
>making it easy to verify IDs, etc.
I see that BIOS setup utility on the cards as a sympthom of the low level of
integration. The cards behave as an alien entity in the computer.
>A few years ago I was building 80MHz 486 systems using Rancho Technologies
>RT1000 8-bit SCSI cards which I got for about $8 each. The cards were
>shipped labeled "not Win95 compatible," but, not only did they work with
>Win95, I never saw a faulty card, or a system in which the RT1000 wouldn't
>work. (Okay, so they were *slow*)
>So, my SCSI experiences have been good.
So have mine, only not with PCs.
>> and they don't behave like the IDE hard drives.
>Well, that's the point, right? ;>)
Not really. IMO a drive is a drive is a drive is a drive.
>> They
>> have their own little BIOSes and things which I'm not used to from other
>> systems.
>Those little BIOSes (the ones with a setup program) are a *big* advantage.
>Just today I was cursing the fact that the BIOS on the ATA-66 controller I
>was installing didn't have a setup program. It took me two hours to get
>all six IDE drives working properly. With a decent SCSI card it would have
>been 15 minutes, tops (barring any bad drives or cables).
Why? I've never had any need for a SCSI BIOS on my SCSI computers.
>> In fact, non-PC systems tend to see IDE as a kind of bastard SCSI
>> instead.
>They may be onto something there . . .
For example, the onboard IDE controller in the Amiga 4000 and 1200 appear to
the system as scsi.device. And NetBSD/amiga only recently switched from "IDE
on SCSI" to the machine-independent WDC device. =)
--
En ligne avec Thor 2.6a.
Every Horse has an Infinite Number of Legs (proof by intimidation):
Horses have an even number of legs. Behind they have two legs, and in front
they have fore-legs. This makes six legs, which is certainly an odd number of
legs for a horse. But the only number that is both even and odd is infinity.
Therefore, horses have an infinite number of legs. Now to show this for the
general case, suppose that somewhere, there is a horse that has a finite
number of legs. But that is a horse of another color, and by the [above]
lemma ["All horses are the same color"], that does not exist.
Received on Fri Oct 12 2001 - 08:12:52 BST