[ILUG] Re: ide cd writer

[Rick Moen]

begin Tony Bolger quotation:
 
> Right and wrong. Connect a narrow/slow scsi device to a fast/wide bus
> and it will kill performance. 

That's why you don't do that.  Slow, high-latency SCSI devices (tape
drives, CD drives, scanners) go on the slow-device chain.

> Though IDE supports disconnects in the latest versions AFAIK....

Yes, we keep hearing these claims.  ATA kinda-sorta can support
disconnect with mumbledy-mumble provisos, ATA will have hot-fix areas
(or hot-fix doesn't matter because of mumbledy-mumble reason), ATA is 
stable and interoperable (no really; this time for sure), ATA
kinda-sorta supports hot-swap with mumbledy-mumble provisos, ATA
kinda-sorts supports elevator seeking and scatter-gather (or they don't
matter because of mumbledy-mumble reason), ATA supports burst-mode /
bus-mastering with mumbledy-mumble provisos, ATA is finally not as
CPU-intensive (no really; this time for sure), and so on.

And these claims are invariably supported by performance metrics that
measure a single task doing a consecutive-sectors access on a single
disk without even trying to compensate for caching.  (Life is simple when
you think like a DOS user.)

And we're told that, if some or all of the above doesn't really work as
claimed, it doesn't matter, because it's cheaper anyway, and therefore
preferable for real-world concerns that seem to approximate being a DOS
user.

As I said yesterday, it's kind of endearing.  Almost like a real I/O
bus.

-- 
Cheers,        "Interfere?  Of course we'll interfere!  Always do what you're
Rick Moen      best at, I always say."  -- The Doctor, _Nightmare of Eden_
rick at linuxmafia.com