Date: Thu, 28 Sep 2000 21:29:16 -0600 (MDT) From: John Galt <galt@inconnu.isu.edu> To: Dag-Erling Smorgrav <des@ofug.org> Cc: Dan Langille <dan@langille.org>, chat@FreeBSD.ORG Subject: Re: CD writers - recommendations Message-ID: <Pine.LNX.4.21.0009282041540.6683-100000@inconnu.isu.edu> In-Reply-To: <xzpk8bxq7a4.fsf@flood.ping.uio.no>
next in thread | previous in thread | raw e-mail | index | archive | help
On 28 Sep 2000, Dag-Erling Smorgrav wrote: > John Galt <galt@inconnu.isu.edu> writes: > > Adaptec 1542 has a Z-80 onboard to share in the disk processing tasks, I'm > > betting that the better cards have beefier processors. I have never seen > > a GP processor on an IDE card. [more blathering elided] > > IDE used to be processor-intensive back when most disks only supported > PIO, but it's not any more. There hasn't been enough change in the spec in the intervening time to categorically state that. In fact, due to concerns of backwards compatibility, I doubt that it will EVER be safe to categorically state that IDE is not CPU intensive: there will be too many wdc's out there disproving it every CPU-hogging second > The reason you can't see a processor on an IDE adapter, by the way, is > that IDE, as its name suggests, places most of the logic on the disk > itself. That's also the reason why IDE channels have masters and > slaves: the controller is actually located on the master, not on your > motherboard or IDE adapter board. SCSI Disks also have processors on the drives, so that one falls flat. In fact, I submit that the average SCSI device does more on-board processing than the average ATA device. Net results: SCSI:CPU does a little processing, host adapter does a little, and device does a little; IDE: CPU does a LOT of processing, controller does none, drive does a little. Add other tasks to each case's CPU: SCSI: CPU timeslices nicely, host adapter takes up a portion of the slack, device takes up the rest; IDE: CPU timeslices badly, controller takes up no slack, drive gets the brunt of the additional work and often fails to maintain constant datapipe. > The reason why ATAPI CD-ROM burners are so sensitive to CPU load is > that they don't support disconnection or tagged queueing, so commands > must be sent sequentially and the CPU must wait for them to complete. So IOW IDE devices throw an interrupt whenever they're operating, SCSI devices throw an interrupt only whenever the host adapter is starving for commands. Sounds exactly like what I wsas saying in the first place, just using proper terminology... > DES > -- You have paid nothing for the preceding, therefore it's worth every penny you've paid for it: if you did pay for it, might I remind you of the immortal words of Phineas Taylor Barnum regarding fools and money? Who is John Galt? galt@inconnu.isu.edu, that's who! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.LNX.4.21.0009282041540.6683-100000>