Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!unix.cis.pitt.edu!dsinc!bagate!cbmvax!daveh From: daveh@cbmvax.commodore.com (Dave Haynie) Newsgroups: comp.sys.amiga.hardware Subject: Re: Hardrives and the A3000 Message-ID: <20023@cbmvax.commodore.com> Date: 21 Mar 91 19:51:58 GMT References: <1991Mar19.222827.16244@en.ecn.purdue.edu> <17727@crdgw1.crd.ge.com> <1991Mar20.220547.4677@en.ecn.purdue.edu> Reply-To: daveh@cbmvax.commodore.com (Dave Haynie) Organization: Commodore, West Chester, PA Lines: 69 In article <1991Mar20.220547.4677@en.ecn.purdue.edu> wwarner@en.ecn.purdue.edu (William A Warner) writes: >I will have to say that they "seem" concerned. I got e-mail from >3 C= reps, including Dave H. They are aware of the problem and >have been since the release of the 2091. "The problem", which has been around since the A2091 or so, is a conflict between the scsi.device and the WC33C93A SCSI controller chip, which causes reselection to fail. There also seems to be some interrupt timing conflict in one release of the A2091 ROM; I don't believe that has ever been an issue on the A3000 release of the scsi.device. Both of these problems result in the same thing, a hard disk lockup potential when both drives are simultaneously accessed. The software people can elaborate, but to my knowledge the current release of the A3000 software hides all of the user setup issues in the scsi.device, eg, you can set your drives for reselection if you like, and the driver will start to use that feature as soon as it's capable of doing this safely. Anyway, this problem is well known. On the other hand, there have been a variety of complaints from people who have not been able to [a] get two scsi drives working, period, or [b] get the scsi drives working as fast as Randell does. That isn't the result of this bug. As far as we know, it is a user setup issue. That's why the questions get asked as they do, we're only trying to help out. SCSI isn't my responsibility on the A3000 or Randell's in 2.0, we're just trying to get people going in the most expediant way possible. If you appear to be having a SCSI bus problem, I ask the same questions I would ask myself if I had a problem setting up an A3000 with two drives (I use at least two drives on every one of my A3000 and A2500/30 system): - Got those unit numbers set properly? - Got termination only at the bus ends? Realizing, of course, that early A3000s had termination on the motherboard, later ones didn't, so "I have termination on SCSI drives at either end of the bus" isn't necessarily the right answer. - Got the "Segate" bit set, where appropriate? - Are you sure your cables work? I probably destroy them more than the average person, since I'll plug one cable in over and over again in the lab during system development. Makes me distrust an arbitrary cable. - "SCSI black magic" issues? I have run into SCSI strangeness, which may just be based on the drive types. Like I said, I'm not the SCSI wiz around here. We have a Mac IIcx in the lab which is really testy about anything properly added or deleted from its SCSI bus. I have a 100Meg Conner drive that refuses to work together with any Quantum, but plays just fine with a Sony drive on the bus (well, it did until I dropped the Sony off my lab bench, the Sony now just makes grinding noises). Anyway, there seems to be some bit of mystery (to me at least) that's independent of the host adaptor involved. Fortunately, it doesn't seem to show up very often. And for disk speed questions: - Got your MASK, MaxTransfer, and Memory Type values set up for the fastest transfers. - You aren't using Diskperf, are you! - You're not running out of Fast memory in that test, are you? >PS: who knows....maybe even a "know-it-all" like myself can still learn > a few tricks. I certainly hope so. I learn new stuff every day. If you stop learning, you stagnate and die; that's a basic law of nature. >Art Warner -- Dave Haynie Commodore-Amiga (Amiga 3000) "The Crew That Never Rests" {uunet|pyramid|rutgers}!cbmvax!daveh PLINK: hazy BIX: hazy "What works for me might work for you" -Jimmy Buffett