shahn@hstbme.mit.edu (Samuel Hahn) (05/31/90)
1) A posted a question a few days ago asking if APPLE hard disks had their heads automatically parked on shut down by the system software. I received two conflicting answers. One person said yes, System 6.0.4 and later sends a head parking command to the hard disk. Another said no, no such command is sent because if it were, all hard disks would be autoparked. In my opinion these two views are not necessarily conflicting. My original question was if APPLE hard disks (or those formatted with Apple's HD Setup) are parked. Just because 3rd party drives aren't parked, doesn't mean Apple's couldn't be. Can anyone from Apple confirm/deny my suspicions? 2) SCSI termination: There has been a lot of talk lately about internal vs. external termination for external hard disks (most notably SyQuests). My current set up is a Mac II w/ 40 meg INTERNAL HD and a LW II SC (properly termianted of course!). Now if I wanted to add an external SyQuest, does it matter if it is internally or externally terminated? My IISC is externally terminted so if the SyQuest is internally terminated, can I just make it the end of my SCSI chain and take the terminator off of the IISC? AND if I can do that, will things still work properly, i.e. can I use my printer without having to turn on my SyQuest and vice versa? Please respond by email or follow-up...I'll summarize if requested.. Sam Hahn shahn@hstbme.mit.edu
dolf@fwi.uva.nl (Dolf Starreveld) (06/01/90)
shahn@hstbme.mit.edu (Samuel Hahn) writes: >1) A posted a question a few days ago asking if APPLE hard disks had their >heads automatically parked on shut down by the system software. I received two >conflicting answers. One person said yes, System 6.0.4 and later sends a >head parking command to the hard disk. Another said no, no such command is sent >because if it were, all hard disks would be autoparked. In my opinion these two >views are not necessarily conflicting. My original question was if APPLE hard >disks (or those formatted with Apple's HD Setup) are parked. Just because 3rd >party drives aren't parked, doesn't mean Apple's couldn't be. >Can anyone from Apple confirm/deny my suspicions? I am not from Apple, but ... Whether heads are automatically parked depends on the drive intercepting the shutdown using the start manager. Once this is implemened, the driver might actually send a Stop Unit command to "park". I don't know whether or not Apple driver software does this, but I think it doesn't. In any case, whatever driver software, if the driver can't know whether the system is shut down it can't autopark anyway. This is the case with older system versions that do not implement the start manager. It is definitely not true that a single command could park all hard disks at once, simply because every scsi command must contain a target id (0-6) and there is not ID that functions as a "broadcast" id. Therefore a single command can not do this to all disks! It also is not true that 3rd party hard disks are not parked. Again, it depends on the software. A friend and I have implemented a wide range of SCSI drivers. In general, all mechanism park automatically when powered down. Thus, in general, the parking is never needed and we don't implement it. We do implement it, however, on SyQuest and other removable media drives, for instance on the Sony 650Mb optical mechanisms. The reason is that we allow removal of the media whenever no partitions from that media are mounted. The Stop Unit command stops the spindle motor and retreats the heads. Subsequently a Release Media command is used to actually allow you to remove the media. On the Sony we do this slightly different on shutdown. In that case we not only release, but also initiate cartridge eject. >2) SCSI termination: There has been a lot of talk lately about internal vs. >external termination for external hard disks (most notably SyQuests). My >current set up is a Mac II w/ 40 meg INTERNAL HD and a LW II SC (properly >termianted of course!). >Now if I wanted to add an external SyQuest, does it matter if it is internally >or externally terminated? My IISC is externally terminted so if the SyQuest >is internally terminated, can I just make it the end of my SCSI chain and >take the terminator off of the IISC? AND if I can do that, will things still >work properly, i.e. can I use my printer without having to turn on my SyQuest >and vice versa? Yes, it does matter. First of all, you already have two terminators at both ends of the SCSI chain (internal + printer). If you intend to place the SyQuest somewhere external, but before the LaserWriter, the Syquest should not have any termination! If you want to locate it beyond the printer, the SyQuest should be terminated and not the printer. Note: SyQuest mechanisms are particularly sensitive to improper termination and this will usually show by incorrect functioning. In general there is no difference between internal and external termination, provided it is done correctly. This means a number of things, but the most important to remember is that each terminator pack need power to function. This may either come from the device itself (usually only when the device is stiched on), or from the bus itself (this requires something newer than a MacPlus since that does not put terminator power on the bus. Some devices allow a mix. Internal power when switched on, external when device switched off. Note: If you depend on terminator power through the bus, each and every device must connect the terminator power line from in to out connector, or it won't work! --dolf -- Dolf Starreveld Phone: +31 20 592 5056/5022 (FAX: 5155), TELEX: 10262 HEF NL EMAIL: dolf@fwi.uva.nl (dolf%fwi.uva.nl@hp4nl.nluug.nl) SNAIL: Dept. of Math. and Computing Science, University of Amsterdam, Kruislaan 409, NL-1098 SJ Amsterdam, The Netherlands
alex@grian.cps.altadena.ca.us (Alex Pournelle) (06/28/90)
Perhaps more about this subject than you wanted to know, but I've been a hard week recovering from drives that were NOT properly shut down, so... (A hazard of the data-recovery business is getting overly evangelistic, but You Can Never Have Too Many Backups.) RANDO@applelink.apple.com (Damon Rando) writes: >In most newer drive >designs [there is a] parking mechanism that automatically parks the head stack >in the landing zone when power is turned off. Apple only sells one drive >that still uses stepper motor technology (the 20Meg). My understanding is >that the heads are parked on this drive when "Shut Down" is selected under >Finder (99% sure). (1) depends on the 20, some are smarter than others. The great majority are MiniScribe 8425SAs, which are usually strapped to auto-park after 30 seconds (you can hear a seek operation after about that long on some macs, and it's best to wait for it). (2) Shut Down doesn't do it--Option Shut Down will on some models. (3) But Apple's shutdown code doesn't always send the proper SCSI shutdown command. I've a pair of Priam EM330 drives (wonderful, beautiful, great--shame about the company) which don't even offer parking from their (terrible) driver software. SilverLining, which has been used on one of them, will offer to install a shutdown/spindown vector, but it's grotty and doesn't work on these monsters. (4) SCSI Parker, the public-domain CDEV, DOES work on jus' about everything I have. It's pretty scary to hear the Priams spin down (for they interpret the command as a real shutdown) when I use "Shut Down"; the office gets real quiet. Some 8425SAs will actually spin down, others will just park; still others ignore the command. MiniScribe went through a LOT of revisions of the firmware on that drive! I'm always paranoid, so I do use SCSI Parker. Priam users will especially like the fact that their drives don't do a panic seek at powerdown (SMACK!). C'mon, Apple, put the RIGHT SCSI commands in future Shut Down revisions. This should be mandatory! My views and welcome to them, Alex "Later anthropological experts maintain that the true stake in the heart of the primitive, so-called USENET computer network was the ``No Amanda Walker'' clause in the much-debated User's Covenant. This clause, which was required to be signed in blood by the users before they could access, stated that he/she/it/yx would post no messages which could directly or indirectly be interpreted as having anything to do with work actually done by that individual in a for-pay job. "The net result of the clause, says this theory, was that only those who knew nothing about the subject could comment on it. Noise levels spiraled out of sight and use by the self-appointed experts plummeted. "Yet still others maintain that this was not true, that the entire problem was caused by the sudden siezure of the only known working source to the internetworking bridge software by officers of the so-called Department of Justice..." --"Attempts to achieve electronic communications among the ancients", Jones, Klewicky, et al, Journal of Forensic Anthropology -- Alex Pournelle, freelance thinker Also: Workman & Associates, Data recovery for PCs, Macs, others ...elroy!grian!alex; BIX: alex; voice: (818) 791-7979 fax: (818) 794-2297 bbs: 791-1013; 8N1 24/12/3
esvoboda@cup.portal.com (Edward J Svoboda) (06/30/90)
> (Priam EM330 has terrible driver software, company out of business)
The technology and assets of Priam Corporation were divided among a number
companys after Priam filed under Chapter 11. The company I work for, Atasi
Technology, Inc., purchased the Advanced 5 1/4" drive line and now manufactures
them in the former Priam HQ building in San Jose, CA.
We are in the process of releasing an improved 330MB Macintosh subsystem
called the MacDisk II which now includes a metal enclosure and custom driver
software. I have little experience with the EM330 (a.k.a. MacDisk) but
believe the problem with drive spinning down can be cured by changing the
jumper settings on the Priam Model 738 drive inside the enclosure.
If you have any questions, or would like information on the MacDisk II,
feel free to call Atasi Technology, Inc. at (408) 436-8872.
Ed Svoboda, MIS Director
Atasi Technology, Inc.
esvoboda@cup.portal.com