rem@remsit.UUCP (Roger Murray) (09/08/87)
In article <3905@well.UUCP>, ewhac@well.UUCP (Leo 'Bols Ewhac' Schwab) writes: > Apple pulled this stunt years ago. "Lets save a few cents on the > drive," they said. "Let's leave out the track 0 detect switch," they said. > "We'll just ram the head against the physical stop 39 times," they said. > "It can't possibly do any harm; just give the user indigestion." Being an Apple owner [Yes, I still have my ][+...], I've always wondered about that. Why 39 times (or whatever it may be)? Was there a scientific method used, or did the number just look good to somebody? -- Roger Murray UUCP: ...!{ihnp4,randvax,sdcrdcf,ucbvax}!ucla-cs!cepu!ucla-an!remsit!rem ARPA: cepu!ucla-an!remsit!rem@CS.UCLA.EDU [formerly LOCUS.UCLA.EDU]
kamath@reed.UUCP (Sean Kamath) (09/09/87)
In article <341@remsit.UUCP> rem@remsit.UUCP (Roger Murray) writes: >Being an Apple owner [Yes, I still have my ][+...], I've always wondered >about that. Why 39 times (or whatever it may be)? Was there a scientific >method used, or did the number just look good to somebody? > >-- >Roger Murray Because Apple drives have 35 tracks, and they figured that some folks might have 40 track drives. (most apple drives *are* 40 track drives!) so they pull it back 39 times to get to 0, always. Sometimes a reboot only clacks the head 4 times. Thats the short one. Then again, most of the time it bangs about 20 or so (the head is usually around track 17) Sean kamath -- UUCP: {decvax allegra ucbcad ucbvax hplabs ihnp4}!tektronix!reed!kamath CSNET: reed!kamath@Tektronix.CSNET || BITNET: reed!kamath@Berkeley.BITNET ARPA: tektronix!reed!kamath@Berkeley <or> reed!kamath@hplabs US Snail: 3934 SE Boise, Portland, OR 97202 (I hate 4 line .sigs!)
binder@fizbin.DEC.COM ("A few frilly words...") (09/09/87)
>> "We'll just ram the head against the physical stop 39 times," [Apple] said. >> "It can't possibly do any harm; just give the user indigestion." > Being an Apple owner [Yes, I still have my ][+...], I've always wondered > about that. Why 39 times (or whatever it may be)? Was there a scientific > method used, or did the number just look good to somebody? 39 is a good number because the Apple floppy drive has 35 tracks. No matter where you are, 34 whacks will get you home, with a few more for good measure. The fact that certain third party drives have 40 tracks is not a problem - 39 will get them home, too, although with no margin for error. Cheers, Dick Binder (The Stainless Steel Rat) DEC Enet: FIZBIN::BINDER UUCP: { decvax, allegra, ucbvax... }!decwrl!fizbin.dec.com!binder ARPA: binder%fizbin.DEC@decwrl.DEC.COM
wtm@neoucom.UUCP (Bill Mayhew) (09/10/87)
> why 39 times...?
When the original Apple Disk ][ and it's controller came out, Woz
took a very nihlistic approach. Years back there was an article
quoting Woz where he patted himself on the back about the
wonderfulness of the Disk ][ and how few parts it had. By the way,
he claimed that he even did the art work for the PC board himself.
So, the bottom line is that it was essentially a matter of religion
that the Disk ][ didn't have a track zero sensor Leaving the
sensor out probably saved a flip-flop, inverter gate, and perhaps,
a transistor and a few resistors. It probably saved $0.50 to $1.00
in the drive itself too. Now, if they sold 500,000 drives, that
would mean a lot of extra profit, assuming that apple charged
whatever the market would bear in either case-- which they did.
Note that extra money was saved on the original firmware
development too, since there was no need to deal with worring about
the status of the sensor.
What annoyed me was the 5.25 inch Duodisk. It used relatively
chintzy ALPS (I think) mechanisms that really aren't up to the
pounding of the stepper motor repeatedly at track zero. The metal
band that attaches the head carriage to the motor also makes a lot
more racket than the plastic sprial used in the older Shugart
drives. We've had more trouble with the Duodisks.
On some of our Apple // ][ ][+, etc., we've replaced the drives as
they wear out with Nihon Office Corp. Toshiba based drives. The
Nihon units have an integral controller board that does use a track
zero sensor that prevents them from making the b-r-r-r-r-a-a-t!.
JDR Microdevices sells these drives under the label BAL-500. See
the back of Byte for further info.
By the way, I think that the head bangs 35 times. The assumption
is that the head might be on the highest numbered track,
necessitating 33 steps. I think they throw in a couple of extra
steps to make sure that the follower drops into the sprial groove
in the case of the Disk ][.
I have to give Woz a lot of credit. The Disk ][ was quite an
Amazing accomplishment of finesse over brawn. At the time the Disk
][ was introduced around 1979, typical CP/M machines were using
controllers that had 45+ ICs. The elegance of the Apple design is
very apparent when compared to the kludgey 1541 (also a GCR format)
drive that Commodore brought out a few years later. I always
wondered why Commodore used their ridiuculous seial bus that made
the disk have an effective data transfer rate of about 300 baud!
Bill Mayhew, Electrical Engineer
Division of Basic Medical Sciences
Northeastern Ohio Universities' College of Medicine
Rootstown (what a name!), OH 44272-9989 phone: 216-325-2511
(wtm@neoucom.UUCP ...!cbatt!neoucom!wtm)mkao@pnet01.cts.COM (Mike Kao) (09/10/87)
The reason is that Apple drives (at least certain models) have 80 head phases;
with 2 phases per track, that means a 40-track capability. Since there is no
track 0 detection, if the head is indeed on track 0, then the disk drive has
to recalibrate by slamming the head against the stop 39 more times to MAKE
SURE the head will be on track 0 for bootup.
To insure my reception of any replies, please respond via e-mail. Thanks!
-- Mike Kao
UUCP: {cbosgd, hplabs!hp-sdd, sdcsvax, nosc}!crash!pnet01!mkao
ARPA: crash!pnet01!mkao@nosc.mil
INET: mkao@pnet01.CTS.COMprw@meccsd.MECC.MN.ORG (Paul R. Wenker) (09/11/87)
In article <341@remsit.UUCP> rem@remsit.UUCP (Roger Murray) writes: >In article <3905@well.UUCP>, ewhac@well.UUCP (Leo 'Bols Ewhac' Schwab) writes: >> Apple pulled this stunt years ago. "Lets save a few cents on the >> drive," they said. "Let's leave out the track 0 detect switch," they said. >> "We'll just ram the head against the physical stop 39 times," they said. >> "It can't possibly do any harm; just give the user indigestion." > >Being an Apple owner [Yes, I still have my ][+...], I've always wondered >about that. Why 39 times (or whatever it may be)? Was there a scientific >method used, or did the number just look good to somebody? When an Apple boots up, the firmware has no way of knowing which track the head is on. What it does is assume that the head is on track 80 and then seeks the head out to track 0. This way, no matter where the head is, it is pretty much assured to end up at track 0. On an I/O Error, DOS 3.3 pulls the same trick. If DOS can't find the sector it's looking for, it assumes that it's on the wrong track. It then sets the current track to 39 (35 tracks + 4 for good luck), seeks out to 0 to get it's bearings again, and then trys to find the sector it was looking for. -Paul R. Wenker -MECC