burton@mitisft.Convergent.COM (Philip Burton) (04/18/89)
Several weeks ago, there was a posting in this group about the successful use of an Adaptec RLL controller (2372 ??) to get 240 MB out of a Maxtor 1140. The writer noted that all 1224 cylinders of the drive were used. Could this person or anyone else with specific experience help me out. I tried to call Adaptec tech support [ 408 945 2550] but the line is ALWAYS busy. My AT is the old kind with only 15 drive types in the ROM BIOS. The largest possible drive is Type 9, which is 15 heads, 900 cylinders. Does Adaptec provide a workaround? My current (stock IBM) controller is limited to 1024 cylinders. How does Adaptec get around that limit? An RLL controller typically yields 26 sectors per track? Does the Adaptec "fool" DOS into thinking that there are more heads/cylinders, or does it actually support 26 sectors? I.e., what does Norton report for drive information? Is 1:1 operaton supported? Is the Adaptec BIOS support purely in hardware, or a combination of hardware and software? Does it go "low enough" into the hardware to fool IBM diagnostics? Does this support work for all versions of DOS, including 4.0? How about OS/2? about Xenix? Please e-mail me, or preferably post to the net for education of all. --Phil Burton --Convergent Technologies
keithe@tekgvs.LABS.TEK.COM (Keith Ericson) (04/19/89)
In article <623@mitisft.Convergent.COM> burton@mitisft.Convergent.COM (Philip Burton) writes: > >Several weeks ago, there was a posting in this group about the successful >use of an Adaptec RLL controller (2372 ??) to get 240 MB out of a Maxtor 1140. That was me. >The writer noted that all 1224 cylinders of the drive were used. > >Could this person or anyone else with specific experience help me out. I tried >to call Adaptec tech support [ 408 945 2550] but the line is ALWAYS busy. > >My AT is the old kind with only 15 drive types in the ROM BIOS. The >largest possible drive is Type 9, which is 15 heads, 900 cylinders. Does >Adaptec provide a workaround? You Betcha'. The BIOS is on the controller card. You tell the AT that you're running with drive type 1. (Actually I think ANY drive type will work as long as you specify SOMETHING; otherwise the onboard formatting program complains that there is no hard disk to work with.) >My current (stock IBM) controller is limited to 1024 cylinders. How does >Adaptec get around that limit? The drive has (#cylinders x #heads x #sectors/track) sectors to store stuff in. This is just a number. Adaptec translates this (cyl x head x sec) value into a different combination wherein "cyl" is <= 1024. So DOS (or is it the BIOS?) asks for a sector on track 985, say, and the controller translates it to the REAL cylinder/head/sector and returns the data. All automatically. Seems to me (caution: cobweb-encrusted memory at work ahead:) that the drive is formatted to 26 sectors/track and translated to 63 per track. (Don't as me - that's just what the formatting program comes up with.) > >An RLL controller typically yields 26 sectors per track? Does the Adaptec >"fool" DOS into thinking that there are more heads/cylinders, or does it >actually support 26 sectors? I.e., what does Norton report for drive >information? Dunno'. My (limited) experience with Norton utilities is that they tend to scramble RLL drives. >Is 1:1 operaton supported? Absolutely; that's one of the best things about the controller. Of course, I'm using 16, 20 and 25 MHz 386AT's. A 6/8 MHz IBM AT (aka Zippo the Wonder Slug) may not be able to handle the transfer rate. >Is the Adaptec BIOS support purely in hardware, or a combination of hardware >and software? Does it go "low enough" into the hardware to fool IBM >diagnostics? I haven't tried IBM's diagnostics. I'd be surprised if they DID work, and I'm not in a position right now to (potentially) sacrifice the contents of a drive to find out. (I'll put it on my "to do list," though.) >Does this support work for all versions of DOS, including 4.0? How about OS/2? >about Xenix? Again, I haven't tried. I was going to install a copy of DOS 4.01 on my machine until I read about all the (other) utilities breaking with the different sized FATs and clusters (?) so I'm sticking with 3.30A. I don't/won't be running Xenix so I cah't help you there, either. The UNIXes I've seen all claim to understand RLL (and SCSI) (and ESDI) drives but I've not seen any of them working with anything other than MFM. (Anyone else have other experience to share? Please?) >Please e-mail me, or preferably post to the net for education of all. > >--Phil Burton > kEITHe PS - specify the Adaptec 2372B - it has a much friendlier formatting routine than does the "A" version.
keithe@tekgvs.LABS.TEK.COM (Keith Ericson) (04/25/89)
I need to correct something in an earlier posting of mine: Someone's question: >> >>An RLL controller typically yields 26 sectors per track? Does the Adaptec >>"fool" DOS into thinking that there are more heads/cylinders, or does it >>actually support 26 sectors? I.e., what does Norton report for drive >>information? My reply: >Dunno'. My (limited) experience with Norton utilities is that they tend to >scramble RLL drives. Well I had my bytes swapped and my bits inverted. It was MS-DOS's chkdsk /f that scrambled the disk and Norton Utilities that was used to repair it. My apologies to Norton and friends. Just this morning another co-worker scrambled his MS-DOS 3.3 file system, on an RLL drive, with chkdsk /f. I used Norton Utilities to repair it. So the problem seems to be chkdsk vs. RLL. kEITHe
pjh@mccc.UUCP (Pete Holsberg) (04/27/89)
In article <4994@tekgvs.LABS.TEK.COM> keithe@tekgvs.LABS.TEK.COM (Keith Ericson) writes:
=Just this morning another co-worker scrambled his MS-DOS 3.3 file system, on
=an RLL drive, with chkdsk /f. I used Norton Utilities to repair it. So the
=problem seems to be chkdsk vs. RLL.
=
=kEITHe
???? I've been running an Adaptek RLL controller with a Rodime 203E for
years, and have upgraded DOS in steps to 3.3. I use chkdsk, Norton,
Mace, PCTools, etc. with no problem whatsoever. Maybe it's the
particular controller??
Pete
--
Pete Holsberg UUCP: {...!rutgers!}princeton!mccc!pjh
Mercer College CompuServe: 70240,334
1200 Old Trenton Road GEnie: PJHOLSBERG
Trenton, NJ 08690 Voice: 1-609-586-4800keithe@tekgvs.LABS.TEK.COM (Keith Ericson) (04/29/89)
In article <685@mccc.UUCP> pjh@mccc.UUCP (Pete Holsberg) writes: >In article <4994@tekgvs.LABS.TEK.COM> I wrote: >=Just this morning another co-worker scrambled his MS-DOS 3.3 file system, on >=an RLL drive, with chkdsk /f. I used Norton Utilities to repair it. So the >=problem seems to be chkdsk vs. RLL. >= > > >???? I've been running an Adaptek RLL controller with a Rodime 203E for >years, and have upgraded DOS in steps to 3.3. I use chkdsk, Norton, >Mace, PCTools, etc. with no problem whatsoever. Maybe it's the >particular controller?? > This is one of those "what the hey?" situations, because, as Pete bears witness, some Adaptec RLL based machines run chkdsk /f just fine. Tom Almy, in the next ofice, uses his sometimes several times a day and hasn't had it mess up yet (he has an Everex Step/20 with an Adaptec RLL). The only thing in common with the two disk trashes are that they are both Intel 301 Motherboard-based systems. And it's possible the users of the trashed systems (intermediate-level DOSsites) could have had something else that precipitated the whole debacle. When/If I get more info on this RLL vs. (?) chkdsk /f I'll pass it on. Until then it deserved an anecdotal/apochryphal designation. kEITHe (Do I have enough lines for the "followup counter" yet?) (There, that ought to do it.) (A couple more for good measure.)