Re: formatting MFM drives on a IBM PC
On 09/30/2017 04:12 PM, Tom Gardner via cctalk wrote: > I think Chuck has it backwards, AT Attachment as defined by the ANSI > committee publically predates IDE. Although IDE was used internally at WD > it did not surface publically until well after the ANSI committee adopted AT > Attachment, abbreviated ATA. The AT in AT Attachment or ATA has never stood > for "Advanced Technology" although many presume so. "IDE" was a Western Digital term for drives used in the Compaq PC, dating from 1986. I've probably got documents from about that time talking about IDE, if I look. Because of Compaq's introduction of the thing early on, that's what we called it then. The "ATA Standard" began its work in 1988 by the Common Access Method committee of ANSI X3T10 and eventually came out with a standard in 1994, but that was long after "IDE" was the lingua franca term for these drives. The ANSI document: https://ecse.rpi.edu/courses/S15/ECSE-4780/Labs/IDE/IDE_SPEC.PDF In, you'll read: "The application environment for the AT Attachment Interface is any computer which uses an AT Bus or 40-pin ATA interface. The PC AT Bus is a widely used and implemented interface for which a variety of peripherals have been manufactured. As a means of reducing size and cost, a class of products has emerged which embed the controller functionality in the drive. These new products utilize the AT Bus fixed disk interface protocol, and a subset of the AT bus. Because of their compatibility with existing AT hardware and software this interface quickly became a de facto industry standard." So, even ANSI X3 talks about the PC AT bus. And yes, "AT", according to IBM, stands for "Advanced Technology" Pretty much, all you need to connect an ATA-1 drive to the 5170 bus is a couple of transceivers and an address decoder. --Chuck
RE: formatting MFM drives on a IBM PC
On Sat, 30 Sep 2017, Tom Gardner via cctalk wrote:
I think Chuck has it backwards, AT Attachment as defined by the ANSI
committee publically predates IDE. Although IDE was used internally at WD
it did not surface publically until well after the ANSI committee adopted AT
Attachment, abbreviated ATA.
Do you, by any chance have the dates?
A casual look seems to show the WD IDE ("Integrated Drive Electronics")
being used by Compaq in 1986?, although not necessarily NAMED "IDE", yet.
A casual look seems to show ANSI ATA as being early 1990s.
The AT in AT Attachment or ATA has never stood
for "Advanced Technology" although many presume so.
DOES it stand for something else?
It is generally assumed that it refers to the IBM PC/AT (aka 5170), where
AT DID, indeed, stand for "Advanced Technology"
RE: formatting MFM drives on a IBM PC
I think Chuck has it backwards, AT Attachment as defined by the ANSI committee publically predates IDE. Although IDE was used internally at WD it did not surface publically until well after the ANSI committee adopted AT Attachment, abbreviated ATA. The AT in AT Attachment or ATA has never stood for "Advanced Technology" although many presume so. A (P)ATA drive never directly connected to an AT bus, nor for that matter any other bus but always required some form of adapter, albeit very simple in the case of the 5170 type bus. The AT Attachment compatibility is with the Task File register set which can be sent over any interface parallel, serial, carrier pigeon, whatever. A SATA drive could be connected to a 5170 with an 5170 bus to SATA bridge or more like a PATA to SATA bridge (with a PATA HBA in the 5170) - PATA to SATA bridges did exist and u might find one on eBay. Booting would be a problem and so would capacity but it should talk. So Serial ATA makes sense to me Tom -Original Message- From: Chuck Guzis [mailto:ccl...@sydex.com] Sent: Thursday, September 28, 2017 5:37 PM To: cctalk@classiccmp.org Subject: Re: formatting MFM drives on a IBM PC On 09/28/2017 05:12 PM, Jules Richardson via cctalk wrote: > On 09/27/2017 09:59 AM, Ethan via cctalk wrote: >> The idea of IDE, as my understanding, is the controller that existed >> as an ISA card was moved onto the actual drive, and then what became >> the controller was mostly just extending the ISA bus over to the >> drive. > > I actually have an IDE "controller" somewhere which is just a tiny PCB > with an ISA connector on one side and a 40 pin IDE connector on the > other, along with a couple of ICs (presumably buffers/latches, but I > don't know without finding it). It's somewhat unusual, given that IDE > ports were normally included as part of multi-I/O boards, or (a little > later) often incorporated into the motherboard. IDE used to be called "ATA" - "AT Attachment"; i.e. something tailored to the PC AT (5170) 16-bit ISA bus. What I find perplexing is the acronym "SATA" for "Serial ATA". The name would imply that a drive can be connected to a 5170, but I'm not aware of any SATA adapters for the 5170 PC/AT. --Chuck
Re: formatting MFM drives on a IBM PC
On 09/29/2017 06:42 PM, Peter Corlett via cctalk wrote: > On Fri, Sep 29, 2017 at 01:08:24PM -0700, Fred Cisin via cctalk wrote: >>> On 09/29/2017 11:20 AM, Jon Elson via cctalk wrote: > [...] >>> Older BIOS firmware provided no means for the user to define the geometry of >>> a connected drive - just a list of predefined types, and those often maxed >>> out at far less than any 512MB limit. There were various software solutions >>> to get around it, though. >>> Of course operating systems had various limits on the maximum size of a >>> partition on top of that - e.g. I think it was 32MB in earlier versions of >>> MS-DOS. >> What are the current drive size limits? > The latest relevant standards seem to be ATA-6 for ATA, and SBC-3 for SCSI, > which have 48- and 64-bit logical block addresses (LBAs) respectively. > > ATA LBAs are *always* 512 bytes no matter what the physical sector size is, so > this sets a hard limit of approximately 144EB. This limit also implies that > sectors must be a multiple of 512 bytes (e.g. modern 4kiB-sectored disks) and > unusual sector sizes are not supported. > > SBC-3 does not specify the size of a block, but it notes that most disks > support 512 bytes with some also supporting 520 or 4096. Assuming 512-byte > logical blocks, the limit is 9.44ZB. > > Since the hard disk equivalent of Moore's Law seems to be running out of steam > at mere tens of terabytes, we probably won't need to raise either limit for a > while yet :) What not mentioned is that LBA addressed drives have cache. With that sector size is meaningless as the track (most likely) or several are read into the cache. There is LRU applied to keep the media up to date and also insure the cache is not stale. With cache it makes doing 512byte blocks a trivial issue. Now we have SSDs... Whole new game same old protocols. Allison
Re: formatting MFM drives on a IBM PC
On Fri, Sep 29, 2017 at 01:08:24PM -0700, Fred Cisin via cctalk wrote: >> On 09/29/2017 11:20 AM, Jon Elson via cctalk wrote: [...] >> Older BIOS firmware provided no means for the user to define the geometry of >> a connected drive - just a list of predefined types, and those often maxed >> out at far less than any 512MB limit. There were various software solutions >> to get around it, though. >> Of course operating systems had various limits on the maximum size of a >> partition on top of that - e.g. I think it was 32MB in earlier versions of >> MS-DOS. > What are the current drive size limits? The latest relevant standards seem to be ATA-6 for ATA, and SBC-3 for SCSI, which have 48- and 64-bit logical block addresses (LBAs) respectively. ATA LBAs are *always* 512 bytes no matter what the physical sector size is, so this sets a hard limit of approximately 144EB. This limit also implies that sectors must be a multiple of 512 bytes (e.g. modern 4kiB-sectored disks) and unusual sector sizes are not supported. SBC-3 does not specify the size of a block, but it notes that most disks support 512 bytes with some also supporting 520 or 4096. Assuming 512-byte logical blocks, the limit is 9.44ZB. Since the hard disk equivalent of Moore's Law seems to be running out of steam at mere tens of terabytes, we probably won't need to raise either limit for a while yet :)
Re: formatting MFM drives on a IBM PC
On 9/29/2017 9:54 PM, Chuck Guzis via cctalk wrote: On 09/29/2017 07:04 PM, Warner Losh via cctalk wrote: I think there are two barriers to that. No sata is a cards. And BIOS of that era still required you to tell it the chs for the drive... So, let's change the acronym to SSATA - "Serial Sort-of AT Attachment" --Chuck SPATA - "Serial Post AT Attachment" -- J.
Re: formatting MFM drives on a IBM PC
On 09/29/2017 07:04 PM, Warner Losh via cctalk wrote: > I think there are two barriers to that. No sata is a cards. And BIOS of > that era still required you to tell it the chs for the drive... So, let's change the acronym to SSATA - "Serial Sort-of AT Attachment" --Chuck
Re: formatting MFM drives on a IBM PC
On Sep 29, 2017 6:49 AM, "Peter Corlett via cctalk" wrote: On Thu, Sep 28, 2017 at 05:36:40PM -0700, Chuck Guzis via cctalk wrote: [...] > What I find perplexing is the acronym "SATA" for "Serial ATA". The name would > imply that a drive can be connected to a 5170, but I'm not aware of any SATA > adapters for the 5170 PC/AT. SATA has a different electrical interface, but otherwise speaks the same protocol as the older (P)ATA protocol. Converters are cheap and readily available. So yes, you could stick a SATA disk in your old 5170, assuming that the BIOS doesn't immediately faint when presented with a very large disk. I think there are two barriers to that. No sata is a cards. And BIOS of that era still required you to tell it the chs for the drive... Warned
Re: formatting MFM drives on a IBM PC
- Original Message - From: "Ali via cctalk" To: "'General Discussion: On-Topic and Off-Topic Posts'" Sent: Friday, September 29, 2017 5:31 PM Subject: RE: formatting MFM drives on a IBM PC > And, that won't work, either. The BIOS in older machines will not work > right at ALL with a high capacity drive. > In some cases you will get a usable volume with vastly reduced > capacity, but in most cases it will just not recognize the drive at > all. I'm not talking about original AT's here, but mid-90's 386 and > 486 systems. > I can only imagine a real AT would be even less likely to handle a > drive over about 40 MB. This is really not a big issue. All you need is a BIOS on the IDE controller card. It was very common back in the day that many add-in drive controllers had their own BIOS so you are not even breaking new ground here. In fact someone far more talented then I could probably design a 16bit controller card w/ its own BIOS that also has the PATA -> SATA adapter interface built in all on one card and your 5170 can be rocking an SSD drive! Heck the XT-IDE project practically does this already with a CF or SD card plugged into the card... -- ... And of course the drive manufacturers supplied drive overlay programs that bypassed the BIOS settings to accommodate larger or undefined drives. m
RE: formatting MFM drives on a IBM PC
> And, that won't work, either. The BIOS in older machines will not work > right at ALL with a high capacity drive. > In some cases you will get a usable volume with vastly reduced > capacity, but in most cases it will just not recognize the drive at > all. I'm not talking about original AT's here, but mid-90's 386 and > 486 systems. > I can only imagine a real AT would be even less likely to handle a > drive over about 40 MB. This is really not a big issue. All you need is a BIOS on the IDE controller card. It was very common back in the day that many add-in drive controllers had their own BIOS so you are not even breaking new ground here. In fact someone far more talented then I could probably design a 16bit controller card w/ its own BIOS that also has the PATA -> SATA adapter interface built in all on one card and your 5170 can be rocking an SSD drive! Heck the XT-IDE project practically does this already with a CF or SD card plugged into the card...
Re: formatting MFM drives on a IBM PC
"Low level format" is pretty much a relic of the old non-servo MFM drives. I recall that early Maxtor IDE drives implemented a LLF On Fri, 29 Sep 2017, Christian Corti via cctalk wrote: Lowlevel formatting has to be done for *all* ST-506 interface drives (e.g. "MFM" and "RLL" drives). It is the disk controller that needs to write its sector and track layout to the drive (ID marks, data marks, GAPs, CRC, ECC and so on). This is also true for SMD drives, for example. So it didn't make much sense in selling preformatted drives until when disk drives exposed only the disk blocks to the host. Whether a drives uses servo information or not is irrelevant. I don't consider writing servo information as lowlevel formatting. You'd think so. But, the exact same thing applied to floppy disks. Formatting has to be done for *all* floppy disks. It is the disk controller/FDC that needs to write its sector and track layout to the drive (ID marks, data marks, GAPs, CRC, and so on. Obviously, they would not be pre-formatted, since the format could be done in multiple ways. But, SOME manufacturers did not provide the user with a way to do that format (DEC, etc.) AND, once a SIGNIFICANT portion of the market had standardized on one particular format, the floppy manufacturers started to sell them pre-formatted. If you need a different format, then you can manually reformat them yourself. (You can erase a PC formatted disk and RE-format it for a few thousand other formats) Similarly, when the majority of the market wanted IBM/WD1003 "AT" drives, preformatting became feasible. "You can RE-format if it's not what you wnat."
Re: formatting MFM drives on a IBM PC
On 09/29/2017 11:20 AM, Jon Elson via cctalk wrote: I can only imagine a real AT would be even less likely to handle a drive over about 40 MB. 40MB would not be much of a problem. DOS, through 3.30 had a limit of 32MB per partition, but you could upgrade to newer DOS, or break it up into multiple partitions. An ST250 (40MB) could be two 20MB, or a 32MB and an 8MB, depending on your type of need. An ST4096 (80MB) could be 3 partitions. On Fri, 29 Sep 2017, Jules Richardson via cctalk wrote: I think the limit was normally 512MB in the old c/h/s addressing days, wasn't it? For a connected drive, the BIOS set aside 4 bits for the number of heads per cylinder, 6 bits for the sectors per track, and 10 bits for the number of cylinders - i.e. maximums of 16, 64 and 1024. At 512 bytes per sector, that came out as 512MB. In addition, there used to be a 2GB limit imposed by DOS. It should have been a 4GB limit, but they used a SIGNED long int, instead of unsigned, so the size of the drive could be from -2147483648 to 2147483647 bytes. Similarly file size was from -2147483648 to 2147483647. If you stepped on a directory entry, you could change a file size to be NEGATIVE! OK, change a file on a floppy to have a file size of 8000 h. DOS now reports it as -2147483648. Or h (-1). Copy a negative sized file to the drive and that should increase the free space, right? Didn't work. Older BIOS firmware provided no means for the user to define the geometry of a connected drive - just a list of predefined types, and those often maxed out at far less than any 512MB limit. There were various software solutions to get around it, though. Of course operating systems had various limits on the maximum size of a partition on top of that - e.g. I think it was 32MB in earlier versions of MS-DOS. What are the current drive size limits?
Re: formatting MFM drives on a IBM PC
For what it's worth, I found long ago that the IDE interface was far closer to the PC-based ESDI controller command set than the WD MFM drive one. ESDI even responds to commands such as IDENTIFY, where that command doesn't exist in the WD1003-type controller vocabulary. ESDI also supports larger (>512MB) drives without breaking a sweat. --Chuck
Re: formatting MFM drives on a IBM PC
> On Sep 29, 2017, at 10:35 AM, Jules Richardson via cctalk > wrote: > > On 09/29/2017 11:20 AM, Jon Elson via cctalk wrote: >> I can only imagine a real AT would be even less likely to handle a drive >> over about 40 MB. > > I think the limit was normally 512MB in the old c/h/s addressing days, wasn't > it? For a connected drive, the BIOS set aside 4 bits for the number of heads > per cylinder, 6 bits for the sectors per track, and 10 bits for the number of > cylinders - i.e. maximums of 16, 64 and 1024. At 512 bytes per sector, that > came out as 512MB. Yes, and I can say that I’m the one responsible for that restriction. :-o At the time we needed to support our LBA controllers and we figured that the restriction was OK because we’d have a BIOS replacement long before disk drives got that big. We were wrong on both counts: drives got bigger a whole lot faster than we expected and that a BIOS replacement took a whole lot longer to make it into systems. *sigh* > > Older BIOS firmware provided no means for the user to define the geometry of > a connected drive - just a list of predefined types, and those often maxed > out at far less than any 512MB limit. There were various software solutions > to get around it, though. The above (512MB max disk size) came about with the PS/2. Prior to that BIOS just had predefined types. TTFN - Guy
Re: formatting MFM drives on a IBM PC
On 09/29/2017 11:20 AM, Jon Elson via cctalk wrote: I can only imagine a real AT would be even less likely to handle a drive over about 40 MB. I think the limit was normally 512MB in the old c/h/s addressing days, wasn't it? For a connected drive, the BIOS set aside 4 bits for the number of heads per cylinder, 6 bits for the sectors per track, and 10 bits for the number of cylinders - i.e. maximums of 16, 64 and 1024. At 512 bytes per sector, that came out as 512MB. Older BIOS firmware provided no means for the user to define the geometry of a connected drive - just a list of predefined types, and those often maxed out at far less than any 512MB limit. There were various software solutions to get around it, though. Of course operating systems had various limits on the maximum size of a partition on top of that - e.g. I think it was 32MB in earlier versions of MS-DOS. cheers Jules
Re: formatting MFM drives on a IBM PC
On 09/29/2017 10:46 AM, Mike Stein via cctalk wrote: - Original Message - From: "Chuck Guzis via cctalk" To: "Paul Koning" ; "General Discussion: On-Topic and Off-Topic Posts" Sent: Friday, September 29, 2017 11:29 AM Subject: Re: formatting MFM drives on a IBM PC On 09/29/2017 06:07 AM, Paul Koning wrote: There are chips that convert between serial and parallel ATA; one of those could perhaps be used. I'm more used to applying them for attaching a serial ATA controller to a parallel ATA drive, but possibly they might work in the other direction as well. Not clear if they are still common products, given that parallel ATA is pretty old, but they may still be available. Oh, I've got a pile of such converters--workability is variable. What I was saying is that I've never seen an ISA interface card to a SATA drive. Have you? --Chuck -- Well, I haven't seen an IDE interface card to a PATA drive either for quite a while since they started integrating everything into the motherboard but OK, I'll grant you that you'd probably need a PATA>SATA adapter if you wanted to Attach a SATA drive to a genooine IBM AT. And, that won't work, either. The BIOS in older machines will not work right at ALL with a high capacity drive. In some cases you will get a usable volume with vastly reduced capacity, but in most cases it will just not recognize the drive at all. I'm not talking about original AT's here, but mid-90's 386 and 486 systems. I can only imagine a real AT would be even less likely to handle a drive over about 40 MB. Jon
Re: formatting MFM drives on a IBM PC
> On Sep 29, 2017, at 11:29 AM, Chuck Guzis wrote: > > On 09/29/2017 06:07 AM, Paul Koning wrote: > >> There are chips that convert between serial and parallel ATA; one of >> those could perhaps be used. I'm more used to applying them for >> attaching a serial ATA controller to a parallel ATA drive, but >> possibly they might work in the other direction as well. Not clear >> if they are still common products, given that parallel ATA is pretty >> old, but they may still be available. > > Oh, I've got a pile of such converters--workability is variable. > > What I was saying is that I've never seen an ISA interface card to a > SATA drive. Have you? No, but my exposure is pretty limited. I ran into these devices in the first generation of SAN product I worked on around 2002. We adopted SATA because it allows connecting lots of drives (14, in our case) over a backplane, without running out of space. But at the time, only "PATA" drives were available, so we used converters. Also, those converters supplied dual channel operation (one from each of a pair of redundant controllers). Storage protocols are asymmetric, so it isn't necessarily the case that a P-S ATA converter that works for SATA adapter to PATA drive will also work for the other case. But it might well, you'd have to look at the specs (and then verify that they are true). It's not a case I have had to worry about so I don't have any additional information. paul
Re: Reviving ancient MFM drives (was Re: formatting MFM drives on a IBM PC)
On Fri, Sep 29, 2017 at 3:05 AM, Christian Corti via cctalk < cctalk@classiccmp.org> wrote: > On Thu, 28 Sep 2017, Geoffrey Oltmans wrote: > >> Speaking of I've got a couple of old MFM drives (10 and 20 MB of a >> variety whose name and model #'s escape me, I wanna say Tandon, but not >> sure). They seem to work fine when I initially format and partition, but >> as >> they run for a while, they get more and more unreliable. It seems to be a >> function of how long they've been running for rather than a predictable >> pattern of bad tracks sectors? Are there any good sources of >> troubleshooting info at the controller level for these old drives? >> > > Well, that's normal. The usual procedure is to let the drive warm up for > 10-20 minutes before formatting. And it is also normal for some models that > they must be reformatted after, say, a couple of months or years, depending > on make and manufacturer. The Rhodime 50MB drive in my IBM 8550 is such a > beast. My procedure is to run Norton CALIBRAT to reformat the drive > losslessly. > > > Good idea. I'll give that a shot.
Re: formatting MFM drives on a IBM PC
- Original Message - From: "Chuck Guzis via cctalk" To: "Paul Koning" ; "General Discussion: On-Topic and Off-Topic Posts" Sent: Friday, September 29, 2017 11:29 AM Subject: Re: formatting MFM drives on a IBM PC > On 09/29/2017 06:07 AM, Paul Koning wrote: > >> There are chips that convert between serial and parallel ATA; one of >> those could perhaps be used. I'm more used to applying them for >> attaching a serial ATA controller to a parallel ATA drive, but >> possibly they might work in the other direction as well. Not clear >> if they are still common products, given that parallel ATA is pretty >> old, but they may still be available. > > Oh, I've got a pile of such converters--workability is variable. > > What I was saying is that I've never seen an ISA interface card to a > SATA drive. Have you? > > --Chuck -- Well, I haven't seen an IDE interface card to a PATA drive either for quite a while since they started integrating everything into the motherboard but OK, I'll grant you that you'd probably need a PATA>SATA adapter if you wanted to Attach a SATA drive to a genooine IBM AT. Sigh... ;-) m
Re: formatting MFM drives on a IBM PC
On 09/29/2017 01:11 AM, Christian Corti via cctalk wrote: > I don't consider writing servo information as lowlevel formatting. I should have been more precise. What I meant was "embedded servo". Most, if not all, modern drives use this technique rather than dedicate a separate surface to servo. There's no low-level format with this. It's been around since at least the 1970s, but was slow to be adopted by the PC drive manufacturers. An interesting aside is that the Drivetec floppy diskettes used embedded servo--which was why you had to purchase them preformatted. I recall visiting a friend at Tandon's "R&D" operation during the 80s. The servo writer for the lab was an affair mounted on a granite block and employed a laser for determining position. At any rate, I hope this clears up what I meant. --Chuck
Re: formatting MFM drives on a IBM PC
On Thu, Sep 28, 2017 at 7:36 PM, Chuck Guzis via cctalk < cctalk@classiccmp.org> wrote: > > What I find perplexing is the acronym "SATA" for "Serial ATA". The name > would imply that a drive can be connected to a 5170, but I'm not aware > of any SATA adapters for the 5170 PC/AT. > I'm sure you're probably aware that a command set was part of the original ATA and has persistently been enhanced over time. I think that it pays more homage to the command set part of the specification rather than the physical interface.
Re: formatting MFM drives on a IBM PC
On 09/29/2017 06:07 AM, Paul Koning wrote: > There are chips that convert between serial and parallel ATA; one of > those could perhaps be used. I'm more used to applying them for > attaching a serial ATA controller to a parallel ATA drive, but > possibly they might work in the other direction as well. Not clear > if they are still common products, given that parallel ATA is pretty > old, but they may still be available. Oh, I've got a pile of such converters--workability is variable. What I was saying is that I've never seen an ISA interface card to a SATA drive. Have you? --Chuck
Re: formatting MFM drives on a IBM PC
> On Sep 28, 2017, at 8:36 PM, Chuck Guzis via cctalk > wrote: > > On 09/28/2017 05:12 PM, Jules Richardson via cctalk wrote: >> On 09/27/2017 09:59 AM, Ethan via cctalk wrote: >>> The idea of IDE, as my understanding, is the controller that existed >>> as an >>> ISA card was moved onto the actual drive, and then what became the >>> controller was mostly just extending the ISA bus over to the drive. >> >> I actually have an IDE "controller" somewhere which is just a tiny PCB >> with an ISA connector on one side and a 40 pin IDE connector on the >> other, along with a couple of ICs (presumably buffers/latches, but I >> don't know without finding it). It's somewhat unusual, given that IDE >> ports were normally included as part of multi-I/O boards, or (a little >> later) often incorporated into the motherboard. > > IDE used to be called "ATA" - "AT Attachment"; i.e. something tailored > to the PC AT (5170) 16-bit ISA bus. > > What I find perplexing is the acronym "SATA" for "Serial ATA". The name > would imply that a drive can be connected to a 5170, but I'm not aware > of any SATA adapters for the 5170 PC/AT. There are chips that convert between serial and parallel ATA; one of those could perhaps be used. I'm more used to applying them for attaching a serial ATA controller to a parallel ATA drive, but possibly they might work in the other direction as well. Not clear if they are still common products, given that parallel ATA is pretty old, but they may still be available. paul
Re: formatting MFM drives on a IBM PC
On Thu, Sep 28, 2017 at 05:36:40PM -0700, Chuck Guzis via cctalk wrote: [...] > What I find perplexing is the acronym "SATA" for "Serial ATA". The name would > imply that a drive can be connected to a 5170, but I'm not aware of any SATA > adapters for the 5170 PC/AT. SATA has a different electrical interface, but otherwise speaks the same protocol as the older (P)ATA protocol. Converters are cheap and readily available. So yes, you could stick a SATA disk in your old 5170, assuming that the BIOS doesn't immediately faint when presented with a very large disk.
Re: formatting MFM drives on a IBM PC
On Thu, 28 Sep 2017, Chuck Guzis wrote: "Low level format" is pretty much a relic of the old non-servo MFM drives. I recall that early Maxtor IDE drives implemented a LLF Lowlevel formatting has to be done for *all* ST-506 interface drives (e.g. "MFM" and "RLL" drives). It is the disk controller that needs to write its sector and track layout to the drive (ID marks, data marks, GAPs, CRC, ECC and so on). This is also true for SMD drives, for example. So it didn't make much sense in selling preformatted drives until when disk drives exposed only the disk blocks to the host. Whether a drives uses servo information or not is irrelevant. I don't consider writing servo information as lowlevel formatting. Christian
Re: Reviving ancient MFM drives (was Re: formatting MFM drives on a IBM PC)
On Thu, 28 Sep 2017, Geoffrey Oltmans wrote: Speaking of I've got a couple of old MFM drives (10 and 20 MB of a variety whose name and model #'s escape me, I wanna say Tandon, but not sure). They seem to work fine when I initially format and partition, but as they run for a while, they get more and more unreliable. It seems to be a function of how long they've been running for rather than a predictable pattern of bad tracks sectors? Are there any good sources of troubleshooting info at the controller level for these old drives? Well, that's normal. The usual procedure is to let the drive warm up for 10-20 minutes before formatting. And it is also normal for some models that they must be reformatted after, say, a couple of months or years, depending on make and manufacturer. The Rhodime 50MB drive in my IBM 8550 is such a beast. My procedure is to run Norton CALIBRAT to reformat the drive losslessly. Christian
RE: Reviving ancient MFM drives (was Re: formatting MFM drives on a IBM PC)
Old 10 and 20 MB MFM drives are most likely open loop positioning systems which are highly vulnerable to off track due to thermal changes and stiction. They also could have flying height problems due to contamination on the slider. These failure modes can manifest themselves a slow soft problems such as you are describing One thing I would do is format them only after they are warmed up and well exercised and then I would never turn them off. If I did turn them off then I would have a batch file do a bunch of seeks for a minute or so during the boot and if I were really clever find a way to do a seek every 5 minutes or so. Tom -Original Message- From: Adrian Stoness [mailto:tdk.kni...@gmail.com] Sent: Thursday, September 28, 2017 9:41 AM To: Geoffrey Oltmans; General Discussion: On-Topic and Off-Topic Posts Subject: Re: Reviving ancient MFM drives (was Re: formatting MFM drives on a IBM PC) prolly a failing ic of some sort? On Thu, Sep 28, 2017 at 11:21 AM, Geoffrey Oltmans via cctalk < cctalk@classiccmp.org> wrote: > On Thu, Sep 28, 2017 at 11:17 AM, Al Kossow via cctalk < > cctalk@classiccmp.org> wrote: > > > > > > > On 9/28/17 7:38 AM, Mike Stein via cctalk wrote: > > > > > What is it that usually fails when the drive can't read the servo info? > > The data on the platter, or? > > > > I've never dug that far into it beyond fiddling with Micropolis > > trying to mechanically get it to find the servo tracks and calibrate > > to track 0. > > > > One of the problems is schematics and documentation on the servo > > systems are extremely difficult to get. The little that is on > > bitsavers is all I've come up with in 25 years of searching and > > there is practically nothing useful elsewhere on the web about > > fixing servos in old 5" disks. > > > > It could be heads, media, positioners, component aging in the analog > > section, etc etc.. > > > > > > > > > > Speaking of I've got a couple of old MFM drives (10 and 20 MB of a > variety whose name and model #'s escape me, I wanna say Tandon, but > not sure). They seem to work fine when I initially format and > partition, but as they run for a while, they get more and more > unreliable. It seems to be a function of how long they've been running > for rather than a predictable pattern of bad tracks sectors? Are there > any good sources of troubleshooting info at the controller level for these > old drives? >
Re: formatting MFM drives on a IBM PC
On 09/28/2017 06:03 PM, Fred Cisin via cctalk wrote: > For a little while, there were also available some 8 bit IDE cards > and drives! I think that Compaq may have been the first customer for > the Western Digital IDE "Integrated Drive Electronics" Yup, it's sometimes called XTA. Very different from ATA--resembles the Xebec XT disk controller layout. --Chuck
Re: formatting MFM drives on a IBM PC
On 2017-09-28 10:03 PM, Fred Cisin via cctalk wrote: On Thu, 28 Sep 2017, Chuck Guzis via cctalk wrote: IDE used to be called "ATA" - "AT Attachment"; i.e. something tailored to the PC AT (5170) 16-bit ISA bus. For a little while, there were also available some 8 bit IDE cards and drives! I think that Compaq may have been the first customer for the Western Digital IDE "Integrated Drive Electronics" That is the persistent story apparently Compaq was looking for a compact solution to cram into their luggables some of the first one where just a regular MFM drive with the controller grafted onto them ( had one came out of a portable II) later the electronics became compact enough to put everything onto a single card. What I find perplexing is the acronym "SATA" for "Serial ATA". The name would imply that a drive can be connected to a 5170, but I'm not aware of any SATA adapters for the 5170 PC/AT. It evolved, and the name persisted, in spite of no longer being applicable. It is call SATA because it is ATA protocol over a serial link, just like SAS is SCSI over a serial link. Paul.
Re: formatting MFM drives on a IBM PC
On Thu, 28 Sep 2017, Chuck Guzis via cctalk wrote: IDE used to be called "ATA" - "AT Attachment"; i.e. something tailored to the PC AT (5170) 16-bit ISA bus. For a little while, there were also available some 8 bit IDE cards and drives! I think that Compaq may have been the first customer for the Western Digital IDE "Integrated Drive Electronics" What I find perplexing is the acronym "SATA" for "Serial ATA". The name would imply that a drive can be connected to a 5170, but I'm not aware of any SATA adapters for the 5170 PC/AT. It evolved, and the name persisted, in spite of no longer being applicable.
Re: formatting MFM drives on a IBM PC
On 09/28/2017 05:12 PM, Jules Richardson via cctalk wrote: > On 09/27/2017 09:59 AM, Ethan via cctalk wrote: >> The idea of IDE, as my understanding, is the controller that existed >> as an >> ISA card was moved onto the actual drive, and then what became the >> controller was mostly just extending the ISA bus over to the drive. > > I actually have an IDE "controller" somewhere which is just a tiny PCB > with an ISA connector on one side and a 40 pin IDE connector on the > other, along with a couple of ICs (presumably buffers/latches, but I > don't know without finding it). It's somewhat unusual, given that IDE > ports were normally included as part of multi-I/O boards, or (a little > later) often incorporated into the motherboard. IDE used to be called "ATA" - "AT Attachment"; i.e. something tailored to the PC AT (5170) 16-bit ISA bus. What I find perplexing is the acronym "SATA" for "Serial ATA". The name would imply that a drive can be connected to a 5170, but I'm not aware of any SATA adapters for the 5170 PC/AT. --Chuck
Re: formatting MFM drives on a IBM PC
On 09/27/2017 09:59 AM, Ethan via cctalk wrote: The idea of IDE, as my understanding, is the controller that existed as an ISA card was moved onto the actual drive, and then what became the controller was mostly just extending the ISA bus over to the drive. I actually have an IDE "controller" somewhere which is just a tiny PCB with an ISA connector on one side and a 40 pin IDE connector on the other, along with a couple of ICs (presumably buffers/latches, but I don't know without finding it). It's somewhat unusual, given that IDE ports were normally included as part of multi-I/O boards, or (a little later) often incorporated into the motherboard.
Re: formatting MFM drives on a IBM PC
- Original Message - From: "Al Kossow via cctalk" To: Sent: Thursday, September 28, 2017 12:17 PM Subject: Re: formatting MFM drives on a IBM PC > > > On 9/28/17 7:38 AM, Mike Stein via cctalk wrote: > >> What is it that usually fails when the drive can't read the servo info? The >> data on the platter, or? > > I've never dug that far into it beyond fiddling with Micropolis trying to > mechanically get it to > find the servo tracks and calibrate to track 0. > > One of the problems is schematics and documentation on the servo systems are > extremely difficult > to get. The little that is on bitsavers is all I've come up with in 25 years > of searching and there > is practically nothing useful elsewhere on the web about fixing servos in old > 5" disks. > > It could be heads, media, positioners, component aging in the analog section, > etc etc.. > - Too bad. I'm probably not the only one who has drives that sound fine (i.e. no bad bearings, scratched surfaces etc.) but will not initialize; they twitch back and forth on startup, presumably looking for track 0, and then give up. Seems like a shame to scrap them if only there were some remedy... m
Re: formatting MFM drives on a IBM PC
- Original Message - From: "Guy Sotomayor Jr via cctalk" Sent: Thursday, September 28, 2017 12:30 PM Subject: Re: formatting MFM drives on a IBM PC > Generally drives now-a-days will read an entire track at a time. >TTFN - Guy -- Cromemco's STDC ST412 MFM controller also read a whole track at a time way back when; quite an improvement over the IMI controllers they'd used previously. Speaking of IMI drives, some of the 5" versions (as opposed to the aforementioned 8" 7710) were available in both ST412 and 34-pin IMI interface versions, and conversion kits were even available. m
Re: formatting MFM drives on a IBM PC
> On Sep 28, 2017, at 12:04 PM, Jon Elson via cctalk > wrote: > > On 09/28/2017 10:31 AM, Chuck Guzis via cctalk wrote: >> And even the sector size on recent drives is fictitious. Larger drives use >> 4K sectors but "translate" them to the 512 byte standard. > So, for every write, it needs to read the 4K sector, alter the bytes and then > rewrite it? Hmmm, explains why writes are slower than reads even on spinning > disks. Of course, if you write entire tracks, it can just hold the partial > data until it has all the data for that track, and write out the buffer in > one rotation. Yes, "512 emulation" drives are a performance problem and I don't trust the firmware either. It's a whole lot more complex than a regular disk drive. If you have 4k physical sectors, the better answer is to use 4k sectors natively. Current OS versions have no problem with this. That way you avoid hairy and slow drive firmware. On geometry, it's quite some time ago that drive designers reinvented non-uniform track lengths. CDC did this way back in the early 1960s, but then for a long time sector counts tended to be the same on all tracks. CDs changed that, and hard drives started doing so also. As a result, the C/H/S values you get from a drive are unavoidably a fiction, because even if the designers were willing to give you the real number, they couldn't possibly do so because there isn't a single answer. paul
Re: Reviving ancient MFM drives (was Re: formatting MFM drives on a IBM PC)
prolly a failing ic of some sort? On Thu, Sep 28, 2017 at 11:21 AM, Geoffrey Oltmans via cctalk < cctalk@classiccmp.org> wrote: > On Thu, Sep 28, 2017 at 11:17 AM, Al Kossow via cctalk < > cctalk@classiccmp.org> wrote: > > > > > > > On 9/28/17 7:38 AM, Mike Stein via cctalk wrote: > > > > > What is it that usually fails when the drive can't read the servo info? > > The data on the platter, or? > > > > I've never dug that far into it beyond fiddling with Micropolis trying to > > mechanically get it to > > find the servo tracks and calibrate to track 0. > > > > One of the problems is schematics and documentation on the servo systems > > are extremely difficult > > to get. The little that is on bitsavers is all I've come up with in 25 > > years of searching and there > > is practically nothing useful elsewhere on the web about fixing servos in > > old 5" disks. > > > > It could be heads, media, positioners, component aging in the analog > > section, etc etc.. > > > > > > > > > > Speaking of I've got a couple of old MFM drives (10 and 20 MB of a > variety whose name and model #'s escape me, I wanna say Tandon, but not > sure). They seem to work fine when I initially format and partition, but as > they run for a while, they get more and more unreliable. It seems to be a > function of how long they've been running for rather than a predictable > pattern of bad tracks sectors? Are there any good sources of > troubleshooting info at the controller level for these old drives? >
Re: formatting MFM drives on a IBM PC
> On Sep 28, 2017, at 9:04 AM, Jon Elson via cctalk > wrote: > > On 09/28/2017 10:31 AM, Chuck Guzis via cctalk wrote: >> And even the sector size on recent drives is fictitious. Larger drives use >> 4K sectors but "translate" them to the 512 byte standard. > So, for every write, it needs to read the 4K sector, alter the bytes and then > rewrite it? Hmmm, explains why writes are slower than reads even on spinning > disks. Of course, if you write entire tracks, it can just hold the partial > data until it has all the data for that track, and write out the buffer in > one rotation. Given that many drives have megabytes of “cache” it’s not really an issue. There is *a lot* of buffering going on in drives. Generally drives now-a-days will read an entire track at a time. They don’t even wait for the index (sector 0), they start reading at the first sector and fill in the buffer as the sectors come under the head. Rotational latency still shows up at the interface if you’re wanting a *specific* sector, you have to wait for it to be read but generally the drive will read the entire track so the next sector will be just uCode latency for the command and the transfer time over the interface. Writing is a bit more of a challenge (because you *do* want the data to make it to the surface) but in general they are buffered also and there’s some interesting stuff going on in the drive to make sure that all of the data is written properly. TTFN - Guy
Re: formatting MFM drives on a IBM PC
On 9/28/17 10:21 AM, Geoffrey Oltmans wrote: On Wed, Sep 27, 2017 at 11:16 AM, allison via cctalk mailto:cctalk@classiccmp.org>> wrote: IDE disks format usually meant high level only. SCSI could be either depnding on the specific controller and media. Seems like the omission of low level formatting of IDE drives had more to do with preserving the servo track data, (and maybe aforementioned firmware/bad sector data) yes? IDE was supposed to be near plug and play so low level formatting was not required and undesired. With that most did not low level format, just leaving room for the likely exceptions. The key was the drive interface was embedded on the drive to remove the need for a board in the bus. Even IDE cam in flavors, LBA, bus interface speeds, then large LBA types as we know them now. That also includes the CF cards as they had an IDE mode as well as 8bit data bus interface mode. Its all PATA... (IDE, EIDE, ATAPI...) SCSI/SASI was the exception for all. It ws a system to device interface and the endpoint could be a drive controller with formatting capability and a raw ST506 drive or it could be a Drive very similar to IDE (Embedded Disk Electronics) with a SCSI interface. Again I've used all flavors, CP/M (Micromint SB180 with SCSI adapter) and Xybec controller to a 20mb ST225, PC with Adaptec and D540, microVAX with CMD SCSI (bus card) and M3100s with internal SCSI and RZxx type drives. Were any earlier MFM/RLL voice coil/servo controlled, or were they all stepper drives? YES to all. The early drives came in all flavors. SA412(10mb) was stepper, Quantum D540 was voice coil with embedded servo yet it could be reformatted (RQDX1/2 and RQDX3 had different formats and I've used the same after formatting on Teltek(CPM), PC (1003MFM and 1005-RLL controllers), plus my PDP11s/MicroVAX (RQDX3 andRQDX2) which are all very different. Generally Drives of the MFM or RLL (ST506/412) interface were low level formatted though exceptions existed. There was a lot of evolution in disks within the PC world some ideas and developments originated from outside that and others were exported. SCSI and MFM interfaces were from outside PCs realm and added in and IDE was uniquely PC in origin though it was a morph of the "host controllers" which were bus level interfaces similar to but not IDE. For a few years you could get the same controller for a MFM drive that could have Host interface, SCSI, IDE, and all the various PC busses (MicroChannel, ISA8, ISA16, EISA a and more), and then buses from DEC, DG and others. That list is far from complete but serves the example. Allison
Reviving ancient MFM drives (was Re: formatting MFM drives on a IBM PC)
On Thu, Sep 28, 2017 at 11:17 AM, Al Kossow via cctalk < cctalk@classiccmp.org> wrote: > > > On 9/28/17 7:38 AM, Mike Stein via cctalk wrote: > > > What is it that usually fails when the drive can't read the servo info? > The data on the platter, or? > > I've never dug that far into it beyond fiddling with Micropolis trying to > mechanically get it to > find the servo tracks and calibrate to track 0. > > One of the problems is schematics and documentation on the servo systems > are extremely difficult > to get. The little that is on bitsavers is all I've come up with in 25 > years of searching and there > is practically nothing useful elsewhere on the web about fixing servos in > old 5" disks. > > It could be heads, media, positioners, component aging in the analog > section, etc etc.. > > > > Speaking of I've got a couple of old MFM drives (10 and 20 MB of a variety whose name and model #'s escape me, I wanna say Tandon, but not sure). They seem to work fine when I initially format and partition, but as they run for a while, they get more and more unreliable. It seems to be a function of how long they've been running for rather than a predictable pattern of bad tracks sectors? Are there any good sources of troubleshooting info at the controller level for these old drives?
Re: formatting MFM drives on a IBM PC
On 9/28/17 7:38 AM, Mike Stein via cctalk wrote: > What is it that usually fails when the drive can't read the servo info? The > data on the platter, or? I've never dug that far into it beyond fiddling with Micropolis trying to mechanically get it to find the servo tracks and calibrate to track 0. One of the problems is schematics and documentation on the servo systems are extremely difficult to get. The little that is on bitsavers is all I've come up with in 25 years of searching and there is practically nothing useful elsewhere on the web about fixing servos in old 5" disks. It could be heads, media, positioners, component aging in the analog section, etc etc..
Re: formatting MFM drives on a IBM PC
On 09/28/2017 10:31 AM, Chuck Guzis via cctalk wrote: And even the sector size on recent drives is fictitious. Larger drives use 4K sectors but "translate" them to the 512 byte standard. So, for every write, it needs to read the 4K sector, alter the bytes and then rewrite it? Hmmm, explains why writes are slower than reads even on spinning disks. Of course, if you write entire tracks, it can just hold the partial data until it has all the data for that track, and write out the buffer in one rotation. Jon
Re: formatting MFM drives on a IBM PC
On 09/28/2017 07:34 AM, Al Kossow via cctalk wrote: > Most drives 40mb and up are closed-loop with dedicated > servo surfaces and many have servos that don't work any more. > > Maxtor and Atasi are early examples of embedded servo drives. > You can tell if a drive is embedded or dedicated by the number > of heads, odd if dedicated, even if embedded. For a lot of larger IDE drives, the "geometry" shown on the drive label (if it's shown) is a convenient fiction. Drives have used schemes such as zone recording for many years, but software expects the same number of sectors per track on a drive. And even the sector size on recent drives is fictitious. Larger drives use 4K sectors but "translate" them to the 512 byte standard. "Low level format" is pretty much a relic of the old non-servo MFM drives. I recall that early Maxtor IDE drives implemented a LLF operation, which would essentially render the drive unusable. Heck, there wasn't a lot of agreement on the IDE interface initially, so, for instance some Maxtor drives would report the total number of sectors on a drive with the capacity words swapped. Things have come a long way. --Chuck
Re: formatting MFM drives on a IBM PC
On 27 September 2017 at 20:25, Fred Cisin via cctalk
wrote:
> On Wed, 27 Sep 2017, Liam Proven wrote:
>>
>> ... as usual, lots of high-quality info. Can't disagree with any of it.
>
>
> thanks. many errors, due to inadequately refreshed dynamic wet-ware RAM
> Chuck will probably notice most of them.
:-)
>> Vanilla PC-DOS and MS-DOS didn't get past 3.3, TTBOMK.
>
>
> My turn to nit-pick.
>
> There never was ANY version 3.3 !
> And certainly no three point three!
> You mean version 3 point THIRTY! (3.30)
Oh, OK.
Honestly, as a holder of a battered dusty old science degree, the
number _after_ the decimal is only used digit-by-digit. It's never
three point thirty, it's three point three zero, and that is
equivalent to three point three.
> 3.30 gave an AX of 1E03 ("three point thirty")
> 3.31 gave an AX of 1F03 ("three point thirty-one")
> 2.10 gave an AX of 0A02 ("two point ten")
> 2.11 gave an AX of 0B02 ("two point eleven")
> "three point three" would be AX of 0303, and would be 3.03
Makes me wince, but I defer.
> How about ZENITH?
Never saw them here. Virtually a US-only company AFAIK.
> Were the tweaks done BY Compaq, or by Compaq and Microsoft, in preparation
> for support in later versions of DOS?
Don't know.
> 3.31 was only available as OEM versions of MS-DOS. There was no PC-DOS
> 3.31.
AIUI, yes.
> Some OEMs had enhancements.
So I heard, I think. Long time ago. Don't think I ever _saw_ any myself.
> 2.11 (0B02) was another version that was ONLY OEMs, often with some very
> strange changes/enhancements, such as support for odd drives such as 3.5"
> (not supported mainstream until 3.20 (1403)), MODE for switching between
> internal/external video and 8 or 16 line internal displays (Gavilan 2.11),
> etc.
> BTW, Gavilan started (NCC 1983) with 3.0" drive, then single sided 3.5"
> (SA300), and about the time that Gavilan destroyed itself (1985), they had
> double sided 3.5" (SA350). Gavilan 3.5" double sided disk formats were
> different from PC-DOS 3.5" formats until Gavilan 2.11K (unofficially
> released after Gavilan was GONE.) In converting a Gavilan to 720K, to get a
> clean look, transfer the bezel from a Gavilan SA300 to a stock SA350.
> Gavilan was one of MANY early laptops, a full year later than Grid Compass,
> but Gavilan appears likely to have been the first to use the term "laptop".
Before my time, I'm afraid.
I came in at 3.20 -- as in some old machines still had it.
In the UK, PCs were too expensive for most people except wealthy
businesses until Amstrad launched the PC 1512 and 1640.
So the clone industry happened in the UK after 1986.
> PC-DOS 4.00 (0004) was "buggy". "The new DOS is so buggy that Norton
> Utilities won't work!" How many of those "bugs" were simply CHANGES that
> Norton fUtilities and the like were not prepared for?
> PC-DOS 4.01 was supposedly "fixed" (think in a veterinary context?).
> I had a copy of PC-DOS 4.01 that returned 0004! "ALL bugs fixed?"
Famously so. :-)
I didn't study internal version numbers, though.
> Prior to 5.00 (0005), MS-DOS was "only for sale with a computer, or as
> upgrade to such". In THEORY, it was not available for retail sale, but
> there was a GIANT grey market, with no difficulty at all finding and buying
> copies. There was NO apparent effort by Microsoft to rein in the gray
> market.
Yup.
DR-DOS changed that.
First version, 3.41 (?), very minor. Never saw it.
DR-DOS 5 though was a huge hit. We sold loads. I liked it. Preferred it, even.
> 5.00 was the first version with a RETAIL channel. It was ten years after
> IBM and Microsoft signed - their contract obviously permitted Microsoft
> selling to OEMs, but was there a clause in their contract that forbade
> RETAIL sales [for ten years]?
Only because MS gazed upon DR's sales and became (even more) covetous.
> BTW, "PC-DOS" was "descriptive" ("Personal Computer Disk Operating System"),
> and was NOT a trademark. I personally confirmed that in the stacks of the
> Patent nadTrademark Office in Virginia in 1987.
(!!)
> At least until after DRI brought "Concurrent PC-DOS" to market.
> IBM was not amused.
> http://online.wsj.com/public/resources/documents/googleglassuspto.pdf
> "MS-DOS", however WAS trademarked.
No flies on MS.
> SETVER (starting with 5.00) was even more fun.
> Prior to that, one of the early assignments that I gave my Assembly Language
> class was to modify EXE2BIN (our copy which came with PC-DOS 2.00 and the
> IBM release of MASM 1.0, was locked to DOS 2.00) to be DOS version tolerant.
See my comment to Alison -- I need to wrangle that now for PC DOS 7.1.
> There was another, even more bizarre way to handle large drives, even larger
> than 512MB!
> In DOS 3.10 (0A03), they introduced the [undocumented?] "network
> redirector". Remember MSCDEX?
> 3.10 had the 32MB limit. But, a CD-ROM was 2/3 GB (about 660MB). They
> handled it with smoke and mirrors. If you tried to CHKDSK a CD-ROM
> CHKDSK D:
> It returned an error message:
Re: formatting MFM drives on a IBM PC
On 27 September 2017 at 18:29, allison via cctalk wrote: >> > Everyone forgets Norton Utilities... Well, no! Still use it, in fact. I have a copy of the last DOS version. A low-priority project of mine is that I am trying to make a running VirtualBox VM with PC-DOS 7.1. Not 7.01, but the version from the IBM Server Scripting Toolkit, with LBA and FAT32 support. My plan, if I can, is to get a few of the significant GUIs & my favourite apps running. FreeGEM, ViewMax, GEOS, DOSShell, DESQview & DESQview/X. And maybe MS Word 6, PC Outline, WordPerfect 6. Some stuff I could do actual productive work in, even now. But only a little bit of PC DOS 7.1 was released -- no SETVER, no SYS, no FORMAT, etc. So making a bootable disk is proving tricky. Norton is helping but only slightly. No full success yet: only a bootable floppy image, nothing more. But I don't remember NU doing low-levels. I know NDD could do sector testing and surface analysis, but only on DOS-formatted volumes, IIRC. -- Liam Proven • Profile: https://about.me/liamproven Email: lpro...@cix.co.uk • Google Mail/Talk/Plus: lpro...@gmail.com Twitter/Facebook/Flickr: lproven • Skype/LinkedIn/AIM/Yahoo: liamproven UK: +44 7939-087884 • ČR/WhatsApp/Telegram/Signal: +420 702 829 053
Re: formatting MFM drives on a IBM PC
- Original Message - From: "Al Kossow via cctalk" To: Sent: Thursday, September 28, 2017 10:34 AM Subject: Re: formatting MFM drives on a IBM PC > > > On 9/28/17 7:21 AM, Geoffrey Oltmans via cctalk wrote: >> Were any earlier MFM/RLL voice coil/servo >> controlled, or were they all stepper drives? >> > > Most drives 40mb and up are closed-loop with dedicated > servo surfaces and many have servos that don't work any more. > > Maxtor and Atasi are early examples of embedded servo drives. > You can tell if a drive is embedded or dedicated by the number > of heads, odd if dedicated, even if embedded. > What is it that usually fails when the drive can't read the servo info? The data on the platter, or? m
Re: formatting MFM drives on a IBM PC
On 9/28/17 7:21 AM, Geoffrey Oltmans via cctalk wrote: > Were any earlier MFM/RLL voice coil/servo > controlled, or were they all stepper drives? > Most drives 40mb and up are closed-loop with dedicated servo surfaces and many have servos that don't work any more. Maxtor and Atasi are early examples of embedded servo drives. You can tell if a drive is embedded or dedicated by the number of heads, odd if dedicated, even if embedded.
Re: formatting MFM drives on a IBM PC
On Wed, Sep 27, 2017 at 11:16 AM, allison via cctalk wrote: > >> > IDE disks format usually meant high level only. SCSI could be either > depnding on the specific controller and media. > > Seems like the omission of low level formatting of IDE drives had more to do with preserving the servo track data, (and maybe aforementioned firmware/bad sector data) yes? Were any earlier MFM/RLL voice coil/servo controlled, or were they all stepper drives?
Re: formatting MFM drives on a IBM PC
On 9/27/17 10:59 AM, Ethan via cctalk wrote: IIRC, the first time I had problems with the low level format was with one of the early IDE controllers and a 230MB Maxtor. Crapped out the entire firmware, was never able to get it to admit who it was again. Seemed to work okay with earlier MFM/RLL 40 MB and 80 MB Conner drives (I think, it's been a while). AFAIK a lot of IDE drives store part of the firmware on the spinning disk in a special section of the disk. Not sure if those early models used that trick to cut costs or not? Not so for MFM. All and IDE drive is in essence is a MFM with a WD1003 controller as IDE is the buss level view of the 1003 (or the later 1005 RLL encoded version). The idea of IDE, as my understanding, is the controller that existed as an ISA card was moved onto the actual drive, and then what became the controller was mostly just extending the ISA bus over to the drive. Correct. My first hard drive was a SCSI-1 ?Fuji? on a Seagate 8-bit ISA card. Families Tandy 1000sx. I remember in the end playing with low level formatting tools and interleves, then the drive dying at the same time. I correlated the two together then, but looking back I think the issue was drive motor/bearings/stuck rotation of platters. My first drive was a ST506 on S100 using the Teltek controller (CP/M) I was an early adopter and 5MB felt like a a lot of space when floppies were maybe 720K (2side double density 80track) back in 1980. After that it included over time St506, st412(RD50), RD51(also sold as MFM for PCs 31mb) then SA225(RD31), sa250 (RD32), Quantum D540 (RD52) and a long list of others including most of the drives to 150mb. Those larger than 30mb woere used in my QBUS PDP11's and MicroVAX systems. Least that was true up to about 93ish when I started using PCs (XT class running dos/WIN3.11). DEC for MFM RDxx disks had different formats depending on system those being Rainbow, PRO, QBUS(RQDX1,2 or RQDX3) which was PDP11 and MicroVAX( MV2000 and QBUS). That's for Low level format. FOr High level format for the different file systems were handled by the OS. While they were sometimes called format it was really writing a bland initial filesstem on formatted media. IDE disks format usually meant high level only. SCSI could be either depnding on the specific controller and media. PCs forced the issue with multiple controlled not all the same For example the ST225 could be formatted with WD1003(plain MFM) or WD1005 (RLL encoded). There were many versions. In the CP/M s100 days you had teltek, Konan and many others all different. Media must be formatted for the controller and there was code supplied for that. Those were the bus installed controller as Xyebc, WD, Adaptec also had host controllers as well as SASI/SCSI interfaced versions to drive MFM drives. The whole of that is 5.25 disks and the later 3.5inch IDE class devices. Other formats were well, different. Allison
Re: formatting MFM drives on a IBM PC
On 9/27/17 12:04 PM, Liam Proven via cctalk wrote: On 26 September 2017 at 21:33, Phil Blundell via cctalk wrote: Low-level formatting (which, at the time, was just called "formatting") used to be quite a routine operation on ST-506 MFM and RLL hard disks. They usually came completely blank from the factory and you had to format them according to whatever sector layout and interleave your particular controller wanted before they were usable. Once the drive was formatted you then had to run a separate process to lay out an actual filesystem. All true, although by the time I entered the industry in 1988 or so, it was normal for drives to ship low-level formatted, at least. I remember that Netware came with a special tool called COMPSURF to do a "comprehensive surface analysis". There's still a passing mention of this here: https://support.novell.com/subscriptions/readmes/114.html Everyone forgets Norton Utilities...
Re: formatting MFM drives on a IBM PC
I'm suffering from TL;DR disease this morning, so I didn't have the inclination to follow all of the links cited in the discussion, so my apology is presented in advance. However, there *was* another way to handle large drives in earlier DOS before 4.0. It was far from satisfactory, because it broke a lot of utilities. The basic scheme was to insert a "layer" in-between the PC Int 13H services and DOS and block I/O transactions up into larger virtual sectors.There was nothing sacred about a hard drive sector being 512 bytes; indeed, the PC98 versions of MS-DOS use 1024 byte sectors on their floppies. So, for example, if you were limited by 16-bit relative sector addresses to 33MB (calling 1MB = 1,000,000 bytes), you could block 2 512-byte physical sectors up into a single "virtual" sector of 1024 bytes and extend the reach of DOS to 65 MB. Make it larger, say, 2048 bytes and you get 134MB. As I said, this came at a cost of memory (you need buffers for these bigger sectors) and a lot of utilities that assumed a 512 byte sector, but as a desperate dodge, it worked surprisingly well. There was also a slight performance penalty, as most MFM hard drives there used 17 sectors per track--or 25 if they were RLL, so there was the awkward problem of a write possibly crossing a cylinder boundary. --Chuck
Re: formatting MFM drives on a IBM PC
On Wed, 27 Sep 2017, Liam Proven wrote:
... as usual, lots of high-quality info. Can't disagree with any of it.
thanks. many errors, due to inadequately refreshed dynamic wet-ware RAM
Chuck will probably notice most of them.
There were several additional programs, that were sometimes needed, such as
if you wanted to have a partition larger than 32MB on DOS 3.30 or earlier
(MS-DOS 3.31 was first to accept larger partitions).
Until this bit!
Vanilla PC-DOS and MS-DOS didn't get past 3.3, TTBOMK.
My turn to nit-pick.
There never was ANY version 3.3 !
And certainly no three point three!
You mean version 3 point THIRTY! (3.30)
Internally, starting with 2.00, the major version was stored as an
integer, and the minor version was a two digit decimal number.
INT21h Fn30h (place 30h in AH and INT21h) returned the major version in
AL, and the minor version in AH.
3.30 gave an AX of 1E03 ("three point thirty")
3.31 gave an AX of 1F03 ("three point thirty-one")
2.10 gave an AX of 0A02 ("two point ten")
2.11 gave an AX of 0B02 ("two point eleven")
"three point three" would be AX of 0303, and would be 3.03
A program to replicate the functionality of the VER command in a
program was an assignment that I used to give my Assembly Languuge
class.
In C, formatting would be
printf("$d.%02d",_AL,_AH);
to avoid 4.01 coming out as 4.1
My nitpick might only be important if you have 6.01 and 6.10, etc.
The first DOS I saw that could handle >32MB partitions was Compaq DOS
3.31 -- they tweaked it slightly.
How about ZENITH?
I thought that I had seen other 3.31s, but I never used them, so I'm not
sure.
Were the tweaks done BY Compaq, or by Compaq and Microsoft, in preparation
for support in later versions of DOS?
3.31 was only available as OEM versions of MS-DOS. There was no PC-DOS
3.31. Some OEMs had enhancements.
2.11 (0B02) was another version that was ONLY OEMs, often with some very
strange changes/enhancements, such as support for odd drives such as 3.5"
(not supported mainstream until 3.20 (1403)), MODE for switching between
internal/external video and 8 or 16 line internal displays (Gavilan 2.11),
etc.
BTW, Gavilan started (NCC 1983) with 3.0" drive, then single sided 3.5"
(SA300), and about the time that Gavilan destroyed itself (1985), they had
double sided 3.5" (SA350). Gavilan 3.5" double sided disk formats were
different from PC-DOS 3.5" formats until Gavilan 2.11K (unofficially
released after Gavilan was GONE.) In converting a Gavilan to 720K, to
get a clean look, transfer the bezel from a Gavilan SA300 to a stock SA350.
Gavilan was one of MANY early laptops, a full year later than Grid
Compass, but Gavilan appears likely to have been the first to use the term
"laptop".
PC-DOS 4.00 (0004) was "buggy". "The new DOS is so buggy that Norton
Utilities won't work!" How many of those "bugs" were simply CHANGES that
Norton fUtilities and the like were not prepared for?
PC-DOS 4.01 was supposedly "fixed" (think in a veterinary context?).
I had a copy of PC-DOS 4.01 that returned 0004! "ALL bugs fixed?"
Prior to 5.00 (0005), MS-DOS was "only for sale with a computer, or as
upgrade to such". In THEORY, it was not available for retail sale, but
there was a GIANT grey market, with no difficulty at all finding and
buying copies. There was NO apparent effort by Microsoft to rein in the
gray market.
5.00 was the first version with a RETAIL channel. It was ten years after
IBM and Microsoft signed - their contract obviously permitted Microsoft
selling to OEMs, but was there a clause in their contract that
forbade RETAIL sales [for ten years]?
BTW, "PC-DOS" was "descriptive" ("Personal Computer Disk Operating
System"), and was NOT a trademark. I personally confirmed that in the
stacks of the Patent nadTrademark Office in Virginia in 1987.
At least until after DRI brought "Concurrent PC-DOS" to market.
IBM was not amused.
http://online.wsj.com/public/resources/documents/googleglassuspto.pdf
"MS-DOS", however WAS trademarked.
SETVER (starting with 5.00) was even more fun.
Prior to that, one of the early assignments that I gave my Assembly
Language class was to modify EXE2BIN (our copy which came with PC-DOS
2.00 and the IBM release of MASM 1.0, was locked to DOS 2.00) to be DOS
version tolerant.
That employer only sold IBM and Compaq kit, and being that the UK was
poorer back then, mostly I only saw those and other cheaper clone PCs.
We didn't get to see some of the other States-side premium ranges, or
*I* didn't, until later, in the early-to-mid 1990s. So other vendors
may have had larger-disk-partition hacks, too. I recall reading of
some -- Wikipedia claims Commodore, Leading Edge, AST, NEC, AT&T,
Tandy, Sperry & Unisys all fiddled with FAT formats.
There was another, even more bizarre way to handle large drives, even
larger than 512MB!
In DOS 3.10 (0A03), they introduced the [undocumented?] "network
redirector". Remember MSCDEX?
3.10 had the 32MB limit
Re: formatting MFM drives on a IBM PC
On 26 September 2017 at 21:33, Phil Blundell via cctalk wrote: > > Low-level formatting (which, at the time, was just called "formatting") > used to be quite a routine operation on ST-506 MFM and RLL hard disks. > They usually came completely blank from the factory and you had to > format them according to whatever sector layout and interleave your > particular controller wanted before they were usable. Once the drive > was formatted you then had to run a separate process to lay out an > actual filesystem. All true, although by the time I entered the industry in 1988 or so, it was normal for drives to ship low-level formatted, at least. I remember that Netware came with a special tool called COMPSURF to do a "comprehensive surface analysis". There's still a passing mention of this here: https://support.novell.com/subscriptions/readmes/114.html -- Liam Proven • Profile: https://about.me/liamproven Email: lpro...@cix.co.uk • Google Mail/Talk/Plus: lpro...@gmail.com Twitter/Facebook/Flickr: lproven • Skype/LinkedIn/AIM/Yahoo: liamproven UK: +44 7939-087884 • ČR/WhatsApp/Telegram/Signal: +420 702 829 053
Re: formatting MFM drives on a IBM PC
On 26 September 2017 at 20:53, Fred Cisin via cctalk wrote: ... as usual, lots of high-quality info. Can't disagree with any of it. > There were several additional programs, that were sometimes needed, such as > if you wanted to have a partition larger than 32MB on DOS 3.30 or earlier > (MS-DOS 3.31 was first to accept larger partitions). Until this bit! Vanilla PC-DOS and MS-DOS didn't get past 3.3, TTBOMK. The first DOS I saw that could handle >32MB partitions was Compaq DOS 3.31 -- they tweaked it slightly. It didn't help me inasmuch as at the time, the main use for such "huge" disks was in fileservers -- and my company used 3Com 3+Share, a weird MS-DOS based fileserver with internal multitasking. It was _very_ picky about what additional DOS utilities it would work with. For larger hard disks, the only supported tool was something from Golden Bow Systems, later known for their Vopt disk-defragger. I tried with Compaq DOS 3.31 -- it died messily, as I recall. I also tried a special memory managed for 80286 PS/2s which could turn the 384 kB of XMS into EMS, which 3+Share could use for a disk cache. Yeah, that crashed a customer's live fileserver, corrupting thousands of files. And I did it, so I had to restore all the files they'd edited, one by one, from backups. On floppy diskette. And for the hundred-odd they'd edited since the last backup, undelete them, one by one. Days of work. It was the only disciplinary measure I got, as it was unpaid for the client and immensely tedious (and humiliating) for me. One customer had a Model 80 (a 386) machine with a 330MB hard disk, but wouldn't spring for Golden Bow's disk manager. It had drives C:, D:, E:, F:, G:, H:, I:, J:, K: and L:. Muggins here had to arrange a pattern of disk shares to try and usefully employ all that space. I did manual hashing of user's home directories for some of it. So, anyway, yes, circa 1989, DOS supported disk partition sizes were a subject of intense professional interest for me, and really, seriously, the only 3.x era MS-DOS family OS I ever saw with >32MB support was Compaq's. It ran fine on any machine. I used it on several. I may still have a copy somewhere. That employer only sold IBM and Compaq kit, and being that the UK was poorer back then, mostly I only saw those and other cheaper clone PCs. We didn't get to see some of the other States-side premium ranges, or *I* didn't, until later, in the early-to-mid 1990s. So other vendors may have had larger-disk-partition hacks, too. I recall reading of some -- Wikipedia claims Commodore, Leading Edge, AST, NEC, AT&T, Tandy, Sperry & Unisys all fiddled with FAT formats. Golden Bow's _didn't_ work with MS-DOS 4 and later, which had built-in support for larger partitions. Compaq DOS did -- I think MS, or rather IBM, picked up Compaq's schema and used it. Later it was named FAT16B or BigDOS. https://en.wikipedia.org/wiki/File_Allocation_Table#Logical_sectored_FAT > And, you needed an overlay, such as ONTRACK, to use a drive larger than > 504MB. > Also, if you had a drive whose geometry was too incompatible with your > computer - not all CMOS/SETUPs had a user-defined drive parameters option. Oh my yes -- that was a whole 'nother world of pain. Even in the late 1990s, I was occasionally dealing with this. I incrementally expanded a friend's cheapo Dell Pentium-133 PC for him. 64 MB RAM, IDT WinChip CPU upgrade, bigger hard disk. The original drive was a 120MB or so. The BIOS couldn't handle drives over 512MB. (No Logical Block Addressing.) So I installed OnTrack Disk Manager, moved his Win95B install to the new drive, converted it to FAT32 with PartitionMagic, hooked up his old drive as a slave, made it drive D:, and secured it in place with duct tape and cable ties. Years later he handed it on to his dad. Later, his dad asked a friend to look at the machine for him. Word was passed back to dad, to my mate, and to me: "Blimey. Whoever your son's mate was who upgraded this PC for him, he did an amazing job. I've never seen such an old PC tricked out as much as this, and it's great workmanship." I remain inordinately pleased by that, some 20y later... -- Liam Proven • Profile: https://about.me/liamproven Email: lpro...@cix.co.uk • Google Mail/Talk/Plus: lpro...@gmail.com Twitter/Facebook/Flickr: lproven • Skype/LinkedIn/AIM/Yahoo: liamproven UK: +44 7939-087884 • ČR/WhatsApp/Telegram/Signal: +420 702 829 053
Re: formatting MFM drives on a IBM PC
> On Sep 26, 2017, at 10:20 PM, Chuck Guzis via cctalk > wrote: > > On 09/26/2017 09:53 PM, Mike Stein via cctalk wrote: > >> Ah yes, the interesting 11MB IMI 7710. Cromemco also used them in their >> Z-2H; I still have one somewhere, not a bad drive when it worked ;-) > > The really crazy thing is that we were taking our hardware over to > Viking Labs to do temperature, humidity and shake table testing, along > with hipot. We were pretty proud that things kept working even after > the CRT neck fractured. That reminds me of one of the “chamber” tests that was being run on an XT. The XT was in the chamber running diagnostics while the chamber was doing temperature and humidity cycling. The lab techs left it running over a long weekend. Over the weekend there was a power failure (not infrequent in southern Florida) and the chamber when it rebooted ran it’s default program which was to raise the temperature to 70C and remain there for 24 hours. The sight that greeted us when we came into the lab was funny. The monitor (I think it was monochrome) melted (I should say the plastic did) and the floppy disk in the drive was rippled. However, we rebooted the XT (after removing the floppy from the drive) and it worked (including the monitor). TTFN - Guy
Re: formatting MFM drives on a IBM PC
IIRC, the first time I had problems with the low level format was with one of the early IDE controllers and a 230MB Maxtor. Crapped out the entire firmware, was never able to get it to admit who it was again. Seemed to work okay with earlier MFM/RLL 40 MB and 80 MB Conner drives (I think, it's been a while). AFAIK a lot of IDE drives store part of the firmware on the spinning disk in a special section of the disk. Not sure if those early models used that trick to cut costs or not? The idea of IDE, as my understanding, is the controller that existed as an ISA card was moved onto the actual drive, and then what became the controller was mostly just extending the ISA bus over to the drive. My first hard drive was a SCSI-1 ?Fuji? on a Seagate 8-bit ISA card. Families Tandy 1000sx. I remember in the end playing with low level formatting tools and interleves, then the drive dying at the same time. I correlated the two together then, but looking back I think the issue was drive motor/bearings/stuck rotation of platters.
Re: formatting MFM drives on a IBM PC
> On Sep 27, 2017, at 7:55 AM, Bill Gunshannon via cctalk > wrote: > > ... > I have never had to do that. The only boxes I ever LLF disks for > were PDP-11's and VAX. And then, only once, on initial install. DEC was all over the map as far as disk formatting goes. Some packs were delivered blank and had to be formatted first (RK05 for example). Some were delivered formatted and there wasn't any format operation (RL01, RP04). And then there were the fixed head devices (RC11, RF11, or more accurately RS64 and RS11 drives) which were delivered formatted, but might have to be reformatted in the field by specialized field service devices for certain repairs. I've seen it done on an RF11 that had a head crash, in 1975 in college. Stranger still are the Pro hard drives, which can be formatted but the tools are deeply buried; I remember seeing RT11 code that can do it but I don't remember if other operating systems include the capability. What VAX drives had a format capability? The PDP-11 ones I can think of weren't supported on VAX. paul
RE: formatting MFM drives on a IBM PC
From: cctalk [cctalk-boun...@classiccmp.org] on behalf of Fred Cisin via cctalk [cctalk@classiccmp.org] Sent: Tuesday, September 26, 2017 7:53 PM To: Ali; General Discussion: On-Topic and Off-Topic Posts Subject: Re: formatting MFM drives on a IBM PC > > I remember at least one manufacturer >recommending it for their > > drive(s) if they were ever tilted through 90 degrees - >presumably > > there were tiny effects on the head positioning and so not >doing a > > LLF would result in problems. On Tue, 26 Sep 2017, Ali via cctalk wrote: > This was pretty common wisdom back in the day. Not quote sure how wise > it was but it was generally recommended in the magazines of the time. I > remember reading an article about it in PC Magazine. There were some interesting discussions of that when companies, such as Compaq, first started to put hard drives in portables! I have never had to do that. The only boxes I ever LLF disks for were PDP-11's and VAX. And then, only once, on initial install. bill
Re: formatting MFM drives on a IBM PC
On Tue, 26 Sep 2017, emanuel stiebler wrote: So, what is the best(?) or easiest piece of software, to format the drives, check for bad blocks, etc.? I like the "CMS Fixed Disk Diagnostics" very much, the file is FDIAG.COM It can be found here: ftp://computermuseum.informatik.uni-stuttgart.de/utils/FDIAG.COM Christian
Re: formatting MFM drives on a IBM PC
On 09/26/2017 09:53 PM, Mike Stein via cctalk wrote: > Ah yes, the interesting 11MB IMI 7710. Cromemco also used them in their Z-2H; > I still have one somewhere, not a bad drive when it worked ;-) The really crazy thing is that we were taking our hardware over to Viking Labs to do temperature, humidity and shake table testing, along with hipot. We were pretty proud that things kept working even after the CRT neck fractured. Something like the IMI drive was downright ludicrous. I know that shortly thereafter, they merged with Onyx--you could see their building on N. First Street with the "Onyx + IMI" logo on it. --Chuck
Re: formatting MFM drives on a IBM PC
- Original Message - From: "Chuck Guzis via cctalk" To: "Jules Richardson via cctalk" Sent: Tuesday, September 26, 2017 8:38 PM Subject: Re: formatting MFM drives on a IBM PC > On 09/26/2017 03:53 PM, Jules Richardson via cctalk wrote: >> On 09/26/2017 01:19 PM, Ethan via cctalk wrote: >>> I don't know if it's a good idea to low level format a drive or not. >> >> I remember at least one manufacturer recommending it for their drive(s) >> if they were ever tilted through 90 degrees - presumably there were tiny >> effects on the head positioning and so not doing a LLF would result in >> problems. > > I remember quite vividly evaluating the then-new IMI hard drive--the > thing with a smoke-black plexiglas top and about the size of a shoebox. > I think Corvus sold them as Apple II addons. > > It used a voice-coil positioner. If you lifted the front of the drive > to about a 10 degree (or more) inclination, the positioner would go > nuts, making all sorts of noise trying to stay on track--and eventually > erroring out. > > --Chuck - Ah yes, the interesting 11MB IMI 7710. Cromemco also used them in their Z-2H; I still have one somewhere, not a bad drive when it worked ;-) http://www.s100computers.com/Hardware%20Manuals/Cromemco/IMI_7710%20Hard%20Disk%20Brochure.pdf https://amaus.org/static/S100/cromemco/photos/Cromemco%20Z2H/IMI%20drive%202.jpg m
Re: formatting MFM drives on a IBM PC
On Tue, 26 Sep 2017, Chuck Guzis via cctalk wrote: I remember quite vividly evaluating the then-new IMI hard drive--the thing with a smoke-black plexiglas top and about the size of a shoebox. I think Corvus sold them as Apple II addons. It used a voice-coil positioner. If you lifted the front of the drive to about a 10 degree (or more) inclination, the positioner would go nuts, making all sorts of noise trying to stay on track--and eventually erroring out. "Let's put that in a computer, and put a handle on it, so that people can lug it around!, just like we doodled on the pieshop placemat. If anybody has a problem, just remind them that we warned them to low-level format anytime that they move it."
Re: formatting MFM drives on a IBM PC
On 09/26/2017 03:53 PM, Jules Richardson via cctalk wrote: > On 09/26/2017 01:19 PM, Ethan via cctalk wrote: >> I don't know if it's a good idea to low level format a drive or not. > > I remember at least one manufacturer recommending it for their drive(s) > if they were ever tilted through 90 degrees - presumably there were tiny > effects on the head positioning and so not doing a LLF would result in > problems. I remember quite vividly evaluating the then-new IMI hard drive--the thing with a smoke-black plexiglas top and about the size of a shoebox. I think Corvus sold them as Apple II addons. It used a voice-coil positioner. If you lifted the front of the drive to about a 10 degree (or more) inclination, the positioner would go nuts, making all sorts of noise trying to stay on track--and eventually erroring out. --Chuck
Re: formatting MFM drives on a IBM PC
IIRC, the first time I had problems with the low level format was with one of the early IDE controllers and a 230MB Maxtor. Crapped out the entire firmware, was never able to get it to admit who it was again. Seemed to work okay with earlier MFM/RLL 40 MB and 80 MB Conner drives (I think, it's been a while). On Tue, Sep 26, 2017 at 7:58 PM, Warner Losh via cctalk < cctalk@classiccmp.org> wrote: > On Tue, Sep 26, 2017 at 5:53 PM, Fred Cisin via cctalk < > cctalk@classiccmp.org> wrote: > > > > I remember at least one manufacturer >recommending it for their > > >> drive(s) if they were ever tilted through 90 degrees - >presumably > > there > >> were tiny effects on the head positioning and so not >doing a > LLF > would > >> result in problems. > >> > > > > On Tue, 26 Sep 2017, Ali via cctalk wrote: > > > >> This was pretty common wisdom back in the day. Not quote sure how wise > it > >> was but it was generally recommended in the magazines of the time. I > >> remember reading an article about it in PC Magazine. > >> > > > > There were some interesting discussions of that when companies, such as > > Compaq, first started to put hard drives in portables! > > > > A lot of the conventional wisdom of the time has turned out to be not so > wise... > > Warner >
Re: formatting MFM drives on a IBM PC
On Tue, Sep 26, 2017 at 5:53 PM, Fred Cisin via cctalk < cctalk@classiccmp.org> wrote: > > I remember at least one manufacturer >recommending it for their > >> drive(s) if they were ever tilted through 90 degrees - >presumably > there >> were tiny effects on the head positioning and so not >doing a > LLF would >> result in problems. >> > > On Tue, 26 Sep 2017, Ali via cctalk wrote: > >> This was pretty common wisdom back in the day. Not quote sure how wise it >> was but it was generally recommended in the magazines of the time. I >> remember reading an article about it in PC Magazine. >> > > There were some interesting discussions of that when companies, such as > Compaq, first started to put hard drives in portables! > A lot of the conventional wisdom of the time has turned out to be not so wise... Warner
Re: formatting MFM drives on a IBM PC
> I remember at least one manufacturer >recommending it for their > drive(s) if they were ever tilted through 90 degrees - >presumably > there were tiny effects on the head positioning and so not >doing a > LLF would result in problems. On Tue, 26 Sep 2017, Ali via cctalk wrote: This was pretty common wisdom back in the day. Not quote sure how wise it was but it was generally recommended in the magazines of the time. I remember reading an article about it in PC Magazine. There were some interesting discussions of that when companies, such as Compaq, first started to put hard drives in portables!
Re: formatting MFM drives on a IBM PC
>I remember at least one manufacturer >recommending it for their drive(s) if >they were ever tilted through 90 degrees - >presumably there were tiny >effects on the head positioning and so not >doing a LLF would result in >problems. This was pretty common wisdom back in the day. Not quote sure how wise it was but it was generally recommended in the magazines of the time. I remember reading an article about it in PC Magazine. -Ali
Re: formatting MFM drives on a IBM PC
On 09/26/2017 01:19 PM, Ethan via cctalk wrote: I don't know if it's a good idea to low level format a drive or not. I remember at least one manufacturer recommending it for their drive(s) if they were ever tilted through 90 degrees - presumably there were tiny effects on the head positioning and so not doing a LLF would result in problems.
Re: formatting MFM drives on a IBM PC
On 09/26/2017 12:52 PM, Fred Cisin via cctalk wrote:
> Get SpeedStor (SSTOR.EXE) and use it.
> Once you do, you will never want to go back to the "Advanced
> Diagnostics" nor the BIOS routine.
That'll work. You can also scribble up your own formatter using the
BIOS calls--or check the SIMTEL20 archive for hard disk utilities. In
any case, it's not a big deal if you don't have ROM BIOS support for a
formatting utility.
> Actually one of the most common complaints on the early ST506/412 drives
> was noise from the spindle wiper! There was a little springy piece of
> copper? that rubbed on the end of the spindle. Over time, it would wear
> a divot, polish that, and start to squeal. Do NOT just rip it off!
> ("that fixed the noise"); push it sideways very slightly, so that it
> can start to wear a NEW divot.
That, and some drives used a solenoid operated spindle brake (just a
felt pad). Activating the solenoid released the brake. Occasionally
the thing would fail and the spindle motor couldn't overcome the drag.
And then there's rubber gone goopy, stiction, etc. Fun and joy--well,
not so much.
--Chuck
Re: formatting MFM drives on a IBM PC
g=c800:5
On Tue, 26 Sep 2017, Chuck Guzis via cctalk wrote:
Unfortunately, I agree with Fred. The stock IBM WD1003 controller ROM
did not have a format routine. Other models of the WD1003 did,
however--you'll see that the controller is a WD1003-something and that
may shed some light on its capabilities.
Get SpeedStor (SSTOR.EXE) and use it.
Once you do, you will never want to go back to the "Advanced Diagnostics"
nor the BIOS routine.
Periodically run SPINRITE, if you have it.
Pre-IDE drives had reliability issues.
Use it to lock out blocks that test bad, but do NOT let it "return to
service" blocks that pass, unless they are NOT on the manufacturer's list
and have been reFORMATted since the time that they had previously failed.
Actually one of the most common complaints on the early ST506/412 drives
was noise from the spindle wiper! There was a little springy piece of
copper? that rubbed on the end of the spindle. Over time, it would wear a
divot, polish that, and start to squeal. Do NOT just rip it off! ("that
fixed the noise"); push it sideways very slightly, so that it can start
to wear a NEW divot.
Re: formatting MFM drives on a IBM PC
One other trivial thought, . . . A FAST AT could handle 1:1 interleave. On many slower ones, an interleave could give dramatic improvement in throughput. Not having the optimum interleave will not interfere with usage, it is entirely a performance optimization. There were even programs that purported to determine the optimum interleave for you. Unfortunately, the application software that you use (dBase, Wordstar, Supercalc, etc.) does more processing of the data between sectors than only reading it. So the true optimum interleave was also dependent on what programs you used. Either Speedstor or SPinrite included capability of changing the interleave (by reading a track, reformatting the track with sectors in different sequence, and then writing the data back to the track.) Don't worry about the interleave until you get it to work. An old hot-rodding adage (that few abided by): "get it to RUN, before you try to get it to RUN FAST."
Re: formatting MFM drives on a IBM PC
On Tue, 2017-09-26 at 14:19 -0400, Ethan via cctalk wrote: > I don't know if it's a good idea to low level format a drive or not. Low-level formatting (which, at the time, was just called "formatting") used to be quite a routine operation on ST-506 MFM and RLL hard disks. They usually came completely blank from the factory and you had to format them according to whatever sector layout and interleave your particular controller wanted before they were usable. Once the drive was formatted you then had to run a separate process to lay out an actual filesystem. For MFM controllers on ISA cards I think the formatter was usually in the BIOS. For separate MFM controllers with a SCSI interface (Xebec S1410 kind of things) you used the FORMAT UNIT command. Newer ATA/SCSI drives with integrated electronics tended to come preformatted and there often wasn't any way to execute a low-level format even if you wanted to. At about the same time they started using embedded servo data on the disk itself for head positioning which made it impossible to do a low-level format in the field. I can't immediately think of any class of device on which attempting to execute a low-level format would be an actively bad idea (apart from destroying your data of course). On older ones it would work, on newer ones the drive would refuse the command, but in neither case is there likely to be any bad consequence. Was there a time in the middle when something bad would happen? p.
Re: formatting MFM drives on a IBM PC
On 09/26/2017 11:57 AM, geneb via cctalk wrote: > g=c800:5 Unfortunately, I agree with Fred. The stock IBM WD1003 controller ROM did not have a format routine. Other models of the WD1003 did, however--you'll see that the controller is a WD1003-something and that may shed some light on its capabilities. --Chuck
Re: formatting MFM drives on a IBM PC
Many hard drive controllers had a crude low level format program in their ROMs. With DEBUG, you could JMP to it, typically G=C800:5 although some were C800:0 or other offsets. U C800:0 to look at the code and find it. G=C800:5 or G_c800:800 for Western Digital controllers G=C800:CCC for Adaptec controllers. G=C800:5for DTC (Data Technonolgy) controllers G=C800:6for OMTI controllers according to https://kb.iu.edu/d/aaoa
Re: formatting MFM drives on a IBM PC
On Tue, 26 Sep 2017, Ethan via cctalk wrote: Hi, trying to check some MFM drives I have on my shelf. Have an IBM PC AT, with an WD1003 controller in it. So, what is the best(?) or easiest piece of software, to format the drives, check for bad blocks, etc.? I think I remember something like "ontrack" for doing it, but didn't touch PCs for a while ... Ideas? Links? Thanks! There were utilities like SpinWrite, and of course dos format. Some of the controller cards have a utility in rom that you can access via debug.com (I forget what the address is.) g=c800:5 g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind. http://www.diy-cockpits.org/coll - Go Collimated or Go Home. Some people collect things for a hobby. Geeks collect hobbies. ScarletDME - The red hot Data Management Environment A Multi-Value database for the masses, not the classes. http://scarlet.deltasoft.com - Get it _today_!
Re: formatting MFM drives on a IBM PC
On Tue, 26 Sep 2017, emanuel stiebler via cctalk wrote: trying to check some MFM drives I have on my shelf. Have an IBM PC AT, with an WD1003 controller in it. So, what is the best(?) or easiest piece of software, to format the drives, check for bad blocks, etc.? I think I remember something like "ontrack" for doing it, but didn't touch PCs for a while ... Ideas? Links? IBM did not supply a low-level FORMAT with the 5150, 5160, 5170. And there was no hard drive low-level FORMAT supplied in MS-DOS/PC-DOS. The PC-DOS/MS-ODS FORMAT.COM/FORMAT.EXE did NOT do a low-level format! A brand new drive came unformatted, and needed 1) low level format 2) partitioning (FDISK) 3) high level format (Directory creation, etc.) (FORMAT.COM/FORMAT.EXE) About the time of IDE drives, drives began to be shipped low-level formatted, and sometimes even partitioned and high level formatted. It was not rare for a drive to need to be RE-formatted. And, on 5160, low-level format was often incompatible between different controllers! There was no low level FORMAT in the OS, nor in the "DIAGNOSTICS" that came with the machine. There was an option in IBM's "ADVANCED DIAGNOSTICS", which was not supplied with the computer, and was not always readily available. Many hard drive controllers had a crude low level format program in their ROMs. With DEBUG, you could JMP to it, typically G=C800:5 although some were C800:0 or other offsets. U C800:0 to look at the code and find it. It would not over-ride the CMOS parameters, nor do much else. The best aftermarket software for formatting MFM drives on a 5170 was "SpeedStor" (SSTOR.EXE). It took a long time, but it was thorough and versatile. http://vetusware.com/download/SpeedStor%206.5/?id=9884 For testing already formatted drives, and periodic retesting, Steve Gibson's "SPINRITE" was best. https://www.grc.com/cs/prepurch.htm(not FREE, and there has never been a FREE version https://www.grc.com/sr/faq.htm ) BUT, some versions had a "feature" that wass on by defayult, and you had to turn off. Some versions of it, would test a block marked "BAD", and if it passed, would return it to service. That might sound OK, but it didn't know or care which ones were on the manufacturer's "BAD BLOCK" list. If the manufacturer of a drive says, "This block is defective, do not use it or trust it", then I do NOT want some program to say, "Well, it seemed to work when we tried it, so we know more about it than the manufacturer." Eventually, he got enough feedback about THAT, and later versions defaulted to NOT do that. There were several additional programs, that were sometimes needed, such as if you wanted to have a partition larger than 32MB on DOS 3.30 or earlier (MS-DOS 3.31 was first to accept larger partitions). And, you needed an overlay, such as ONTRACK, to use a drive larger than 504MB. Also, if you had a drive whose geometry was too incompatible with your computer - not all CMOS/SETUPs had a user-defined drive parameters option. -- Grumpy Ol' Fred ci...@xenosoft.com
Re: formatting MFM drives on a IBM PC
Maybe simply run the debug utility supplied with DOS and at the prompt enter this: G=C800:5 Normally, all the necessary tools to check and mark bad blocks are accessible by this way. However, you will have to encode manually the HDD specifications (heads, sectors, etc.) On 26/09/2017 20:08, emanuel stiebler via cctalk wrote: Hi, trying to check some MFM drives I have on my shelf. Have an IBM PC AT, with an WD1003 controller in it. So, what is the best(?) or easiest piece of software, to format the drives, check for bad blocks, etc.? I think I remember something like "ontrack" for doing it, but didn't touch PCs for a while ... Ideas? Links? Thanks!
Re: formatting MFM drives on a IBM PC
On Tue, Sep 26, 2017 at 2:08 PM, emanuel stiebler via cctalk wrote: > Hi, > > trying to check some MFM drives I have on my shelf. > Have an IBM PC AT, with an WD1003 controller in it. > > So, what is the best(?) or easiest piece of software, > to format the drives, check for bad blocks, etc.? > > I think I remember something like "ontrack" for doing it, > but didn't touch PCs for a while ... Hi, Emanuel, I remember OnTrack existing but I never used it. I don't think the 5170 PC/AT had a BIOS-based formatter, but you could toss the WD1003 into a newer machine (386 and up) and just use the BIOS menus to get to the formatter. With older 8-bit controllers, there was often a formatter in the controller's BIOS ROM, but I don't think the WD1003 has one. -ethan
Re: formatting MFM drives on a IBM PC
Hi, trying to check some MFM drives I have on my shelf. Have an IBM PC AT, with an WD1003 controller in it. So, what is the best(?) or easiest piece of software, to format the drives, check for bad blocks, etc.? I think I remember something like "ontrack" for doing it, but didn't touch PCs for a while ... Ideas? Links? Thanks! There were utilities like SpinWrite, and of course dos format. Some of the controller cards have a utility in rom that you can access via debug.com (I forget what the address is.) I don't know if it's a good idea to low level format a drive or not. -- : Ethan O'Toole
