Floppy Drive Interface Tinkering

unless you count oddball stuff like LS120

Bye. Jasen

That's misleading, since you're comparing the non-Apple 5.25" disks with the Apple 3.5" disks. We are talking about Apple's 5.25" disks, as seen on the Apple II which came about years before the Macintosh arrived. Michael is right, those Apple floppies had pretty lousy capacity.

But, your statement is also misleading, because that 400KB disk was a 3.5" floppy, and when other computers made use of them, they fitted

The only time Apple had comparable floppy disk capacity was when they went for compatibility with the rest of the world, and and dropped their out of the mainstream scheme.

I should reinforce the point that higher integration had nothing to do with leaping forward to 400KB floppy capacity. It was the same basic arrangement used on the original Apple II floppy system, just integrated into a single package, and the introduction of 3.5" floppy drives made that big leap forward.

Michael

How about LS-120 type drives? They are IDE and read/write 1.44 Mb floppies.

Bernoulli type high density flexible media drives started out as expensive SCSI equipment. The later items like LS120 also turned out to be niche markets, as did the Sygate 1.5 GB cartridge drives.

They may read them, but i doubt it; physical media encapsulation issues. Do you remember "stringy floppies"?

What in the world does "stringy floppies" have to do with it? They were tape drives, with the tapes about the level of a microcassette tape, intended to be an improvement over using cassette recorders to save programs, but not nearly as expensive as floppy drives.

The LS-120s were a hybrid between a floppy and an optical drive, which is why they were commonly called floptical drives. They used a laser to get more precise placement of the read/write head, so the density could increase greatly, while leaving the disks the same size as 3.5" floppies.

And yes, they could read and write 1.44meg floppies, which was the point. You could replace an existing floppy drive with one, giving you far greater storage capacity, but not lose the ability to deal with floppy disks.

Michael

AKA "Exatron"? I had a couple, and some of the micro tapes that they used.

Do you remember the IOMEGA 20 MB "Floptical" drive? I have a new one, still sealed in the original retail display box.

Avoiding the entire building of a floppy controller aspect, I've decided to just use an ISA multi I/O card in my project. Though I figured I'd better learn how to operate it via its registers first in a PC before I go and try to make one work standalone. Since I don't have any systems with an ISA slot handy though, I'm testing code on my laptop. And I've run into a problem.

I've opted to not use DMA if at all possible, since that's just more components. So after much effort, I managed to get my laptop to properly initialize its floppy controller and complete a read operation of the first sector. The problem, however, is that even though I've read an entire 512 bytes, it's still resulting in an error code from status register 0, which translates to: "command execution started but terminated abnormally". Status 1 has bit 7 set, which translates to "end of cylinder; sector# greater than sectors/track".

Short of posting a mess of code here, I thought someone might know what I'm missing from prior experience. For the record, my IRQ 6 handler acknowledges the IRQ to the interrupt controller and increments an IRQ count variable (which I use to track when an interrupt hits), though once I go into the read loop, I pretty much ignore the interrupt and poll via the status register bits (though my handler still takes care of resetting the interrupt controller of course). I'm thinking I can probably ignore the IRQ altogether whenever I use an ISA card stand-alone by just polling for everything. I can't seem to ever get reliable interrupts to happen anyway. One time I test my program I'll get'em, the next time I won't in some places, yet the program generally works either way (since my wait_for_irq routine is set to timeout after so long if it detects none).

Anyhoo, the 512 bytes it reads are the actual intact data that they ought to be, so there's no problems getting it from the disk. After it's done reading, I'm reading out the seven result bytes like I should. So I'm not understanding why I'm getting an error when I have the data I requested. Is there something that needs to be done to complete the read or something?

One thing that confuses me is when I'm sending the End of Track value in the read command. The way I understood it is that this meant to put the last sector on the track, which in my case would be 18, though I noticed that when I put 18, it reads off all 18 sectors. So I ended up using 1 for this, which I can only assume is correct.

Anyway if posting code is necessary I can, I just didn't want to clutter up this post anymore than my ramblings already have. ;) Thanks!

How about the laptops that let you use a special cable to plug a floppy drive into the printer port?

I actually thought about finding a mostly external drive already, but a.) it defeats part of the fun aspect of making something myself, and b.) without the proper laptop to test it on, I might not know for surely if it's working properly. My own laptop is a Thinkpad 600e, which can use its floppy drive externally through the special port on the side, but I generally just swap out the dvd-rom for the floppy when doing my testing, so it's plugged in directly. I have no idea what the signals/protocol for dealing with it in its external caddy would be, though it might be interesting to find out, especially since I'm sure it doesn't require 12v to run its motors like a normal one does.

Michael,

The one time I looked, those floppy drives are actually electrically identical to a standard 26-pin "normal" floppy drive (they just omit a ground pin or something -- I don't recall). ...and presumably one of the "parallel port mode" pins is used to detect that it's a floppy rather than a print, and mux all the signals over to the floppy drive controller rather than the printer latch.

---Joel

In my old Toshiba T2100 laptop the LPT connector goes via a multiplexer either to the printer interface or to the uPD765 floppy controller. The function is selected by a toggle switch. I don't think that any laptop uses an ordinary standard printer port for this purpose.

Piotr

Same here, although I wouldn't be surprised if such a thing existed. I've seen both parallel to SCSI converters and SCSI 3.5" floppy drives, so strictly as an exercise in nerdiness it probably could be done.

And of course there used to be many parallel port to CD ROM drive adapters out there...

I've still got a couple that were donated, without the driver software. Of course, they only worked with win 95, so they are just collector's items these days. I also have a couple SCSI external CDR drives, but I wasn't able to find the driver software for the oddball SCSI interface cards that came with them. I just picked up a HP sheet fed scanner with a SCSI interface that I'd like to get up and running to scan old manuals with.

The LS120 media was the same shape as regular floppies and the drives would read ordinary floppies at an accelerated rate. I never tried to write a floppy in one.

wafadrive, microdrive?

Bye. Jasen

Exatron "Stringy floppie drive"

Magnetic tape, a la cassettes.

Cheers! Rich

jasen wrote in news:ebmof4$s4r$ snipped-for-privacy@gonzo.homenet:

The magnetic media is identical between the "standard" floppy and the "super floppy" or LS-120. The difference is that the super floppies have concentric rings of "pits" burned into them by a laser during the manufacturing process. This allows for optical tracking and closer spacing of tracks. It's an adaptation of CD technology. LS-120 IDE drives read and write to "normal" floppies at a much higher rate because 1) the heads are designed to be able to do this, and 2) the interface allows higher data transfer rates.