Remember the ZIP Drive ?

[helius]

Jaz drives were also used in 5-spindle arrays for the video post production industry. Each drive was capable of 6MB/s transfer, and 5 of them together was 30 MB/s, fast enough to stream SD video with the hardware systems of the mid-1990s. The drives may have been specially modified to provide spindle synchronization as was often required for such systems. There's a fascinating history of "digital framestore" systems that has mostly vanished from the internet.

[tooki]

The other day, I came across a printed copy of a motion control program, for UT inspection tank, that I wrote in the early 1990s (guessing). The printout has a Zip disk attached to it. I doubt I could find a drive to read it.

Zip drives are plentiful on eBay, etc. Even factory-sealed ones aren't that hard to come by.

[james_s]

You could also send the disk to any of us who have Zip drives and get the contents copied to a more current media like a CD-R.

[tooki]

Or a zip file on dropbox.

[Rick Law]

Jaz drives were also used in 5-spindle arrays for the video post production industry. Each drive was capable of 6MB/s transfer, and 5 of them together was 30 MB/s, fast enough to stream SD video with the hardware systems of the mid-1990s. The drives may have been specially modified to provide spindle synchronization as was often required for such systems. There's a fascinating history of "digital framestore" systems that has mostly vanished from the internet.

And don't forget, there was the REV drives which was the next generation after JAZ - REV was 35gb at introduction, 70gb later models.

I sold my two REV drives and two dozen cartridges on eBay around 5 years ago.  If even for non-professional use I found that inadequate, I thought best to unload it to someone who can still use it before everyone jumped ship and I ended up with something worthless.

[CatalinaWOW]

ZIP drives really had no competition when they came out.  Reliable with enough room to transfer all the data most people had at the time.  Then as programs, pictures and movies got big they weren't nearly as roomy as they once seemed, and they had reliability problems for a while to nail the coffin shut.  I still have a USB version, just in case, but I also have just about every other storage mechanism I have used except for the 8 inch floppies.  Probably should have kept one of those too, but really can't think of 128 kbytes of data that is still important to me.

[mtdoc]

ZIP drives really had no competition when they came out.

True. IIRC, they had about a 6 month or so head start on the Syquest EZ-135 drives which were faster and had more storage. But that head start was all that was needed to make the Zip drive the de-facto standard for small, high capacity removable storage. PC innovations were rapid fire those days.  The Bernoulli Boxes had been around for a while but they were too big.

[james_s]

The Bernoulli drives were way too expensive for consumers too. I don't recall the exact pricing but the cartridges alone were on par with the cost per megabyte of hard drives. IIRC the drives were >$500 on top of that.

[Jwalling]

I remember another unit that was a regular floppy drive, plus it could use some special floppy up to 120 MB.

LS-120. Many are still in use as they were/are in a lot of the early Agilent Infiniium scopes.

[David Hess]

I always stayed away from the ZIP drive simply because I knew that any system which used an embedded magnetic servo on a soft magnetic media would be unreliable.  As soon as you demagnetize the disk, it becomes unusable.

[timb]

I remember another unit that was a regular floppy drive, plus it could use some special floppy up to 120 MB.

LS-120. Many are still in use as they were/are in a lot of the early Agilent Infiniium scopes.

Man, I loved my LS-120 drive! Much more reliable than a Zip Drive, used a standard IDE interface (so you could boot from them with no special drivers), the media wasn't terribly expensive and they supported regular floppies as well! You could also "overformat" regular floppies using the drive, so they held 2.88MB (IIRC).

It was a nice system, unfortunately not long after they started becoming popular CD-R drives and media prices started to come way, way down. Also, flash memory prices crashed and SD/USB storage started taking off. Both of these events conspired to kill LS-120. If only LS-120 would have come out at the same time as the Zip Drive, it really would have crushed it.

[NiHaoMike]

Jaz drives were also used in 5-spindle arrays for the video post production industry. Each drive was capable of 6MB/s transfer, and 5 of them together was 30 MB/s, fast enough to stream SD video with the hardware systems of the mid-1990s. The drives may have been specially modified to provide spindle synchronization as was often required for such systems. There's a fascinating history of "digital framestore" systems that has mostly vanished from the internet.

And don't forget, there was the REV drives which was the next generation after JAZ - REV was 35gb at introduction, 70gb later models.

I sold my two REV drives and two dozen cartridges on eBay around 5 years ago.  If even for non-professional use I found that inadequate, I thought best to unload it to someone who can still use it before everyone jumped ship and I ended up with something worthless.

Actually have one of those drives in my PC because the PC I salvaged the case from had one and I didn't have anything else to cover the hole that would be left if I removed it. Pretty amazing how the disks are about as big as two stacked 3.5" floppies and store more than a Blu-ray. (That is, until affordable >32GB Flash storage came along...)

[tooki]

I always stayed away from the ZIP drive simply because I knew that any system which used an embedded magnetic servo on a soft magnetic media would be unreliable.  As soon as you demagnetize the disk, it becomes unusable.

And has that come to pass?? As best I can tell, provided the disk wasn't put into a click-of-deathing drive that damaged it mechanically, Zip disks are doing just fine. (Assuming it was stored in a reasonable environment, like any soft magnetic medium.)

When is demagnetizing a disk a normal scenario?

[ebastler]

I still have it on my retro system.

I actually bought two ZIP drives just a couple of years ago -- one USB, one SCSI, as a way to transfer data to old Mac computers (Mac Plus and SE/30). Very convenient for data transfer, and if I remember correctly the Macs can even boot from the ZIP disks, so you can play with different operating system versions.

The ZIP drives are better than their reputation, in my opinion. I have not had one fail on me yet, and I could read the disks from my old parallel ZIP drive on the "new" ones I bought. The ZIPs are certainly holding up better than the 1980's 3.5" not-so-floppy disk drives in those Macs...

[tooki]

I just bid on a zip drive that has USB and SCSI, for that purpose. I have been working to get my ancient Macs up to speed.

Yeah, Macs can boot from Zip, no problem, as long as one of the following conditions is met:

- SCSI Zip drive on a Mac with onboard SCSI (or a bootable SCSI interface card; not all were)
- Internal ATAPI Zip drive on a Mac with onboard IDE
- FireWire Zip drive on a Mac with onboard FireWire (not via expansion card)
- USB Zip drive on a Mac with onboard USB and AGP, PCI-X, or PCIe expansion (the earliest USB Macs, with straight PCI expansion, didn't have firmware support for USB booting, ruling out models released before 1999)

In pure theory, you could boot a modern Mac from a USB or FireWire Zip drive, if only the capacity were distantly enough to hold the OS!


I used Zip extensively on my 1994 Performa 475. The Zip drive was actually fully as fast as the 160MB hard drive in that Mac. That was an external SCSI drive.

On my 1999 Power Mac G3, I chose the internal Zip option when ordering. For my 2000 PowerBook, I bought an internal Zip module from VST (which is currently the only Zip drive I have!!). That was a neat thing, a Zip drive that went into the drive bay in place of the DVD-ROM drive.

I always wanted an LS-120 drive, too. I haven't been able to confirm this, but I want to say that it was one of the only drives aside from the internal floppy drives on Macs that can read 800K Mac-format DD floppies. (Ordinary USB floppy drives cannot.) For those who don't know, 400K (single sided) and 800K (double sided) Mac floppies used a clever system of variable motor speed to increase the disk storage by using more sectors on the outer tracks than on inner ones, while maintaining the data rate, as old systems required. But since the entire rest of the personal computing world used "dumb" sector allocation, when HD floppies came out, Apple went with the pack for compatibility.

Anyone remember the 1.7MB Microsoft Distribution Format floppy format? MS used it on both Mac and PC, by doing low-level control of the floppy drive, dispensing with some aspect of the formatting that normally chews up space, so they could ship Windows and Office on fewer floppies. Became totally obsolete with the advent of CD-ROM of course.

[ebastler]

Thanks for the additional detail, tooki! You are obviously much deeper into ancient and semi-ancient Macs than me. I never had any of the old Macs when they were current -- far outside of my price range, and actually too "closed" for my taste. But the packaging and user interface was so cute...

When the original Mac was launched in 1984, Apple ran a promotion in Germany where you could borrow one for free for a few days, to give it a spin. (The local Apple dealer handed you a complete system in that nice carry bag, after having taken a copy of your ID. They did not check your bank account to see whether you stood any chance to buy a Mac, so even poor students qualified for the promotion. ) That's as close as I got to one of these, back in the day...

[helius]

Demagnetizing disks is a normal scenario whenever data destruction is involved. It was also required when re-using magnetic media on incompatible devices—for example, with single-density floppy disks written by double-density drives; since the head gap is narrower, it is unable to erase the full track width and the consequence is remanent magnetic fields in the sides of the track. Demagnetization is also a required step in calibrating tape drive heads, so that the magnetic developer will correctly show the track pattern and not remanent fields. For these reasons they were common equipment in repair workshops.

The factory servo tracks are an issue in many different media types, which cannot be used at all after being bulk erased. Some QIC formats, LTO, and any hard disk since the mid-1990s has the same issue.

[helius]

I always wanted an LS-120 drive, too. I haven't been able to confirm this, but I want to say that it was one of the only drives aside from the internal floppy drives on Macs that can read 800K Mac-format DD floppies.

No, they are not able to read variable-speed GCR floppies at all. Their support for regular floppies stops at MFM.
Adding to the confusion are that there were multiple models (SD-USB-M, SD-USB-M2, SD-USB-M3) labeled "For Macintosh" and yet without any 800K support. I think the changes were more to do with software compatibility.

[tooki]

Thanks for the additional detail, tooki! You are obviously much deeper into ancient and semi-ancient Macs than me. I never had any of the old Macs when they were current -- far outside of my price range, and actually too "closed" for my taste. But the packaging and user interface was so cute...

When the original Mac was launched in 1984, Apple ran a promotion in Germany where you could borrow one for free for a few days, to give it a spin. (The local Apple dealer handed you a complete system in that nice carry bag, after having taken a copy of your ID. They did not check your bank account to see whether you stood any chance to buy a Mac, so even poor students qualified for the promotion. ) That's as close as I got to one of these, back in the day...

[background]1993-2008 was when I was seriously deep in the Mac tech world. Back in the 90's, you could tell me any Mac model ever released and I could recite the specs, its quirks, etc...

From 1999-2006, I was a moderator and then admin of the forums at MacNN.com, at the time a leading Mac news site. At its peak, we had over 60K registered members, of which about 30K were active. To the best of our knowledge, the biggest Mac forum at the time until Apple's forums came along.

In 1998-99 I worked at Apple dealers, 2001-02 I worked as an onsite Mac consultant (oh, the joys of extension conflicts... back then I could tell you what every single of the around 200 extensions on a typical graphics pro system did, which ones didn't play nice together, and what order they had to load in to work properly), and 2006-08 I worked sales at Apple retail.[/background]

The closedness of the Mac was always wildly overstated by critics if you ask me. Other than being single-source for the computers themselves, the ecosystem was fairly open. There were commercial and shareware/freeware programs that tweaked damned near every aspect of the OS. Classic Mac OS wasn't locked down with code execution protection like it is now -- any extension could "trap"* (patch) the OS in any way it felt like, and it was widely done. Third-party hardware was actually quite plentiful (though if it was Mac-only, due to the small numbers it was made in, was often more expensive). The thing that made people think that stuff didn't exist for Mac was the simple fact that you couldn't find it at mass-market retail, as you could for PC.

As for weird knowledge of the old systems, I've started to worry that a) my own knowledge is fading, and b) it somehow needs to be documented for future computer archaeologists. From SCSI voodoo, to the insanely complex rules for populating RAM in some machines, to the display adapters, to the simple reminder that ADB is not hot-pluggable, to what all the extensions did, to how fonts worked, to the neato old " snd" resource format**, or that most 68K Macs won't boot if the backup battery is dead, or that the Mac Classic contains System 6.0.5 in ROM, with a key command to boot from it... etc etc etc.

To the chagrin of many, Apple took down its vintage support documents a few years ago. They're incompletely archived on archive.org, and some other sites have amassed a ton of it, but a lot has simply gone missing.





I think Apple did the try-before-you-buy thing in other places, too.



*People don't realize this, but on Macs until about 1997, a substantial portion of the core OS was in actual mask ROM. In the first Mac, it had 128K of RAM and 64KB of ROM, containing much of the OS, allowing things to be run directly from ROM instead of having to consume RAM. But because it's impossible to write bug-free software, the OS included a mechanism, called trapping, for the disk-loaded portion of the OS to "trap" a resource or instruction, to run a replacement version on disk instead of the one in ROM so that a buggy routine can be replaced with a fixed one. So unlike some OSes, you didn't have to rely on weird runtime code injection and the like to alter the software. The same trapping mechanism works for a system extension to trap a resource or instruction in the OS files themselves. This was widely used in the late 80s and 90s for third-party software to alter the OS in pretty much every way imaginable, from serious stuff like enhancing the memory manager, to pure fun like an extension that made the item being dragged by the mouse dangle with gravity and inertia.

**in addition to allowing PCM samples, it supported various forms of compression that were decoded in hardware, to FM sound synthesis instructions that were sorta compiled in software by the Sound Manager to run on the sound hardware, and the rendering of which varied between versions of Sound Manager. Those sounds -- like the "Simple Beep" system sound -- not only sounded different by Sound Manager version, but literally cannot be played back or converted by any known software on a modern OS. (The Simple Beep, for example, is fundamentally a sine wave, but with various filtering applied at different parts of the sound. Sound Manager 1 and 2 could only synthesize a stepped square wave approximation of a sine, which after filtering gave the Simple Beep a very distinct and IMHO pleasant sound. On Sound Manager 3, the same instructions are rendered using a true sine wave, giving it an entirely different, sharper sound.)

I always wanted an LS-120 drive, too. I haven't been able to confirm this, but I want to say that it was one of the only drives aside from the internal floppy drives on Macs that can read 800K Mac-format DD floppies.

No, they are not able to read variable-speed GCR floppies at all. Their support for regular floppies stops at MFM.
Adding to the confusion are that there were multiple models (SD-USB-M, SD-USB-M2, SD-USB-M3) labeled "For Macintosh" and yet without any 800K support. I think the changes were more to do with software compatibility.

OK, thanks. I wasn't sure about it. Too bad. AFAIK, the only way to get data off a 400K or 800K Mac floppy is with an actual Mac floppy drive.

[tooki]

Demagnetizing disks is a normal scenario whenever data destruction is involved. It was also required when re-using magnetic media on incompatible devices—for example, with single-density floppy disks written by double-density drives; since the head gap is narrower, it is unable to erase the full track width and the consequence is remanent magnetic fields in the sides of the track. Demagnetization is also a required step in calibrating tape drive heads, so that the magnetic developer will correctly show the track pattern and not remanent fields. For these reasons they were common equipment in repair workshops.

The factory servo tracks are an issue in many different media types, which cannot be used at all after being bulk erased. Some QIC formats, LTO, and any hard disk since the mid-1990s has the same issue.

Deliberate demagnetization isn't something a typical user would ever encounter; if you wanted to erase a disk you'd reformat it in software. High security organizations wouldn't trust demagnetization, so they'd shred the disk anyway. Ergo, for whom exactly is this a meaningful real-world limitation?

None of the other things you mentioned are even distantly relevant to Zip.

But anyway, my gripe was that you David Hess said specifically that the embedded servo would make Zip unreliable. And I see no shred of evidence that this turned out to be true.

[helius]

But anyway, my gripe is that you said specifically that the embedded servo would make Zip unreliable. And I see no shred of evidence that this turned out to be true.

Check your headers, that must have been someone else.

I don't think Iomega ever sold Zip as an archival format (the way MO drives were positioned), so its durability is really not relevant. If you store magnetic disks in a high-flux environment (like on top of your CRT) you will lose data anyway; the fact that the disk itself is also damaged is beside the point.

[tooki]

But anyway, my gripe is that you said specifically that the embedded servo would make Zip unreliable. And I see no shred of evidence that this turned out to be true.

Check your headers, that must have been someone else.

Oops, yep, sorry, wasn't you! I've edited my comment to correct this.

I don't think Iomega ever sold Zip as an archival format (the way MO drives were positioned), so its durability is really not relevant. If you store magnetic disks in a high-flux environment (like on top of your CRT) you will lose data anyway; the fact that the disk itself is also damaged is beside the point.

Well, I don't remember if they used the word "archival" specifically, but for sure they implied that long-term storage was a usage scenario. But unlike MO, which was marketed to pros, zip was mostly marketed to consumers.

That said, don't you think that the dangers of floppy (and similar) demagnetization from CRTs have been wildly overstated? Everyone used CRTs during the heyday of the floppy, and nothing happened. Heck, classic Macs had the floppy drive just inches from the CRT, and those were machines that booted from floppy, so those disks were near CRTs constantly. People had their floppy disk storage boxes next to the computer. Floppy disks sitting on top of CRTs was an everyday thing.

Similarly, I've never heard of a Zip disk being demagnetized by a CRT.

My suspicion is that many, perhaps most, cases of supposed CRT-caused disk failure were either spontaneous (disks weren't perfect, especially towards the end of the format when most media was crap), or caused by something else, like humidity and heat.

[P90]

speaking of floppies, I still have my original Macintosh SE/30 with travel bag, wonder if it's worth anything...   lol

[helius]

For sure there were magnetic fields from the CRT yoke, but it's really the flyback that has the potential to corrupt magnetic media. That would be on the left side of a Mac, and books from the era do warn about putting floppy disks near there.

[David Hess]

I always stayed away from the ZIP drive simply because I knew that any system which used an embedded magnetic servo on a soft magnetic media would be unreliable.  As soon as you demagnetize the disk, it becomes unusable.

And has that come to pass?? As best I can tell, provided the disk wasn't put into a click-of-deathing drive that damaged it mechanically, Zip disks are doing just fine. (Assuming it was stored in a reasonable environment, like any soft magnetic medium.)

When is demagnetizing a disk a normal scenario?

The problem was that if it happened, the disk became unusable as there was no way to rewrite the servo tracks.  With damaged servo tracks, the disk will be stuck in click of death mode.  That this could damage the drive and drives could damage the disks was a separate problem with the hardware but are they really separate if they both came from poor engineering?

At the time my opinion was that relying on an embedded servo track on soft media would lead to poor reliably and was in indication of poor engineering.  The floptical designs like the LS-120 and MO drives avoided this problem entirely by using optical tracking.  Did the LS-120 designers do that just to avoid patents?  The rigid disk designs like from Syquest were "harder" so unintentional damage to the servo track would be less likely and I never heard about it happening with them.  If it did, at least the drives did not destroy themselves.

So my memory of the ZIP drive and Iomega's products in general was other people complaining.  Based on the lawsuit and Iomega's own statements, (1) I think my early assessment leading me to avoid them was accurate.

The factory servo tracks are an issue in many different media types, which cannot be used at all after being bulk erased. Some QIC formats, LTO, and any hard disk since the mid-1990s has the same issue.

Voice coil actuators all suffer from this if a magnetic servo track is used whether it is embedded or not; early drives had a separate servo track (2) but the result was the same and this eventually became untenable because the track density became high enough that alignment between the servo surface and data surfaces could not be maintained after formatting.

Didn't tape drives have a method to "step" the head so they could align on the servo of already written tracks while writing the new tracks including the servo?  I do not remember any tapes becoming unusable after being degaussed but I rarely dealt with high end tape storage.  Colorado Memory Systems (4mm and 8mm DDS) cured me of any enthusiasm for tape backup although Summit Memory Systems made a reliable QIC drive. (3)

(1) Iomega said only 1 in 200 people had click of death failures?  Well, I guess it is not a problem then.

(2) So you ended up with drives which had for instance 8 or 16 surfaces but only 7 or 15 were available for data.  The other surface held the servo tracks and was written at the factory.

(3) I really liked the Summit Memory Systems QIC drives because of their performance and reliability but they seemed to be the exception.