Did it have hydraulic positioners like the old CDC 3300 drives? On Wed, Jun 10, 2026 at 6:29 PM Paul Koning via cctalk <cctalk(a)classiccmp.org> wrote:

Well, I said that it CAN have higher density, not quaranteeing that it would :-) The original 23FD was 80K TLDR: long (too long) ramble about disk capacities: 8" SSSD (77 tracks * 26 SPT * 128 bytes poer sector) was 250K. 8" DSDD (77 tracks * 26 SPT * 256? bytes per sector) was about 1000 K 5.25" 500K data transfrer rate ("high density") was 2 sided 80 tracks per side 15 secors per track of 512 bytes each),was 1200K, or 1.17M, NOT 1.2M unless you define meg as 2**10 * 10**3. instead of an honest 2**20 (1048576) The earliest 5.25" (35 track SA400) had a capacity of 87.5K when formatted with FM, ten sectors of 256 bytes each. (TRS80 and early Osborn). other choices of size of sectors and number of sectors will vary that a little. But VERY soon they lengthened the slot for 40 tracks (100K) GCR (Apple) gave an slmost 50% increase But, FM double sided, then double density (MFM), and 96tpi for 80 cylinders (80 tracks per side) MFM doubled the capacity (Higher density / more bits per track) Going to 96tpi gave twice the total capacity, at 80 tracks per side, but each track was the same as a 48tpi track. Doubling the data transfer speed, was still MFM ("double density), but the changed speed enabled almost twice as many bits per track Yes, I know that later, there were barrium ferrite disks (E[xtended] D[ensity]) even some 192 track per inch, andsome bizarre ways to get a lot of data onto a disk. Changing a single sided format to double sided doubled the capacity, but changing capacity is not the same as changing density. One company (Intertec superbrain) insisted on calling the 340K double sided version of its 170K single sided, "Double Density"! Then when they went to 340K double sided, double density, they called that "quad density"; everybody else in the world called it double sided double density, and used "quad density" to refer to 640K - 800k 96tpi double sided double density. But then, when Superbrain went to 780K (which almost everybody else was calling "quad density", Superbrain had already used up that name! So, they called their 780K format, "SUPER Density"! Then 3.5" The Epson PF10 had 40 cylinders with 67.5 traxks per inch, but everybody else immediately went to 135 tpi, giving it essentially the same geometry visible to the computer as 5.25" double sided 96tpi. As opposed to ST225, with 4 heads for one platter. :-) . . . or the Lisa Twiggy. :-) They claimed that the second thumb slot was for the other heads, not just to make sure that all media would have plenty of thumbprints. Yes, switching to "side B" of a double sided disk was obviously much quicker than stepping to another track. -- Grumpy Ol' Fred cisin(a)xenosoft.com

"RAMAC" is not the name of a disk drive! RAMAC refers to a complete computer system that implements the "Random Access Method of Accounting and Control", such as the IBM 305 RAMAC of which the IBM 350 disk storage unit is a *component*. So the name you're looking for is "IBM 350" which was the name of the first disk drive.

On Wed, Jun 10, 2026 at 10:44 PM Gavin Scott via cctalk <cctalk(a)classiccmp.org> wrote: Back at one of my early jobs in the 80s, the owner told me a story from when his dad worked at Columbus Coated Fabric (before they were bought by Borden Chemical, the makers of Elmer's Glue). As the story goes, CCF was either the first or second big company in Columbus to buy a computer, an IBM 305 RAMAC, specifically for payroll. It was then that I learned about one's-complement machines and 'negative zero'. -ethan

My understanding of it is, it makes the logic simpler. But the cost is software complexity. ok bear.

On Thu, 2026-06-11 at 11:29 -0400, Paul Koning via cctalk wrote: Univac 1108 "test zero" would skip on +0 or -0. "Jump if zero" wolud jump if +0 or -0. It also had a field in instructions  to select a part of a word, so if the word is a number and you want a part of it, say one sixth, you need to be careful. The ALU didn't produce negative zero, at least not for addition and subtraction, but the Fortran compiler had the subtract backward for some reason. A = B - C was "load C, subtract B, store negative A," so if B==C you ended up with A = -0. It was common to see a button or sticker "Down With Negative Zero."

On Thu, Jun 11, 2026 at 3:31 PM Tom Gardner via cctalk <cctalk(a)classiccmp.org> wrote: I think I have one or two Sony OA-D3XV drives, I'd have to go up in the attic to look for a never-released terminal server product I worked on in 1984-1986 to find out exactly which model. I remember it had a chunky full-height Sony 3.5" drive but I don't remember if it was so old we had to pinch the shutter on the diskettes or if it had an autoshutter mechanism. The first 3.5" drives I used were on a 128K Mac, around July/August 1984 and I definitely wasn't pinching the shutter on those. -ethan

On Fri, Jun 12, 2026 at 11:42 AM Carey Schug via cctalk < cctalk(a)classiccmp.org&gt; wrote: The model 1 was sold for a number of years as a cassette storage ayatem before people started getting disk drives for them. Tandy, when they started selling a disk drive was mist concerned with profit and reliability, compared with cassette. As disk drives and disk doublers brought a new life to the model 1 after 1980, all kinds of OS's popped up that har a denser allocation scheme than the original trs DOS. It was a unique case, not many other home PC model had so many OS's to choose from. Bill

https://www.bitsavers.org/pdf/ibm/305_ramac/22-6264-1_305_RAMAC_Manual_of_O… has the answer: 50 surfaces, 100 tracks per surface, 10 sectors per track, 100 characters per sector. Multiples of 10 show up in later products too, like the 1311 disk packs for the IBM 1620. Then of course there's weird stuff like the IBM/360 disks with their variable length sectors. Or the CDC mainframes with 322 12-bit words per sector. Or Electrologica where you could choose one of five sector sizes on a per-track basis. Or the DEC RS64 with 64 byte sectors, or the word-addressable RF11. paul

I am pretty sure it was 100 six-bit characters plus a parity bit and a stop bit, so the recorded character was indeed 1 byte but to say it was 100 bytes that would be like adding the formatting, ECC and header to the 512 or 4 KiB industry standard sector. Many do, I don’t 😊 Tom -----Original Message----- From: Fred Cisin &lt;cisin(a)xenosoft.com&gt; Sent: Thursday, June 11, 2026 1:35 PM To: General Discussion: On-Topic and Off-Topic Posts &lt;cctalk(a)classiccmp.org&gt; Subject: [cctalk] Re: Some small corrections RE: Ramac : The real first disk drive? and RE: Floppy disk On Thu, 11 Jun 2026, Tom Gardner via cctalk wrote: Thank you, I couldn't remember how many bytes per track to calculate accurately. They might not have been multiples of 128? Long, long ago, I was told 100 bytes per sector?