On Thu, 26 May 2005, Paul Koning wrote: Yes. Many years (I believe about six). And disks are much more efficient than drums anyway. -- Sellam Ismail Vintage Computer Festival ------------------------------------------------------------------------------ International Man of Intrigue and Danger http://www.vintage.org [ Old computing resources for business || Buy/Sell/Trade Vintage Computers ] [ and academia at www.VintageTech.com || at http://marketplace.vintage.org ]

On Thu, 26 May 2005, Paul Koning wrote: I was thinking more in terms of space. A disk is much more efficient in this regard. You can stack N times the number of disks in the same space that a drum takes up. -- Sellam Ismail Vintage Computer Festival ------------------------------------------------------------------------------ International Man of Intrigue and Danger http://www.vintage.org [ Old computing resources for business || Buy/Sell/Trade Vintage Computers ] [ and academia at www.VintageTech.com || at http://marketplace.vintage.org ]

Tom wrote: I'd be very interested to hear of any working magnetic disk drive, commerical or otherwise, predating the IBM RAMAC. In hindsight it looks obvious. I'm not convinced that it was obvious 50 years ago. Eric

I found a website (pointer from an Electronic Design article on RAMAC): http://www.magneticdiskheritagecenter.org/ . Not a great site, but it has a bunch of neat information, including PDF files with several articles about the Model 305. One interesting detail from one of those articles: the 305 contains a drum, too. That's the program store and scratchpad. paul

On Fri, 27 May 2005, John Allain wrote: What? And take all the fun out of it?! Seriously -- past some trivial level of discussion this is absolutely called for. I don't trust computer-history books, with a few exceptions (eg. Randall, but he doesn't follow specific technologies). They're all filled with parrot talk. There are some very early references here, I do not have the refered-to docs unless stated, some really seem worth following up on. I won't summarize my summary. I'm in my lab and will see what I can find. NOTABLE: Booth's 1947+ paper magnetic disk Harvard Computation Laboratory, various Cohen, drum, 1950 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! "Digital computer components and circuits", RK Richards, 1957. In 'storage on a magnetic surface' p348, ref to IBM disk project (no name/model). Flying heads (bernoulli). Mentions a multi-disk system where non-referenced disks are stopped, to save energy of pumping all that air. Early EnergyStar anyone? Bibio data: DRUMS: "Magnetic drum digital storage system", AD Booth 1949 [see Booth below]. "Design, construction and performance of a large-scale general-purpose digital computer" BW Pollard, Proccedings AIEE-IRE Comp. Conf., Dec 1951,m pp62-70 published Feb 1952). "Desription of a magnetic drum calculator", Annals of Harvard Computation Lab., Vol 25, 318 pages, 1952. "Engineering description of the ElectroData digital computer", JC Alrich, IRE Transactions on Elec. Comp. Vol EC-4 pp 1-10, March 1955. "The magnetic storage drum on the Ace Pilot Model". DO Clayton et al, 1956. DISKS refs WJCC 1956 Noyes/Dickensen IBM [below] IBM 650 drum (AIEE-IRE-ACM conf, 1954). "Air floating, a new principle in magnetic recrding of information", GE Hagen, 'computers and automation', vol 2 #8 pp23-25, Nov 1953). "The notched-disc memory", J Rabinow, 'electrical engineering', vol 71, pp 745-749, Aug 1952. "magnetic memory device for business amchines", SJ Begun, 'electrical engineering', vol 74, pp 466-469, June 1955. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! MEDIUMS chapter discusses drums at length. Table 14-1 p338 and text coyly discusses a half dozen drum designs, but does not mention models/products; and the footnoted reference is to "STORAGE OF NUMBERS ON MAGNETIC TAPE" Coombs 1947 pamphlet, so I assume this is vaporware. The whole book actually reeks of vapor. A couple of tantalizing references p381: 33. Moore, BL, "magnetic and phosphor coated disks" in Proceedings of a symposium on large-scale digital computing machinery" Harvard Univ. Press, 1948. 37. SHepard, CB, "Transfer between internal and external memory", Annals of the computation lab. of Harvard Univ. Vol XVI pp267-273, 1948. The other 30+ refs apepar to be magnetic tape. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Proc. Western Joint Comp. Conf. 1956 "Magnetic recording head design", AS Hoagland (of UC Berkeley), head design for the IBM 305 magnetic disk file. Ref to authors 1954 AIEE Transactions paper "magnetic drum recording of digital data". "Engineering design of a magnetic-disk random-access memory", T. Noyes & WE Dickenson, IBM. 50-platter, 5,000,000 character storage. 24" platters. BIG F'ING MONSTER. 1.5 HP motor, 1200 rpm. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! "Preparation of programs for an electronic digital computer" Wilkes 1957 edition. I don't know if this is in the 51 edition, but EDSAC had a drum added sometime. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Automatic digital calculators, booth and booth, 2nd ed 1956 I don't have AD Booths earlier book. This one mentions his ARC calculator, drum and "disk" memory. Built 1947; completion date not given (see other book). The later APE(x)C machines (annoying variable name) are drum and electromechanical (pin) memory. ARC originally used Fe oxide coated paper, spun vertically (horizontal axle) to keep paper flat; the fixed "C" shaped head passed flux through the paper one side to the other (clearly, only on the very outer edge). Contact heads. Apparently this actually worked. page 133. Capacity must suck. Clearly this is the first magnetic "disk" type storage, though I won't call it a disk in modern terms, but it points out the utter futility of trying to name "firsts". refs "Magnetic drum storage for digital information processing systems", 'Mathematical Tables and other AIds to Computation' (journal), 1950. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Electronic Digital Computers, Franz ALt, 1958 Too new to be revealing, but does mention Rabinow's notched disk system, which is a toroidal "jukebox" for mag disk. Early drum ref: "Magnetic drum storage for digital information processing systems", 'Mathematical Tables and other AIds to Computation' (journal), 1950. !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! COMPUTER HANDBOOK, Huskey, 1962 Though it's very late, Huskey always covers previous art/history each chapter, so he's always a great author to follow. AS Hoagland writes about early mag storage; wire (1898), first mag data storage (1947), and drums. Talks of disks as "relatively new" (1962). RAMAC. Air bearing with explicit compressor (non-bernoulli).

On 27 May, 2005, at 10:29, Eric Smith wrote: The least obvious part of it was how to make a system where the head did not scrape away the coating. Poulsen had shown how to record to a fast-moving wire. His telegraphone had severe head wear. Agfa had shown how to use a paper tape coated with iron oxide to be able to work at lower head to tape speeds and get less head wear, and the tape backing was subsequently improved to acetate film. How to make this work when the magnetic coating was on a rigid surface was far from obvious. Magnetic drum assemblies were extremely tricky to adjust. There was a reason why they used fixed heads - the principle of moving along the drum axis had been known since Edison made the phonograph. The rason why drum storage used fixed heads was that nobody had figured out how to make flying heads yet, so they needed a very rigid assembly where you could use screws to make micrometer adjustments. -- -bv

Eric Smith wrote: I can imagine a couple of engineers in the 1950s hanging out at the diner for lunch and discussing "we need something with capabilities similar to drums but with a lot more storage capacity". One of the engineers gets up from his seat and puts a nickel in the jukebox, and watches the tone-arm move amongst the stack of disc records...

Tom> Rotating magnetic drum memories were certainly in use by 1948. Tom> Surely, other forms of rotation are obvious enough. And it's not Tom> like people weren't looking for storage then! Take a look at the technical papers in http://www.magneticdiskheritagecenter.org/MDHC/1956%20PUBLICATIONS/ -- and for that matter other material on that site. It's not well organized but it does have interesting content. Keep in mind that drums were head per track devices. A big part of the innovation here is the moving head actuator. Other aspects are questions like "can you make disks spin close enough to true?", "can you coat disks with iron oxide smoothly enough?" and other questions that were unanswered problems at the time. Remember, doing something the first time often requires genius. Doing something later often requires only basic craftmanship. Today, you and I could build a RAMAC clone in the garage workshop, with care and good workmanship. (Hey, that would be a neat project!) But that doesn't mean the original inventors deserve any less credit for serious innovation. paul

On Sat, 28 May 2005, Ulf Andersson wrote: The problem with histories is that they oversimplify. In this instance it's not too bad; RAMAC is a disk drive by modern definitions, so there's a reasonably bright line between it and similar things. But things like Booth's paper memory defies modern categorization; though rotating, magnetic, disk-like, it was a main-memory for a non-stored-program calculator. Since in 1947 there were no computers-as-we-know-them-today -- meaning an electronic machine with an "operating system" and a a data abstraction known as a "file" -- this disk/not-a-disk doesn't fit into "timelines" and Famous Person/Famous Product histories, so they get dropped. This problem exists in other areas of history and historiography, but in most areas of human culture -- clothing, food, language, housing, warfare, artwork, etc -- massive paradigm shifts don't occur every 15 minutes.

spare40tire wrote: I don't think it was known as a RAMAC. Maybe some people called it that *after* the introduction of the 305 and 350. IBM 1302, 87 MB, introduced September 1963 IBM 3330, 100 MB removable packs, introduced June 1970, expanded to 200 MB pack capacity in 1973 IBM 3340, first "Winchester" drive (sealed HDA), used 35 and 70 MB removable IBM 3348 "Data Modules", March 1973 DEC RP04 (manfactured by ISS/Sperry), 120 MB removable pack, mid 1970s Fujitsu Eagle M2351 14-inch hard drive, 380 MB, early 1980s Maxtor XT1065 67MB 5.25-inch hard drives, 1983.

Still have a partial copy of the handbook here for a GI (general instruments) 500FR 14 inch single platter HDD- 16 megaBITS capacity, platter spun up to 3600 rpm by a 'small' 1/2 HP motor. Dates on the drawings range from early 1971 to 1974. The platter is still here-makes a nice gong. The box this was in was std rack width and about 8-10 inches high-and nearly three feet front to rear! DaveB, NZ ----- Original Message ----- From: "Eric Smith" <eric at brouhaha.com> To: "General Discussion: On-Topic and Off-Topic Posts" <cctalk at classiccmp.org> Sent: Friday, May 27, 2005 9:01 AM Subject: Re: Remembering RAMAC -- No virus found in this outgoing message. Checked by AVG Anti-Virus. Version: 7.0.322 / Virus Database: 266.11.17 - Release Date: 25/05/2005

On Thursday 26 May 2005 18:29, Vintage Computer Festival wrote: The IBM 650 did have a RAMAC option. IIRC, Ampex Corporation had an IBM 650/w RAMAC in 1961 (I visited Ampex at that time and saw it running ;-) It WAS called a 650 RAMAC... See: http://www-03.ibm.com/ibm/history/exhibits/650/650_tr2.html Cheers, Lyle -- Lyle Bickley Bickley Consulting West Inc. Mountain View, CA http://bickleywest.com "Black holes are where God is dividing by zero"