On 3/7/23 20:34, Zane Healy via cctalk wrote: > On Mar 7, 2023, at 5:32 PM, Paul Koning via cctalk <cctalk@classiccmp.org> > wrote: >> >> I'm a bit puzzled by "6250 700'" because the reel size has no bearing on the >> format. 10 inch reels (1200 feet) were by far the most common but >> occasionallly the smaller 600 foot ones would be seen. and in rare cases >> (the infamous DEC TS05 comes to mind) 600' was all that they could handle. > > I’ll be the first to admit my question is a bit strange. Basically I’m > trying to use the date that various media types were first introduced to show > the oldest possible date for a bunch of media I’m trying to date. The > 9-Track tapes have been inventoried as “700 6250 BPI”, and I know they’re the > smaller reels. Doing some digging, it looks like 6250BPI tapes date back at > least as far as 1971 with the IBM 3400 series drives, I’d thought that 6250 > came about in the 80’s.
10.5" 2400' reels were the lingua franca since at least 1953. Companies bought them by the semi-trailer load. Quite often, just wrapped and palletized. Smaller reels (e.g. 5" and 7") were often used to economize (particularly when shipping; got a bunch of those). 2400' on a 1.5 mil mylar base was the most common, but you could find 3600' on 1.0 mil base. 1/2 tape adoption was typically a lot slower than the "bleeding edge" we're used to today. 7-track tapes were still common in the early 1980s, particularly if you were running older iron (CDC 6000/CYBER, UNIVAC 1100). Fortunately, the tape doesn't really care what you put on it, so long as the grain size of the oxide supported the recording density. I've handled tapes recorded at 556 that were labeled as 6250-certified. You have to appreciate the fact that tape was for many years, the preferred interchange medium between different computers. The IRS for years accepted tax returns prepared on 9 track tape, for example. I recall having to babysit the customer at McClellan AFB because the tapes written on CDC 669 drives weren't being read on the customer's'IBM 3420 drives. It turned out that the 669 when operating start-stop would "crowd" the first few frames of a block; the drive wasn't quite at speed when those were written. While the 669s could read them fine, the 3420s barfed. It eventually meant flying in one of the firmware techs to tweak things. "Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway." Andrew Tannenbaum, Computer Networks, 3rd ed., p. 83. (paraphrasing Dr. Warren Jackson, Director, University of Toronto Computing Services (UTCS) circa 1985) --Chuck
