Seymour J. Metz asks: > I need a reference for Wikipedia as to whether the cells on a > 3380 only relate to error checking or whether they actually > reflect the physical layout of the track.
I've never seen this documented. But I never looked that deeply, either, so it might have been in my face since 1981. Regardless, I was told it was only for purposes of allocating space on the actual track -- AS IF the device actually wrote 32-byte (the cell size) physical blocks (or multiples thereof). At the time, prior to PCs, this meant nothing special to me. Of course fixed sector sizes for PC drives made more sense, and I assumed the underlying 3380 and 3375 hardware, like the 3370, used a fixed block [or sector] size, which obviously had to a multiple of 32. Later, I was told (by IBM) that this was, in fact, the case. The 3310 and 3370 were FBA above the covers. The 3375 was, as has been acknowledged, FBA under the covers. The 3380 has been claimed to be genuine CKD under the covers, but I was informed authoritatively (supposedly, or I felt so at the time) that it was also FBA under the covers. FBA was an acronym developed for public consumption/marketing purposes. Internally, they were called "cellular" [DASD] devices, and such terminology used to appear in MVS/SP and DF/DS and DF/EF source code and macros. It also appeared in APAR and PTF text. Sadly, since all such material remains licensed and proprietary, so none of it ever made it into the public domain, and since the WWW effectively started (for IBM) no earlier than 1995, none of this sort of stuff appears to be anywhere online where Google can discover it. Nonetheless, despite (easily Google-able) claims that the 3380 was CKD under the covers, I was told (at the time, circa 1981) by presumably authoritative sources that it was, in fact, an FBA (or cellular) device. One had to use the same track balance calculation algorithm for 3380s as one did for the 3375, and the cell size was recorded as 32 bytes. If the 3380 was a CKD device under the sheets, then San Jose went to an awful lot of trouble to make it look like an FBA device below the spread so that it would be compatible with a genuine below the covers FBA [cellular] device. Other than hearsay, I don't have any authoritative reference, so I don't have any (currently-online) material to offer you. Immediately after I was told (by the late Ed Daray and Jack Gelb) how 3380s worked (after they felt they could actually "tell" me), I never gave it any further thought, unfortunately. Today I feel that I _should_ have thought to be more curious and dug deeper. But that did not happen. <sigh> So I'm as interested as you are. Perhaps Lynn Wheeler has some old (actually physically archived) reference material that can shed light on this subject. -- WB ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
