On Tue, Aug 26, 2014 at 2:37 AM, Pontus Pihlgren <pontus at update.uu.se> wrote: I wouldn't think so... ISTR the RP02/RP03 were a Memorex drive, and probably the RP04 and RP06 but I'm not as sure... Google tells me: "The RP02 is a Memorex 660, IBM 2314 (drive) / 2316 (pack) - compatible, the RP03 uses the same pack (????), the RP04 is IBM 3330 compatible." (the RP07 was not made by Memorex AFAIK) -ethan

On Aug 26, 2014, at 3:22 AM, Ethan Dicks <ethan.dicks at gmail.com> wrote: Correct. RP03 is physically very different from the RP04/5/6 packs. The recorded area is much wider. The IBM 2314 was mentioned as comparable. You?re going to need an RP03, or a spin table. RP04 and RP05 are in fact compatible; the same pack can be read and written on both. The RP06 pack is mechanically the same, but the layout is different and it can only be read/written on an RP06. By the way, when drives are described as ?equivalent to the IBM 3330?, chances are that means mechanically identical. The format is different, and you?re quite unlikely to be able to read a pack for the one on the other drive. (Well, not unless the drive doing the reading is very forgiving about foreign formats, which I don?t believe is the normal practice.) IBM used variable length sectors and, if requested, keyed sectors, on the IBM 360 and 370 series systems. DEC used 512 or similar fixed length sectors. CDC used this kind of drives too: a CDC 844-21 is the same thing as an RP04, pretty much, but it uses sectors of 322 12-bit words. paul

On Aug 26, 2014, at 10:27 AM, Al Kossow <aek at bitsavers.org> wrote: I wonder how hard it would be to create a spin table from scratch (given compatible heads). The critical parts would be a high quality spindle, pack mounting adapters for that spindle, and a precision x/y positioner for the head. That would do the job if the tracks are wide enough that you can get away with fixed radial positioning. If you need to follow a servo track in real time during rotation, it gets harder. The fixed positioning would certainly apply to RP03, RK05, 1311, and the like. paul

My string of 3330s is really ugly. Really, really ugly. At best, all I can do is have a cosmetic restoration done. Yes, hopefully - but somehow I doubt it. It is too far for me, and that is saying something. I have a Calcomp D114, which is perhaps compatible with the RP03? I pulled it from the Black Hole a couple of years back. Does the RP03 use that odd looking IBM-raw-drive-control-thing plug, as they used with the 2311 and 2314 (and the plug compatibles, as well)? -- Will

AFAIK RP01 = Memorex 630-1 uses 1316 disk pack RP02 = Memorex 660-1 uses 2316 type disk pack (sector slots) RP03= ISS 715? uses double density 2616 type disk pack RP04= ISS 730? uses 3336-1 disk pack RP05= Memorex 670 uses 3336-1 disk pack RP06= Memorex 675 uses 3336-11 disk pack (best drive ever according to Grant Saviers) RP07=??? I think it was an ISS fixed disk drive not unlike the IBM 3350 (worst drive ever) Since the RP03 used an external transducer for positioning it should be possible to read it on a spin stand with a relatively narrow head and possibly a means for adjusting the center to minimize runout. Extracting the information from the bit stream should also be possible given the fixed sector size. Altogether a relatively straight forward process if u have a spin stand and the means to change the mount to match the pack hub. Tom -----Original Message----- From: Paul Koning [mailto:paulkoning at comcast.net] Sent: Tuesday, August 26, 2014 7:06 AM To: General Discussion: On-Topic and Off-Topic Posts Subject: Re: RP03 disk pack imaging On Aug 26, 2014, at 3:22 AM, Ethan Dicks <ethan.dicks at gmail.com> wrote: wrote: wrote: Correct. RP03 is physically very different from the RP04/5/6 packs. The recorded area is much wider. The IBM 2314 was mentioned as comparable. You're going to need an RP03, or a spin table. else? RP04 and RP05 are in fact compatible; the same pack can be read and written on both. The RP06 pack is mechanically the same, but the layout is different and it can only be read/written on an RP06. By the way, when drives are described as "equivalent to the IBM 3330", chances are that means mechanically identical. The format is different, and you're quite unlikely to be able to read a pack for the one on the other drive. (Well, not unless the drive doing the reading is very forgiving about foreign formats, which I don't believe is the normal practice.) IBM used variable length sectors and, if requested, keyed sectors, on the IBM 360 and 370 series systems. DEC used 512 or similar fixed length sectors. CDC used this kind of drives too: a CDC 844-21 is the same thing as an RP04, pretty much, but it uses sectors of 322 12-bit words. paul

On 8/28/14 2:13 AM, Philipp Hachtmann wrote: I supported RP02/3 and RP11 where I was working in 1980. They were a pain in the butt to keep going then, I don't think 35 more years have made them any more reliable. My desk was 3 feet from the backs of the RPs. At least the RP03 didn't have a mechanical latch that the RP02 had that set every time the heads moved (click, click, click click click click click) The good news is the head technology is pretty primitive, they fly fairly high (compared to 3330) so there is a chance that the mech and pack can be inspected and cleaned so that it doesn't crash the second you try to load the heads. It is roughly equivalent to RK05 tech in the DEC world. Mechanical restoration and corrosion/rust is just going to be time consuming to deal with. That is one of the things I've been dreading in trying to get the Century Data T-80 or T-300 working. These are roughly the same age as CDC first generation 80 and 300 mb SMD drives. There are a LOT of M-series modules in an RP-11. I remember the data separator (and all the other twiddly pot stuff) to be pretty touchy.

Interesting you should mention the RK05, the 11/34 with the RP03 drives has both an RK05 F and J. I'm fairly confident we can get the RK drives working, and grab images of the / and /usr file systems, which is a great start on getting this version of unix back. The source code for the OS is on an RP03 pack, and I'm not sure I will have access to the hardware long enough to try and get the controller and a drive working. It's interesting to note that this system actually has a second RP11 controller, condition unknown, not connected (and apparently never powered up since it was bought at auction 17 years ago). The one connected to the drives has a controller fault of some kind at power up, no specific info available yet.