Hi Richard, Sorry I missed you earlier I'll send you some pictures of what I have so you can decide if it's worth your drive. About the 6250 / 4132 OS. I've been talking to Rick Bensene down in Portland. He had a bootable UTek tape at one point, but it seems to have gone bad. I don't know if anyone else has either a drive that could be imaged, or a boot tape. Sadly, that piece of history may be gone. Jim Richard wrote:

On 5/10/12 9:46 PM, Dave McGuire wrote: I have to get one 8560 working to recover some data from an 8550, after that it all can go. I have a couple more 8560's as well. There is only one complete 8540 and a bunch of parts. Unfortunately, Will was going to Black Hole on the way back, so there was no space in the van for anything going east. There was an 8560 service manual in the pile, so that will get scanned tomorrow. It looks like the floppies are V7 file systems.

On 5/10/12 11:15 PM, William Donzelli wrote: that was a good find. stuff ain't made no more SDS made a tape cartridge system. CHM has one in the SDS 930 from NOAA.

On 5/11/12 1:30 AM, Simon Owen wrote: They are ds/dd 26 sector 256 bytes/sector non-interleaved, except for t0/side 0 which is sd 128. Does remind me, has anyone noticed that imagedisk incorrectly reports the track number when the first track differs from the rest? When it switches to double-density, it says it is at track 4/0. I was mistaken about them being V7. A few of them are, but most are in Tek 'fbr' format, which I don't know anything about. They appear to have tables of contents at the front, with full path names. The data appears to be contiguous, so it should be fairly easy to figure out the format.

On 5/11/12 12:26 PM, Jochen Kunz wrote: The one problem you might have is the first track is single density.

On 12 May 2012 at 9:30, Jochen Kunz wrote: Some can, some can't. Some can only read single density (FM); some can neither read nor write FM. It all depends on the chip (or combination thereof) used. Oddly, later systems with SMSC "SuperIO" chips have better luck with this. I remember complaining to Intel that the 82077AA-1 stepping completely broke FM writing (and some reading). They said that FM wasn't important enough to deal with then--or in the future. If you have an old PC system with a National Semi 8473 or 8473, you've got the best of the lot. WD37C65 controllers are also okay. Hope this helps, --Chuck

On Sat, 12 May 2012, Jochen Kunz wrote: The original 5150 FDC board was hard-wired for MFM ("double density"). It could not do single density, even if you gave it the single density commands. It COULD be modified. (Flagstaff Engineering did!) SOME later FDC chips did the MFM/FM switch entirely in software. SOME of those can do single density. David Dunfield studied the issues, and even wrote some test software to check compatability with FM and MFM 128 byte sectors (another NEC style chip issue) Well, THAT would certainly be enough to prevent any access using a WD or NEC style FDC chip! Probably. The 1.2M floppy was developed as an 8" replacement. In fact, the first one that I got (Mitsubishi) had a 50 pin connector on it. You would have some struggles working out jumpering, and "minor" details, such as head load, etc. People tend to make their "last hope" be something that they know even less about, and have little chance of success if their other approaches failed. And yet, those same systems can function beutifully in the hands of somebody who is familiar with them and knows what is going on.

On 5/11/12 12:41 AM, Jochen Kunz wrote: I only have volume 1 of the user's manual. I'm guessing there must have been other volumes. I can see there is a man(5) page for fbr on the first disk of the system restore which describes the format. 8560 MUSDU System Reference Manual FBR(5) ___________________________________________________________________________ FBR(5) NAME fbr - file backup and restore format SYNTAX #include <sys/types.h> #include <fbr.h> DESCRIPTION Fbr saves and restores directories and files on a floppy disk archive, preserving aliases (multiple links to the same file). The archive consists of: Boot block - Block zero is reserved for a copy of a stand-alone boot program. Directory area - This area contains directory information for all files and directories on the archive. The first entry in this area is dedicated to the archive label. Data area - The data of all files on the archive is stored here. The format of a directory entry (as given in the include file) fol- lows. To increase the portability of archives, this structure is read and written in PDP-11 format regardless of what machine fbr_is running on. #define PATHLEN 106 /* space for path */ struct fbrent { char fbr_path[PATHLEN]; /* pathname */ unsigned short fbr_mode; short fbr_uid; /* owner's userI^AD */ short fbr_gid; /* owner's groupID */ off_t fbr_size; /* size in bytes */ time_t fbr_acct; /* access date-time */ time_t fbr_modt; /* modify date-time */ unsigned fbr_fblk; /* firs^At data block */ char fbr_zero; /* unused */ char fbr_chks; /* checksum */ }; The path field is the pathname of the file when archived, less any redundant slashes or initial './'. It is null-terminated if less than PATHLEN bytes long. The mode, uid, gid, size, and access and modifi- cation date-times are in the same format as their corresponding i-node fields. As in the file system, an available entry has a mode of zero. The fblk field contains the block number of the first data block allo- cated to this file. The zero field is unused. The checksum field contains a number such that the sum of the first 127 bytes of the 1 FBR(5) 8560 MUSDU System Reference Manual ___________________________________________________________________________ directory entry and the complement of the checksum is zero. Each file's data starts on a block boundary and occupies max(1,((size + 511) / 512) contiguous blocks. At least one archive block is allo- cated to each archived file or directory so that aliases (multiple links) can be properly recorded. Each directory archived is treated as an empty file (I.E. no directory's data is stored). All entries representing links to a given file are identical. In particular, the fblk field in each link's entry contains the same block number. The archive label directory entry contains the following information: path A comment about this archive. mode Set by fbr to a file readable/writeable/executable by all. Other- wise unused. uid The userID of the user who created this archive. gid The groupID of the same. size The total number of bytes in the data area of the archive. This field in combination with the fblk field, records the size of both the directory area and the entire archive. acct The date-time that this archive was created - not its access time. modt The date-time that this archive was last altered. fblk The block number of the first archive block following the direc- tory area. checksum (same as any other directory entry.) SEE_ALSO fbr(1), stat(2). --- I'll be adding scans of the installation and service manual for the 8560 to bitsavers later today. Dug back though the Tek stuff I had bought a while ago, and I have an 8562 (11/73 cpu) and a 8501. Don't have any docs for those. The 8562 appears to use an I/O board with an 80186 instead of the 8088, and a Micropolis 5" drive.