TSX Plus distribution status...

I've received a number of emails over the past few months as to the status of the TSX Plus distribution for hobbiests. I'm finally to the place where I thought I would be a few months ago! Some of this is background material - but I've included a fair amount of technical information which I hope will enable others to recover data from old disk drives that would otherwise be unreadable. -- As you may recollect, some months ago I contacted S&H Computer Systems, Inc. on behalf of collectors to see if I could obtain a free license, binaries, and documentation for TSX-Plus for collectors. After a bit of negotiating regarding the free license, I was granted permission to release TSX-Plus to hobbiests. I was provided with a copy of the latest release, including documentation. -- About the same time, I was told that S&H might be scrapping out their 11/73 system "in a few months". I told them not to scrap it out - but let me know when they wanted to "make it go away". Shortly thereafter, I was told they needed the space for a tenant - and the 11/73 needed to disappear. I put out a "rescue" request on this list (The "Nashville rescue") - and after a good deal of back-and-forth, a couple of folks responded to rescue. The only items I wanted were the disks that contained TSX "anything" - and that I'd be willing to pay whatever shipping costs there were to California. What I got was six RL02 packs and one Fujitsu 2312K (at about 85 pounds) and its associated Emulex SC02/C controller, cables, etc. (The SC02/C makes the 2312K look like two RK07's plus one RK06). I put together an 11/83 system from spare parts I had around, including a "new" TK50 drive. After getting it up - I tried out the 2312K - and it wouldn't even go "ready". When I disassembled the drive electronics (two boards) - I found that one of the jumpers on the second board was bent badly and the jumper had fallen off in shipment. My guess is that it was bent from the getgo - and the jumper fell off in the process of shipping it from Nashville to CA. At any rate - after I straightened out the pins and replaced the jumper - the drive would go "ready". I was able get directories for DM0:, DM1: and DM2:!!! I thought I was "home free" - but then a "nastyness" occurred - I got many errors (temporary and permanent) reading files from the 2312. I put out a manual request on this list - and fortunately, Joe Heck responded to my request and mailed me a copy of the M2312K manual. This was a Godsend - I couldn't have gone much further without it. After reading both the M2312K manual and the Emulex manual - I discovered a feature which saved the day. The drive and Emulex controller support "Servo Offset Plus and Minus". This means that the conroller will, on discovering a sector error, offset the cylinder position either plus or minus 3 micrometers off track and attempt to read the record again. I modified settings on the Emulex board to turn this feature "on". Subsequent to making this setting, the disk could "almost" entirely be read without permanent errors. Unfortunately, when I tried to backup and verify the disk to tape - I got a number of verify errors. Back to the "drawing board". Many of the files on the disk were huge ".dsk" files. Unfortunately, the likelihood of a permanent error in a huge file proved to be very likely. So while the above fix made recovering small files very likely - the huge files (12,000-36000 blocks long) remained error prone. I concluded that I'd have to further enhance my recovery technique. I decided to add a 40 MB RD32 (ST-251) to the system and recover data by disk to disk copies, followed by tape backup of the RD32. I found the best way of recovering data in this situation was to "mount" the "filename.dsk" files to a logical disks - and then copy the contents to a newly created and mounted logical disk on the backup disk. In many cases there were no errors in any of the copied files - even when the original .dsk files had several errors! While this technique worked well, it took a LOT of work and time. I literally spent weeks getting all the data off the 2312, The good news is this: The original release I received from S&H (which is all end users ever received) was 5.6 MB in size. After the 2312 and RL02 data recovery - I have about 73.5 MB of source, data and code. Included is a complete PDP-11 accounting system (which I've also received permission to release after I strip it of data files). I have the TSXPlus licensing software which can generate keys. This may or may not me necessary for hobbiest use - but at least I now have it if we need it. I haven't had the time to look at all the stuff that's in the 73.5 MB - that's a LOT of PDP-11 code! Given that I also have a "real" job - this has taken an extraordinary amount of work. I feel very relieved that the worst is over and now I can actually "play" with whatever I have archived. (Archived on tape, ZIP disks, and my companies RAID5 disk array - I'm not about to loose it now!!!). Additionally, I paid for the shipment of four huge boxes of TSX-Plus source listings to be sent from Nashville to California - and have given them to Al for scanning into bitsavers. While some of the source is in the archived data - most of it was "lost" by S&H at some point in history - except for the hard copy listings... After a bit of recovery time, I'll finally start getting TSXPlus, RTSORT, S&H COBOL and documentation - with licensing forms - on my Website for any and all PDP-11 hobbiests to download. Cheers, Lyle -- Lyle Bickley Bickley Consulting West Inc. Mountain View, CA http://bickleywest.com "Black holes are where God is dividing by zero"