tag:blogger.com,1999:blog-8869371566442567811.post5131679044840980861..comments2024-03-27T23:23:05.548-07:00Comments on Rescue 1130: 2014 Pickup of an IBM 1130 System and More: Verifying 1130 FORTH is working properly, plus continued battle with the demon disk driveCarl Claunchhttp://www.blogger.com/profile/00359841432891565909noreply@blogger.comBlogger3125tag:blogger.com,1999:blog-8869371566442567811.post-48905734057829477522018-03-26T09:29:42.837-07:002018-03-26T09:29:42.837-07:00Hi Dwight. I replied separately by email but for o...Hi Dwight. I replied separately by email but for other readers, here is the summary.<br /><br />Good idea to check the disk speed, we will do that on the next session.<br /><br />Yes, we have thought about ways to recover data from damaged cartridges. We can read the data on either side of the damage but would miss dozens of tracks by keeping the edge of the head away from the bad spot. Filling in and polishing the damaged spot is a way to reduce the lost portion of the pack contents, but would take extreme care and some careful and precise measuring methods to ensure the resulting surface is safe for the flying head. Carl Claunchhttps://www.blogger.com/profile/00359841432891565909noreply@blogger.comtag:blogger.com,1999:blog-8869371566442567811.post-52028534341577731912018-03-25T21:13:34.603-07:002018-03-25T21:13:34.603-07:00Hi
It just occurred to me that if one wanted to se...Hi<br />It just occurred to me that if one wanted to see a printed number, one could stop the computer and look at it on the light panel.<br />Anyway, just a thought.<br />On the Diablo ( I have a Diablo 30 but only spun it up a couple time and not had time to do much more ) I was thinking, have you checked the RPM of the spindle? Mayby it is just spinning too fast. That would be consistent with reading well but poor writing. Also, try to degauss the heads.<br />On my Diablo, I have a disk that had a deep crash. The heads were OK after cleaning. No damage that I could see and seem to fly without issue.<br />The problem is I'd like to recover the data on the working part of the disk. It is code and stuff for the Nicolet 1080 that I have. It is not likely to be read by anything but a digital scope or the Nicolet because of the unusual sector size( 3000 octal of 20 bit data ).<br />My question is, have you ever heard of someone filling in the rut so that one can recover the rest of the data?<br />Dwight dkelvey@hotmail.com<br /><br />dkelveyhttps://www.blogger.com/profile/10461265216802985944noreply@blogger.comtag:blogger.com,1999:blog-8869371566442567811.post-37049634890113943532018-03-25T03:37:36.833-07:002018-03-25T03:37:36.833-07:00Hi Carl,
Thank you for mentioning my coining of t...Hi Carl,<br /><br />Thank you for mentioning my coining of the term "Proto-FORTH". I'm glad you liked it.<br /><br />I had a spare Saturday morning available yesterday, and having found the scanned listings on SV-Fig, I thought I'd spend a few hours trying to work out just how this thing works.<br /><br />The interpreter has just 29 primitives coded, and these have been chosen carefully, so that an assembler can be easily booted. The first 30 lines of the "forth" listing are this assembler. Once a simple assembler has been written, it makes defining the rest of the language a whole lot easier. <br /><br />I puzzled why the "OR" instruction was one of the primitives - an unusual choice? Then I realised that OR was used every time Chuck wanted to assemble an instruction.<br /><br />For example - here are the hex op-codes for some commonly used 1130 operations <br /><br />.LD C000,<br />.ST D000,<br />.ADD 8000,<br />.SUB 9000,<br />.MUL A000,<br />.DIV A800, <br /><br />Chuck then defines the various registers<br />.X1 100 OR,<br />.X2 200 OR,<br />.X3 300 OR,<br />.XIO 800 OR,<br /><br />With just these 12 short phrases he can already build up a large subset of the 1130 instruction set.<br /><br />By line 40 of the first page of listing, Chuck has a working assembler, so he can start to define the more common words such as DUP, DROP, SWAP, etc using the newly created assembler.<br /><br />This is the fascination of how Forth is built up layer by layer.<br /><br />By the bottom of page 2 of the listings, most of the common Forth words needed have been defined, and Chuck begins thinking about the main task in hand - which is to get the disk fully integrated into his system.<br /><br />It would be great to have a machine readable (text) version of the listings - as being able to annotate with comments would be very useful in this piece of detective work.<br /><br />When you think that all that exists here is about 2400 bytes of 1130 assembler - and a lot of that is the EBCDIC character conversion table, plus 220 lines of protoForth, listings - then rebuilding an understanding of 1968 Forth will take a fair bit of mental agility.<br /><br />What would also be helpful is to define the Forth virtual machine model - from the various registers used and stack held in memory.<br /><br />I have started to make some notes on the primitives, the dictionary format and anything else I can glean from working through the listings.<br /><br />I am happy to share these notes with anyone who may wish to pitch in.<br /><br />Ken Boakhttps://www.blogger.com/profile/01168707285621954181noreply@blogger.com