Like you, I use the word 'clone'to mean a machine with the same hardware interface and same ROM BIOS calls as the IBM machine it claims to be a lcone of, and thus it runs the same OS (yuo can boot said machine from a disk that will boot the IBM machine) and which runs the same software. Of coruse soem early clones _did_ have incompatibilities. I tend to use the term 'IBM incompatible' for an 80x86-based personal computer that is not compatible with the similar IBM machine. Things like the Rainbow, HP150, Apricot, etc. There are several levels of incompatibility -- the hardware may be differnt (so that programs that 'hit the mrtal' won;t run, the BIOS calls may be different (so that programs thtat use said calls won't run), etc. Of coruse ther are plenty of other machines that are not compatible with any IBM, but I don;'t tend to refer to thsoe as 'IBM incompatibles'. I've also been known to use the term 'IBM compatible' in a jocular way, for example when I said I'd jsut been given an 'IBM compatible ethernet interface'. This happened to be a Storagetek unit to link to an IBM 370 channel. As for the original question, IIRC if the IBM bios fails to find a boot device, it calls a particular software interrupt, which is set to the cold start point of the BASIC ROM. Whether this ROM would work on not-100%-compatibles depends on just what that ROM does about directly accessign hardwre. -tony

________________________________ From: Diane Bruce <db at db.net> COMPAQ and Hyperion Dynalogic solved this by relocating the BASIC at run time to high memory. ick ick icky poo. But amusing. C: High in DOS memory, or otherwise? Some computers made 768k or even 896k available to DOS.

Unfortunately (and perhaps obviously) the BASIC ROM source listings were never released, and the TechRefs say very little about what ROM BAIC actually dows. I guess one approach would be to saech the ROM BASIC for driect I/O instructions (there must be some, if only for the BASIC statements that allow a user program to directly access I/O ports) and also software interrupts. If there are plenty of the latter and few of the former, it's a good bet it'll run on BIOS-compatible clones (as opposed to running on only hardware-compatible clones). -tony

Tony Duell wrote: Dave McGuire wrote: I'd wager that it's an almost literal translation of a late version of Microsoft BASIC-80, and I seem to recall that there are some commented disassemblies of that, so it shouldn't be too difficult. It's probably not even that different than Microsoft's TRS-80 BASIC.

On 12 Jun 2012 at 12:18, Chris Tofu wrote: I take "clone" at its face value the same way as I view "pregnant".? You either are or you're something else.? An Eagle 1600 runs MS-DOS but is not a PC clone. C: Well I did use clone in my op. It's often used in different ways. I hadn't thought people would always revert to giving the most obvious answer to a question. Of course if you made a totally identical clone mobo of a 5150/5160, the roms would work. My question was (or should have been) given? an empty socket (as is the case on a 5150/5160 mobo, rarely anywhere else) or a means to extend that bus to accept additional roms via a plug in board, would you expect BASIC roms to work in say a Tandy 2000, AT $ T 6300, Compaq Portable (original), Zenith Z-100/120, etc. Wouldn't most of the lower level stuff be BIOS driven? ?And while there are not degrees of pregnant, there are degrees of clonage. Looser usage albeit. Did Compaq *not* seek to clone the operation of an IBM PC? From a s/w vantage, it most certainly was a clone. But I'll agree a true clone is mostly and exact one. ?Did you mean the Columbia MPC1600? I don't personally know of an Eagle 1600, unless you're referring to a little seen portable unit. But I don't follow the rest of your post--perhaps if you used shorter sentences... C: Ok. I will. Next time. Promise. No more. Then two. Words between. Periods. Period.

On 13 Jun 2012 at 12:20, Chris Tofu wrote: Nope--I'm referring to the 1982 8086 system brought out by Eagle Computer. IIRC, the MS-DOS floppy format was 800K (8x1024 byte tracks). Not many probably survive, because it was followed in a few months by the utterly-PC-compatible Eagle PC. At one time, Eagle was the darling of retailers--with the later Z80-based models, you opened the packing box, took out the computer, plugged it in and turned it on. Everything was pre-loaded and configured. The word processing program offered with the Eagles was Spellbinder--a very fine bit of software. Eagle used to hold public contests to determine what was the fastest a first-time user could get the hardware up and running. I seem to recall times around 45 seconds. If you don't know the whole story of Eagle and Dennis Barnhart and the bright red Ferrari, you owe it to yourself to read up on it. There were other 8086-based clone-ish systems at the time of the 5150. One was the Stearns PC that came with a copy of MS-DOS, but also something called ST-DOS that addressed a number of compatibility issues with the 5150. Call it a semi-clone. As far as ROM sockets, well, that's a system-by-system thing. Many Taiwanese clones had 5 or 6 28-pin ROM sockets for 64Kb ROMs. Later ones went to one or two 256Kb sockets. You could take one one of those and play with the PROM binaries a bit to get IBM BASIC to map in at F6000 and be good to go. --Chuck --Chuck