computer KO

There should be a SWTP 6800 with a homebrew floppy controller and drive somewhere in the garage. It's in a homemade case. There were no disks with it, so I have no clue what disk format the controller expects.

Reply to
Michael A. Terrell
Loading thread data ...

I was really temped to play with the very clever 6809 but by then it was clear to me that the IBM PC architecture would dominate and as a tech I should focus on learning it, even though the company I was working for had bet on the DEC LSI-11 for the control console and the TI TMS9900 microprocessor for embedding. jsw

Reply to
Jim Wilkins

If you remember the old days, it's amazing how fast DOS boots from a USB flash drive on a modern machine.

DOS is still needed for low-level hardware diagnostics like Memtest86+ or the hard drive maker's tests.

Reply to
Jim Wilkins

I remember how slow text files scrolled on the original PC/XT class when you used DOs to display them. I used to use Debug quite a bit, and used a hidden menu system under DOS to keep the know it alls from messing with my computer. The program directories were numbered, instead of named, and to get the menu you pressed X, the return to run it. It drove them nuts! ;-)

BTW, that first hard drive was a 5 MB Ampex.

Reply to
Michael A. Terrell

An engineer had me write batch files to make DOS emulate Unix, for instance an ls.bat that interpreted the - switches.

One way to mess with meddlers is to redefine the screen colors, like temporarily change all colors to blue.

formatting link

After 'hit any key to continue" I flashed the screen bright white if the wiseass user pressed the Alt, Shift or Ctrl keys.

jsw

Reply to
Jim Wilkins

formatting link

Reply to
Steve W.

[ ... ]

Likely 5-1/4" floppies with soft sectors, and (maybe) double sided. You would have to look up the controller chip (likely WD (Western Digital) chip to see whether it could handle double density.

For that system, it was likely targeted at either the SSB or the TSC OS. I remember the first SSB floppy controller was rather tricky. The 6800 had a series of I/O ports on the back, which appeared in blocks of 32 bytes (IIRC) with a corresponding 32-byte gap between them. And, because of simplistic address decoding, that series of I/O blocks echoed up through 4K of address space (IIRC). SSB wrote their boot PROM to occupy all of those gaps, so it essentially did not use any of the system's usable address space.

Enjoy, DoN.

Reply to
DoN. Nichols

OS-9 was being written (along with BASIC-09) while Motorola was firming up the design of the 6809. So Microware had some input to what the instruction set extensions should be, resulting in a rather tightly integrated OS for that CPU.

OS-9 still lives, BTW. Mostly in 68000 based systems for industrial controls.

And while the TMS-9900 was quite intersting, it was an amazingly slow CPU for all that it did. It put all the registers in system RAM, with a "workspace pointer" to show where the 16 16-bit registers were at the moment. A subroutine call was BLWP (Branch and Load Workspace Pointer), giving the subroutine a clean set of registers, except for two, one of which pointed back to the previous workspace pointer and the other of which contained the previous status register. You could access the old registers through indexed addresssing off the one of the current set. All in all, quite intersting.

But -- this design meant that every use of a register required a memory access -- or several. (A later version of the CPU chip copied the registers into on-chip registers, which were faster to access, but for everything which *changed* the contents of a register, it had to be written out to RAM anyway, to maintain compatibility with the old code.It still did run a bit faster.

And in its instruction set, it was closer to an IBM 360 than anything else. :-)

Oh yes -- I/O was also weird. A block of 4096 I/O addresses, distinct from the system RAM, but when you wanted to save a 16-bit register to an output device, it took 16 sequential clock cycles, one for each bit. Same for reading from the I/O bus. So, for fast I/O, it was better to use memory-mapped I/O anyway. :-)

It was supposedly a microprocessor implementation of the insruction set of the It 990 minicomptuer. :-)

Enjoy, DoN.

Reply to
DoN. Nichols

"Jim Wilkins" on Thu, 13 Jun 2013 14:04:51

-0400 typed in rec.crafts.metalworking the following:

I still was able to read text downloads at 1200 baud, but after that it got more difficult. B-)

Oooh, I like that one.

I've been rediscovering batch files, trying to sync my class files from home to the various machines at school. I don't have to reset the software, I've the parameter files on the thumb drive.

-- pyotr filipivich "With Age comes Wisdom. Although more often, Age travels alone."

Reply to
pyotr filipivich

Yeah, isn't it wonderful? ;) Remember when we typed characters onto a BBS and had to wait for them to echo onto the display? Those were the days.

Yeah, nice.

Bueno, bwana.

I remember making lots of money fixing computers that the local Marines had flummoxed. They edited the a.bat and c.sys files with Word Imperfect, saving them in .wp3 mode instead of ASCII. I never had the fun youse guys did with the early computers, but old Peter Norton was the God of DOS and we had lots of fun with DOS. That was before he sold his soul to SlymeAntics and created the dread disease we know today, Norton AntiChrist. Oops, I mean "AntiVirus". I loved his Norton Editor and happily used it for a decade.

Oh, and remember DESQview? I loved that pre-Windows windows program. I jumped to Win 3.1 the day it came out after hurting with Win3. It's not uncommon to have fifteen windows open in a session. Love it! But GAWD, whatever happened to programming skills? What used to be done in 6kb now takes them 600MEGs to accomplish! Arrrrrrrrrrgh! I have

2G of memory on this machine and it's slow with lots of windows open. (Win7)

Yeah, lots of fond DOS memories...

Reply to
Larry Jaques

The product development lesson I took from the distressed childhood of small computers was that how little you got wrong was more important than how much you got right. It's hard to criticize your project with a fresh eye; I learned it as an exercise when building theatre scenery.

That may apply to the success of Honda cars too. The car mags tended to rate them as second best in all categories, while the winner of one may be very weak in another. jsw

Reply to
Jim Wilkins

I put a variable Wait state generator on my homebrew PC that would effectively drop the CPU clock to about 1 - 300 memory cycles per second. At low speed I could watch the text scroll one letter at a time.

The speed pot was very useful for stepping through and debugging larger programs. I could speed up stepping through a long subroutine until the changed pattern of the higher address LEDs showed me it had completed. jsw

Reply to
Jim Wilkins

The magic word to exorcise that demon is "CleanWipe".

formatting link

-xyzzy

Reply to
Jim Wilkins

You could set up local echo, and turn off the remote echo, but the remote did let you know that your connection was still up.

And remote echo at 110 Baud was *really* slow. :-)

[ ... ]

I once wrote for a co-worker a program which sort of acted as a klugy cron substitute. It essentially checked whether the current time when invoked was between two times. He used it in the AUTOEXEC.BAT to do some re-configuration when booted after hours.

It seems that another co-worker was spending all day writing letters to his lawyer, and then printing them out on the main secretary's computer, which had the team's Laserjet connected to it -- after she shut down her computer and went home.

The modifications at boot time would redirect all print output to a file, and otherwise log a lot of what he was doing. He never did figure out why the computer stopped working after quitting time. :-)

[ ... ]

They just can't understand the difference.

I wound up compiling my favorite unix editor, jove, to run on MS-DOS, and using it for anything like that.

The advantage of compiling jove for MS-DOS was that I only had to remember the one editor for everything that I worked with, including a strange unix system by BBN (the C-70), which had your choice of ed, or a strange editor compiled from FORTRAN as supplied. Not even vi/ex, which is why I grew up liking jove, and hating vi. :-)

Well ... the main cause of bloat is the transfer to doing everything with a GUI. I keep a *lot* of dtterm or xterm windows up on my unix system -- quite a few of them actually working on a different system than they are displaying on and accepting keyboard input from. :-)

Mine are fond memories of SSB's DOS-68 and DOS-69, and OS-9, and later v7 unix.

And rather sour memories of MS-DOS's implementation of wildcards compared to unix's implementation. (Anyone else ever name a group of temporary files with an 'X' somewhere in their names, and when done with them typed "DEL *X*.X"? It does *not* do what it would in unix. :-)

Enjoy, DoN.

Reply to
DoN. Nichols

I first got onto a Net in 1972, during an Army field exercise. I was the maintenance man at a mountaintop radio relay site that was part of a data network that used Teletypes for control messages. We were ordered to generate random simulated traffic, so the repairmen, the only people in the vans who could functionally read and write English, set up a defacto network to chat and exchange insults about our officers, coded to be readable on the paper tape holes rather than the printout. The paper tape alphabet was my contribution. It gave the illusion of being standard 5-letter secret code groups.

-hal

Reply to
Jim Wilkins

I think my first experience was 300baud but my first owned modem was

1200. My second was a 9600 screamer. [ ... ]

Har!

I heard a lot about/from you UNIX guys. Mostly gripes. ;)

I didn't get into computing until 1987 or so, other than shunning my dad's used Kaypro CP/M machine with dual 191k floppies. It was good enough for his book editing but not at all interesting to an auto mechanic like I was at the time. I got into computing after I injured my back. (Coleman College's Computer Electronics Technology course, where I learned how to wrench on electrons.)

What's with the equal sign? I don't recall its use in DOS.

Reply to
Larry Jaques

Slick. Soooo, when did the NSA let you go, Jim?

Reply to
Larry Jaques

Who?

Reply to
Jim Wilkins

Try a Telebit WorldBlazer -- with a protocol which it knew, such as kermit, uucp, or one other PC flavored one whose name I forget. It had a buffer, and would quickly handshake the data into the buffer, spoofing the other end. It would then use its own proprietary protocol to send the buffer contents to the other end at a blazing speed, without having to wait for handshaking, and then the other end would receive it via another protocol spoofing. That made a *big* difference in the cost when I was downloading usenet newsgroups via uucp from uunet. When uunet ditched the Telebit modems in favor of some nominally faster modems, the resulting throughput was a lot slower, and I had to drop a number of newsgroups to keep it affordable. (I was charged by connect minute, and the connections were automatic, not with me at a console controlling it.) So -- this change was really a benefit from uunet's point of view -- and I moved to another ISP (Digex, from talking to the owner at a hamfest.)

[ ... ]

At a later time, the same fellow who set up the redirect traps dealt with another fellow who thought he was really hot on computers. But he could never figure out why it stopped working in the afternoon, and was fine again in the morning. (What was the problem was that there was a power timer on the monitor's power cord -- and he *never* found it. :-)

[ ... ]

Pro/Con vi?

[ ... ]

What "equal sign"? That is an asterisk '*' in there, a wildcard.

On unix, it would say "Delete ever file which had an 'x' in the name somewhere in the body.". But on MS-DOS, it works differently as I discovered to my disgust.

MS-DOS did the wildcarding in a rather stupid way. It hits the first '*' and replaces it with '?' (single-character wildcards) for everything from where it is found until the end of the body, or the end of the three-character extension. So:

DEL *X*.*

became

DEL ????????.???

which meant the same thing as:

DEL *.*

But -- since it was not typed as exactly "DEL*.*", it did not even bother with the annoying "ARE YOU SURE?", and just happily deleted everything in the directory -- including stuff we wanted to keep. :-(

Enjoy, DoN.

Reply to
DoN. Nichols

Larry Jaques on Fri, 14 Jun 2013

06:06:54 -0700 typed in rec.crafts.metalworking the following:

"Kids these days! Why when I was a boy, we had to stack the electrons ourselves, by hand!"

Half of it is the need to conserve resources doesn't seem to be there, half of it is software engineers having to justify their employment by making "improvements". I've been drawn into battle with Word 2010 - my SO has to use it, because the school is convinced it is "good". Her question boils down to this: am I here to learn about Creative Writing - or how to us MS Word?" I agree with her. What Win7 does to Wordpad is ... insane. Feature creep. I'm not looking for "Word Lite" so much as a bare bones word processor - I enter text, minimally edit it, minimally format it, save it and print it. No graphic options, no hyperlinks, no "ribbon" getting in my way. I think I could make a mint with a WP package called "My Typewriter" - and that is all it does. "You can have any font you want, as long as it is Times New Roman, Courier, or Arial, in 8, 10 or

12 point." Arrgh, it is way late, and I should be asleep - let alone merely in bed.

thousand lines of batch file to "fake" a menu program.

-- pyotr filipivich "With Age comes Wisdom. Although more often, Age travels alone."

Reply to
pyotr filipivich

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.