|
Whatever you call the age, (its color depends upon your historical frame of reference,) vintage computing equipment reminds me of the time when I was first discovering computers as magic devices.
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
Ahh! The Magic Age!  Every day...
|
|
|
|
|
I find your reply fascinating, even if it doesn't make any sense to me.
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
|
(Back then was magic. Now is also magic.)
|
|
|
|
|
Of course it is, I'm not disagreeing. I just think everybody likes to remember the "good ol' days".
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
David O'Neil wrote: Now is also magic.
No. It's quantum mechanics that you may need these days to explain why your junk does not work. On top of the traditional ones like capacities, inductivities, noise in general, thermal problems or timing problems, of course. Even those in all possible combinations could be quite arcane to track down.
I have lived with several Zen masters - all of them were cats.
His last invention was an evil Lasagna. It didn't kill anyone, and it actually tasted pretty good.
|
|
|
|
|
Lead. The correct metal to name the age after would have to be lead, as in soldering together your own hardware. These also still were the days when my old cat used to toast her rear parts on my 4k RAM expansion.
I have lived with several Zen masters - all of them were cats.
His last invention was an evil Lasagna. It didn't kill anyone, and it actually tasted pretty good.
|
|
|
|
|
David O'Neil wrote: I don't miss those old 386's. Bootups in minutes...
Hardly. My first "commercial" PC (bought from the company I worked for when it went bust and sold for scrap) was a 386 with 32Mb disk and 640Kb memory. It booted to DOS in about 3 seconds and loading up Win3.11 took about another 10. These days I avoid turning mu Core i7 off if I can possibly help it, as a cold start to a useable system takes at least 5 minutes. And god forbid if MS decides it's going to apply an update to W10, it can take anything between 15minutes and 4 or 5 hours. So yes, I certainly do miss the boot-up times of "old days". (And of course my very first Compukit UK101, running the 6502 mentioned in the original post, gave the a prompt sub-second from applying power. Warming up the CRT took longer  )
|
|
|
|
|
I think the the early 1980s was more "Cambrian Explosion".
A massive diversification of computers, most of which died out after the dominant species took hold.
|
|
|
|
|
I had the Radio Shack Color Computer (Coco), and it used a 6809 processor, was actually pretty impressive. Beyond the standard DOS OS that a lot of microcomputers of the time, RS also sold a version of OS-9, a small multi-threaded operating system (well, probably more of a task switcher, but still cool on such a small platform).
|
|
|
|
|
Ah, and it came with the very characteristic MC6847 graphics chip. Remember that awful text mode? You only have two choices of colors. Text can only be dark green on a light green background or brown on a bright orange background. Bleeding eyes are guaranteed and the fixed color palettes of the graphics modes are not better.
The 6847 used to be quite common back then and even today is still relatively easy to find for building your own 8 bit computers. I have two of them in my parts box. but when the day ever comes that I will actually build a graphics card for the Zwölf, I will probably take a TMS9918 instead. Or, if I actually have to use it, I will probably 'forget' to correctly mix in the color signals and leave the video signal grayscale only.
I have lived with several Zen masters - all of them were cats.
His last invention was an evil Lasagna. It didn't kill anyone, and it actually tasted pretty good.
|
|
|
|
|
CodeWraith wrote: I will probably take a TMS9918 instead
Was that the chip used inside the TI-99/4A home computer?
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
At the time, I was fascinated by the 9900 chip (the CPU, not the 9918 graphics chip), having all 'registers' in memory. At an interrupt, you could swap context completely simply update the "Workspace Pointer" to another complete register set.
The bad thing was of course that register operations were no faster than memory operations, so all things considered, the chip was terribly slow. But interrupt handling was terribly fast!
Sometimes, when I dig up memories of those quite different solutions that were ditched, I can't help but wondering: What if someone took up the ideas today and modernized them? A 9900 model 2022 would of course be a 32 or 64 bit chip, and I assume it would be built with a caching mechanism for the register blocks. Maybe, for some application areas were super fast interrupt handling was essential, and "unlimited" number of threads was a valuable property, that idea of just setting a pointer to the appropriate register block might have some merit!
(I have worked on a machine with something resembling it: Interrupts were prioritized into 16 levels; a high level could suspend a lower one. Each level had its own physical register set. The first instruction of the interrupt handler started executing 900 ns after the arrival of the interrupt signal, which was rather impressive in the mid 1970s, when these small 16 bit minis (PDP-11 class) where developed and sold.)
|
|
|
|
|
Yes, I remember it well. I learned assembly for the TMS9900, and I recall that the architecture of the TI-99/4A was such that the video RAM was separate from the CPU RAM. In order to write to video RAM, the CPU had to set certain memory locations and then perform several NOP instructions to allow the video processor time to read the values that were just written.
This led to extremely poor graphics speed, and in my view was just a dumb design.
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
I don't expect anyone picking up a 30-40 year old architecture to feel obliged to carry forward any or every specific detail in the old design. Like the example you give: That seems to be something we can leave behind.
The important thing is if there are forgotten solutions that might be more viable today than they were when they were left. Even though 99/4A had a poor video RAM interface, still the idea of a register block pointer to memory may be far more realistic today than forty years ago. As far as I remember, at the time of TI-99 cache memory was completely unknown in the microprocessor world (and for some years to come). I am not stating as a fact that cache makes the register block pointer idea a good one, but it might change the situation so that in some context where fast interrupt handling and thousands of threads is essential and far more important than beating x64 on number crunching.
Whenever I tell my 'hope' that Intel would reconsider the ideas behind the 432, to make a truly object oriented processor, someone instantly jumps up: Forget it! The 432 was limited to 8 Ki objects - that is completely useless today! ... So why not make a 432 model 2022 with 4 Gi objects, then? But, I hear you cry, the 432 objects could be of max 4 Ki bytes! (if my memory is correct - don't quote me on that limit) ... So why not let the 432 model 2022 have object up to 4 Gi bytes?
The same thing here: Even though there may be bad partial solution that you don't want to pick up, don't let that keep you from reconsidering other partial solution that might be much better.
|
|
|
|
|
I also had a fascination with the TI 9900 chip. Loved that register concept. I don't collect vintage computers, but I have my TI99/4A in it's original box, with some of the original paperwork, and many cassettes of old basic and assembler programs. I haven't turned it on for 20 years. I think I tossed my TI cassette player, but I still have it's box. So, no collecting vintage hardware, but one piece that I can't seem to let go.
|
|
|
|
|
Yes, that's the one. The selection of suitable graphics chips that are still available in some way unfortunately is not very large.
I have lived with several Zen masters - all of them were cats.
His last invention was an evil Lasagna. It didn't kill anyone, and it actually tasted pretty good.
|
|
|
|
|
Much more entertainment writing the games (even noddy ones) than playing them IMHO.
|
|
|
|
|
refarding that failures list, for me meeting high expectations is not same failure.
never released to market is likly the story of 100 to 1 products never released. Pippin, 64dd, sega cd, 32x.
oh we only sold 9 million, boo, we wanted a billion: Sega Saturn, Dreamcast, VirtualBoy, WiiU,
I saw enough in the world, but still marketcap expectations: ngage, ps vita.
yeah, the paragraph is enough to explain its dislike: ps classic.
for any others, if the companies broke even, well guess what, stop being greedy.
|
|
|
|
|
I put together a GitHub - Kris-Sekula/mgh80: Simple 5 IC Z80 singe-board computer[^] kit, just for the fun of it. I played with it for a while then I got another simple Z80 board, added a keypad, 6 - 7-segment display and created a monitor program for it in assembler. Had a lo9t of fun with it but haven't had a chance to go back and do any more with it.
The most expensive tool is a cheap tool. Gareth Branwyn
JaxCoder.com
|
|
|
|
|
Looks interesting! Must've been like programming a TRS-80!
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
Probably, never had one. Had a Commodore64, that was a lot of fun.
The most expensive tool is a cheap tool. Gareth Branwyn
JaxCoder.com
|
|
|
|
|
I think that's more computing power than Honey the Codewitch uses on a daily basis.
|
|
|
|
|
That could be true: Not long ago, I was working on an IoT-project (similar to Honey) using an ARM M0 that could be clocked as low as 32 KiHz. Total power consumption is very dependent of clock frequency, and IoT is very dependent on not draining the batteries too fast. This M0 at 32 KiHz had extremely low power requirements. (A common in-house joke was that it could run on the leakage current from the battery  )
I would not be the least surprised if an 1.5 MHz 6509 could do a lot more that that ARM IoT chip. At the same time, I suspect that the factor in power requirement was a lot higher that the factor in processing power.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
