I am getting old. I remember I played with all of these when I was younger. Z80 in my Sinclair ZX81 and later ZX-Spectrum, and then in an MTX-500. The 6502 in a Commodore C64, and an MC68K in an Amiga.
Long time ago... The ZK81 had 1KB or RAM, seems impossible to do anything with that little RAM today. Yet, we wrote games with it.
I always feel a bit nostalgic when I read these kinds of articles.
What amazes me most is how easily one adapts to new technologies. These days I take my 4 core laptop with dedicted GPU running Linux for granted and don't think about it twice.
But that's also the fun part. I'm doing software for two decades professionally now, and there is always something new to learn, something new to pickup and explore! It's exiting, I hope to get to do this for another decade or two.
It was also fun to be part of those "pioneering" year (albeit just as a user of the technology).
To be pedantic, the C64 had a 6510 processor - but that was just a 6502 with some extra I/O capabilities.
I cut my teeth on assembly programming on the 64, writing a VT100-compatible terminal emulator that ran in graphics mode to give a full 80 column display. I had also cracked the problem of printing on the IEEE bus while keeping serial comms happy using a combination of data buffering and fast interrupt handling.
"The first thing that strikes you about VisiCalc is its size, or rather, lack of it. You get a 27,520 byte MS-DOS executable (actually a .COM file) and nothing else. No installer, no uninstaller, no DLLs, just a fully functional, no frills spreadsheet program. Run it from your hard drive, run it from a floppy or run it from a USB key ring if you like; this thing is tiny!" (circa 2004)
- they used the port-direction bits in the parallel I/O chip registers as temporaries. After all, you only care about user input for a tiny part of the video frame
- some games didn't use the stack . . . or not all the time. So the stack pointer was just another register, too
A friend of mine had carpet on his bedroom walls, so when he upgraded from his ZX81 to something better, he took the circuit board out of the casing and had it stuck on the carpet wall, it looked better than having a painting on the wall.
The MIPS R3K is deserved of a mention. This chip was at the heart of early SGI workstations, the machines that pioneered everything 3D whether that be on screen or in some GIS application. Although SGI were known for the movie stuff that was the tip of the iceberg, when it came to meteorology, geology (oil) and military applications a lot was possible thanks to SGI, and, by proxy, the MIPS chips.
Later on there were plenty of gaming consoles that used the MIPS R3K series chips. An expectation that games would be in 3D rather than 2D sprites came along with it.
Clearly a lot of other innovations went on at SGI and around anything with a MIPS chip, however, for a while it was the preferred option for any hardware that did 3D.
I'm not a hardware guy so I didn't really know what to expect from an article about "microchips that shook the world". The title seemed somewhat hyperbolic. I clicked on it to see if there was any mention of the 555, one of the few "chips" I know about.
Some 20 years ago my sister's boyfriend and I made a simple "autofire" for my Amiga. I guess it was some kind of bonding thing. An autofire was a piece of hardware that you would hook up between a joystick and the computer. I didn't know much about electronics so he put together the schematics. It involved a 555 and a potentiometer so that the fire rate could be controlled. I was good enough with a soldering iron that he let me assemble the thing. Fun times.
(Turns out that the article had a few more chips I recognized.)
When I think about Transmeta I see bad journalism on the part of the IEEE Spectrum. Many read the magic show IEEE Spectrum article and incredulous believe that the performance problems were due to low performing 16bit code but when in-fact the idea at the core was flawed. They had recreated the x86 through shadow registers, the only performance gain would be to perform serious optimization which is a job for a software compiler. When the demand was right Intel pulled off Core line.
The 68 000 from Motorola was not "just" in the Amiga and the ST and the early Macs, it was also in the Megadrive, and in most of the early Sega Hardware in the arcades (Outrun and AfterBurner 2 had dual-CPU boards using 68000 in them). This chip was a wonder at that time.
Interesting - to me at least - footnote: Federico Faggin attended the University of Padova and is from Vicenza, not far from here. Sadly, like many people in Italy in the past and present, it seems that he was a lot better off leaving in terms of his career.
An amazing trip down memory lane!
I did so many things with the good old 741 many many years ago, my favourite being making a wah wah pedal from scratch, very simple design, and then I had to go for distortions and phaser effects. nice.
"Everything You Always Wanted to Know About the ICL8038" sure is amusing and interesting. It answers questions like the one given in the article, and my favorite, number 7: "This waveform generator is a piece of junk. The triangle wave is non-linear and has large glitches when it changes slope."
What is funny is that in the end SPARC, PA-RISC, etc probably made things worse because all the workstation vendors used the same 68K chips but they each had different RISC architectures, meaning there was less volume, and of course chip fabrication depended on volume.
I am currently reading Commodore - a company on the edge. Highly recommended to anyone interested in the early days of microprocessors, the 6502, MOS Technology, Chuck Peddle, et cetera.
I'm a little surprised by the fact that none of the Atmel chips made the list, considering how foundational the Arduino is for introductory DIY electronics.
Long time ago... The ZK81 had 1KB or RAM, seems impossible to do anything with that little RAM today. Yet, we wrote games with it.
I always feel a bit nostalgic when I read these kinds of articles.
What amazes me most is how easily one adapts to new technologies. These days I take my 4 core laptop with dedicted GPU running Linux for granted and don't think about it twice.
But that's also the fun part. I'm doing software for two decades professionally now, and there is always something new to learn, something new to pickup and explore! It's exiting, I hope to get to do this for another decade or two.
It was also fun to be part of those "pioneering" year (albeit just as a user of the technology).