• Please review our updated Terms and Rules here

CPU family tree - help needed

6805 and 6808 are mostly microcontrollers for embedded applications. They usually have lots of embedded flash and some RAM and special IO. They are similar to AVR with many memory/IO options so perhaps they are numerous, but minor branches of 6800 Family tree.

683xx are mostly embedded computers. 68328 is well-known because it is in Palm Pilot. 68332/68336/68340/68376 are mostly embedded engine controllers. 68360 is often an IO companion to 68040. I’ve reverse engineered a SBC based on 68302 and ported CPM68K to it. Originally it was a controller for network server.
https://forum.vcfed.org/index.php?threads/porting-cp-m-68k-to-a-recycled-board.61681/#post-746253

P90CE201 is new to me but I ported CPM68K to it 2-3 years ago. I don’t know whether or what commercial computers were based on it. https://www.retrobrewcomputers.org/doku.php?id=builderpages:plasmo:p90mb:p90mb_r1

I think you should remove Z800 so the family line goes directly from Z80 to Z280. Tilmann Reh famously designed CPU280 for hobbyists based on Z280.

I collect CPU and have several tubes of NS32008 but I don’t know any computers based on 32008.
Bill

Thanks, I will remove the 6805 and 6808. The 68328 is a tricky one, it's a microcontroller but was used in some popular early hand-helds. I will include it because Palm Pilots are important devices in computing history.

The P90CE201 seems more like a CPU with some built in peripherals than a microcontroller, especially as it doesn't have RAM or ROM. The line is somewhat blurred. Many modern CPUs have peripherals - memory controllers, PCIe controllers, even USB controllers and SMBUS for mobile parts. I think though if it wasn't used to any known computers I'll leave it off for now.

I'll remove the Z800.
 
Motorola Coldfire ?

I thought about it, but can't find any machines it was used in. I remember there was some talk of it being an upgrade path for Amigas at one point, but nothing ever came of it. I get the impression it was only ever used in embedded industrial system, but correct me if I'm wrong about that.
 
Do you have a date for the introduction of the LSI-11?
I'll look for it. I realized after I posted that the Pro300 line used the F11 and J11 processors, which are versions of PDP-11s, but not the same as the LSI-11. I'll send whatever I can find.

I am not adding microcontroller versions, unless they were used in micro computers. Similar to the no mini/mainframe computers rule. I might make exceptions if there is some compelling argument, like they were used in a popular home computer but not as the main CPU. For example I was thinking of adding the 6500/1, which was used in Amiga keyboards.
When you say "microcontroller versions", do you exclusively mean variants of general purpose processors? If not, you might consider the 8048, which was used in many keyboards, including the original IBM PC's. In your shoes, I wouldn't want to include it, but I thought I should bring it up and let you decide.
I also try to avoid minor variants like versions without an FPU...
So is thay's why you don't have the i486SX. (I never liked that chip, but that's a different story.) What about the 386SX? That was a 16 bit external bus version of the 32 bit 386 architecture and instruction set, analogous to the 8088.
 
kuro68k,
your name reminds me of “kuno“, a 68K-based (P90CE201) SBC for Arduino Mega enclosure that runs CPM68K. ;-)

It's from the Sharp X68000, with "kuro" meaning "black" or "expert" depending on how it's written. Nearly went with "kami" which is god, because that machine was known as the "god computer" as it was so powerful, but I thought that might be a bit much :)

When you say "microcontroller versions", do you exclusively mean variants of general purpose processors? If not, you might consider the 8048, which was used in many keyboards, including the original IBM PC's. In your shoes, I wouldn't want to include it, but I thought I should bring it up and let you decide.

So is thay's why you don't have the i486SX. (I never liked that chip, but that's a different story.) What about the 386SX? That was a 16 bit external bus version of the 32 bit 386 architecture and instruction set, analogous to the 8088.

Thanks. The 8048 is another tricky one. If I include parts used in Amiga keyboards, maybe I should include that. Perhaps a useful rule would be if the part is likely to be found in retro computers/consoles that people own. In that case, that family was also used in TRS-80 computers and in the Maganvox Odyssey 2. It's not related to any other parts as far as I can see though, it was a different instruction set to all other Intel parts.

Hmm, not sure what to do with that one. I need to think about this a bit more, because as you say the 386SX is just an 8 bit bus version and I have included several other "same but smaller bus" chips. It should at least be consistent.
 
Thanks. The 8048 is another tricky one. If I include parts used in Amiga keyboards, maybe I should include that. Perhaps a useful rule would be if the part is likely to be found in retro computers/consoles that people own. In that case, that family was also used in TRS-80 computers and in the Maganvox Odyssey 2. It's not related to any other parts as far as I can see though, it was a different instruction set to all other Intel parts.
Yeah, it's a dedicated microcontroller, not a "microcontroller variant" of something else. That's why I said that, in your shoes, I wouldn't include it, and I only brought it up because you mentioned use in keyboards specifically.
 
I don't get why you would not want to include everything in your chart. All microprocessors, bit-slice processors, microcontrollers, PIC, SoC, the lot. All variants. If there are conscious omissions you've elected to make, you may have people rightly assuming your chart is not "complete".
Is there a problem adding everything?
 
I don't get why you would not want to include everything in your chart. All microprocessors, bit-slice processors, microcontrollers, PIC, SoC, the lot. All variants. If there are conscious omissions you've elected to make, you may have people rightly assuming your chart is not "complete".
Is there a problem adding everything?

Mainly space. It's already bigger than A1 size. I could make the boxes more compact perhaps, but if it is to be readable when printed there is a limit to how big the whole thing can be and how small the fonts can be. I wanted to add some additional notes to some items too, like examples of machines they were used in.

It's mainly aimed at vintage computer enthusiasts like myself, so for any given machine we are likely to encounter we can see where its CPU fits in.
 
I remembered the Geode line, as used in the OLPC and others. Should probably include them for their historical significance, but I can't find dates for the release of the various models.
 
I'm missing the Coldfires as well as 68k dependants, also the Phillips 68070 is missing (that was a µControllor based on the 68000).

I'm also missing the Sharp SM83 (of Nintendo GameBoy fame), which was a Z80/8080 descendant.
 
Thanks. I have added most of the suggestions, and started listing some famous machines that used various parts. I'm not sure how well the machine listings work, might remove them.

I also experimented with a bit of shading, trying to show groups of relation processors. I think it's going to be too awkward though.

tofro: Was Coldfire used in any computers? It seems to have been limited to industrial/embedded systems.
 

Attachments

  • Family Tree 2023-05-31.pdf
    246.3 KB · Views: 8
Update with additional CPUs, and some re-organising. I've started adding shading for related parts, not sure how well it works.

I've also been adding some notes on well known using of different parts. If you know of any famous machines that used particular parts, please let me know.
 

Attachments

  • Family Tree 2023-06-06.pdf
    249.9 KB · Views: 11
Missing the short lived AMD K6-2+ & K6-3+ which were not the same as the K6-2/3's.

Thanks, I will see if I can find some data and add them.

In the mean time I've been working on the system timeline. Now it's just computers (I'll do a separate console one) and the PDF is about 50MB: https://drive.google.com/file/d/1wZa6H-ALBJv_0HDusrNPndsBgo10WNOG/view?usp=drive_link

The images are all from Wikipedia, since they are CC licenced. I need to add attribution of course. There are a lot of missing images, especially for Japanese machines. I'll have to start asking around.

I was thinking of adjusting the timeline to make some years wider, so they don't need to stack so much. I'm running out of vertical space. Not visible on the PDF are Microbee systems, and the PC-9800 series. I have the data but no room for them at the moment. There are some other Japanese 8 bit systems to fit in, and I'm sure some more Western ones that I forgot.

What do you think? Make some years wider and then maybe put the later Mac stuff in a little separate area?
 
The short lived K6-3+ (2+ had smaller cache) could run with a 6x multiplier (Intel 2x setting) plus had a low-power mode since it was designed for OEM laptop use. It also used any existing on-PCB cache memory as L3 cache. It was released for OEM use only and not retail, but they did show up in the secondary market from some suppliers. I used them to upgrade a few models of socket 7 laptops to equal the performance of a 1st gen PII CPU without having to purchase a new laptop. Plus the low-power mode allowed very long battery usage compared to the original Pentium MMX/AMD K6 CPU's. A BIOS update (or boot program) was required to turn on the low-power mode.

See: https://www.tomshardware.com/reviews/socket-7,262-3.html

Note: It was sometimes possible by de-lidding a K6-2+ to convert it into a K6-3+ with the revealed hidden jumper settings in some cases. I guess to use defective cache K6-3+ for better early production yield. Perhaps later just a marketing ploy as often happened with both AMD and Intel pricing schemes back then. See:
 
Thanks. Were the 2+ and 3+ both released in the year 2000?

I'm also trying to understand how they fit in family tree wise. Were they both descendants of the K6-III or is the K6-2+ based on the K6-2? It seems like if you can convert some of them into a III+, and if AMD wanted to reduce the number of cores it was fabricating, basing them off the same one would make sense.

I noticed something interesting. K6-2 CPUs are branded that way, with a "2". But K6-III CPUs have Roman numerals instead. Same with the 2+ and III+.
 
Back
Top