As far as I know, the fastest production Z-80 chip (the Z-80H) ran at 8MHz. This one runs at 84MHz per CPU!

I switched from 8-bit (Z-80B) hardware to 16-bit (i386) hardware once the 16-bit PC could emulate the Z-80 faster than any physical Z-80 processor could run. I feel fortunate to have skipped the 8088 and 80286 PCs which all ran slower than my Z-80B for most things.

I kept one of my old CP/M machines in my parent's garage for 10 years, and then tried to resurrect it. It was an IMSAI 8080 with CompuPro hardware that I had modified to allow the front panel to work while allowing the system to run at 4MHz. I had also implemented software configurable baud rate and start/stop/parity selection by hacking the hardware and customizing the BIOS.

Unfortunately when I resurrected it, it booted and ran for only a short time before the ST851 hard drives scraped the oxide coating from the system disks. (I destroyed all of the backups while troubleshooting.)

I gave the whole system away before I discovered that they are actually still quite valuable, but I have no regrets. I much prefer a more modern computer.

If you expand that to Z80 compatible chips, the fastest was (is, possibly, as I think it's still being made) the Zilog eZ280 at 50 MHz -- which, given that it runs at about three times as fast as the Z80 at the same clock frequency, is like having a 150 MHz Z80.

https://en.wikipedia.org/wiki/Zilog_eZ80

My TRS-80 Model 4 back in the day (circa 1988, I think) had been upgraded with a HD64180 CPU, which apparently could run at least at 10 MHz -- although I'm pretty sure the third-party expansion board I was using ran it at 6.144 MHz.

You may be aware, but to add to your comment...

Good ol' Z80s are still in active production as well, and you can buy a 20MHz chip in a DIP, LQFP, or PLCC package.

Anyone who is interested should check out the RC2014 and retrocomp google groups. Lots of discussion there, and there are a few people selling Z80 and Z180 computer kits.

https://groups.google.com/forum/#!forum/rc2014-z80 https://groups.google.com/forum/#!forum/retro-comp

> I switched from 8-bit (Z-80B) hardware to 16-bit (i386) hardware once the 16-bit PC could emulate the Z-80 faster than any physical Z-80 processor could run.

Isn't the i386 the first 32-bit x86 processor, not 16-bit?

The 386SX was a 16-bit CPU capable of 32-bit operations. Think of it in the same context as the 8088 vs. 8086.

As I recall, it was a low cost 32-bit CPU essentially identical to the original 80386 (subsequently rebranded 80386DX to further distinguish it from the SX) which was introduced several years earlier and is the main CPU referred to when people say i386 without qualification, except that it had a 16-bit data bus and a 24-bit address bus.

> Think of it in the same context as the 8088 vs. 8086.

Yes, both of those are 16-bit but the 8088 is a later low-cost version with an 8-bit data bus, but an identical execution unit to the 8086.

It is the other way around: the 386SX is 32 bit chip with a 16-bit data bus (plus the jiggery-pokery to convert one 32-bit request from the internals into two 16-bit ones) and 24 bit address bus, so it could be used on cheaper boards originally designed for 286s with minimal modifications. It is still considered a 32-bit chip.

Same for the 8088 - that was still a 16-bit chip like the 8086 (almost identical, in fact), but with a smaller (8-bit) external data bus so 16-bit requests to/from memory had to be split in two.

The replies to my above comment have shown that I was not at all clear about the meaning of my analogy. The 8088 was considered a 16-bit CPU, but it used an 8-bit memory bus. Thus two (or more) memory fetches were required for each instruction. The 80386SX was considered a 32-bit CPU, but it used a 16-bit memory bus. Thus the same memory bandwidth issue that applied to the 8088 applied to it as well.

Imagine an 8088 running at 4.77MHz trying to keep up with a Z-80 running at 4MHz. The Z-80 always won because of the memory bandwidth issue.

We are not disagreeing with the "386SX is to 386[DX] as 8088 is to 8086" part, or the effect on memory bandwidth.

But the "386SX was a 16-bit CPU capable of 32-bit operations" is simply an incorrect statement, which we are correcting. The 386SX was a 32-bit CPU with a 16-bit data bus.

Your new Z80 comparison seems flawed too: while it supports some 16-bit operations over paired 8-bit registers, including 16-bit load & save, the Z80 is generally considered to be an 8-bit CPU (so it is an 8-bit CPU capable of 16-bit operations) and in any case it has an 8-bit data bus (see the diagram in https://www.petervis.com/electronics/CPU_Processors/Zilog_Z8..., note that there are only 8 data lines, D0 through D7) so it can't pull/push 16-bits at once externally any more than the 8088 can.

Okay. From an architecture point of view, I agree with you. The context of my post related to the databus width and the speed implications of it. Sorry I wasn't clear in my meaning.

Wrong. The 386SX is a 32-bit CPU with a 16 bit external data bus. It is fully 32-bits internally, same as the DX model.

> I feel fortunate to have skipped the 8088 and 80286 PCs which all ran slower than my Z-80B for most things.

I might believe that compared to an 8088 running at 4.77 Mhz like in the original PC but the 286 is in another league entirely.

The original IBM AT computers ran at 6MHz. (They later upped this to 8MHz.) Even at 6MHz and a 16-bit bus, they were very close to the same performance of a 4MHz Z-80A.

Or if it was "discovered" that bending the Z80 pins inward a bit and stacking them 16 high over each other actually works, and would have worked even back in the 70s. That would be surreal.

I did that with memory on more than one occasion. Stacking them more than 4 high they'd tend to overheat.

Wouldn't you have to at least wire the chip select pins somewhere distinct for that to work? I know little about electronics.

Yes, exactly that, you bend out one pin per chip and wire that up to some decoding logic. But all of the other pins are parallel.