6502 or Z80 Which Is Better?

I've used many old MPUs ranging from COP1802, 2560 SCMP, Z80, 8080, 68xx, 6502 and derivatives (and the list goes on...)

The Z80 and 6502 can't really be compared against each other because of their disparate features.

Here are some aspects;

You can't directly compare clock speeds against each other because the timing of the CPUs differ relative from their system clocks.

The Z80 has many inbuilt registers allowing it to operate without any external RAM whereas the R6502 is dependent upon external RAM at page 0 (0000-00FF) but this RAM at page 0 has special instructions that make extremely versatile.

The Z80 has I/O port addressing which does not consume precious memory space.

The Z80 has inbuilt DRAM refresh circuitry but this is also can be a hinderance.

The Z80 handles the DAA instruction different to the Intel 8080. - It basically corrected it.  This instruction is often used to sense whether an 8080 or Z80 is present.

Z80 peripherals have a daisy chained interrupt priority scheme.

68HC11 was my workhorse processor for small projects in the late '80s and '90s, then Microchip and Atmel started making fun parts.....

Scott

Gotta underline something MK said about 68HC09 and -11.  These were a little more latterly, I think by Motorola.  Parallel multiplication and orthogonal math instruction sets made these chips a relative joy to use.  Interesting that the AMD 9511 math coprocessor didn't catch on

I gotta agree with MK that the 6XXX processors were smoother and more pleasant conceptually, the Z-80 had some neat features, more registers, complete shadow registers for swift context flipping, and that tres groovy sub-rosa dram refresh. . .

Had fun with both

I voted for the 6502 because it was the first microprocessor I ever owned. I was a student back in the mid seventies, but couldn't afford any of the processors that were then on offer (the various Intel and Motorola parts). Then MOS Technology launched the 6502 at a $25 price - they had full page adverts in the various American design magazines. I optimistically sent them an international money order in the hope they would send me one - there was nowhere in Britain I could buy one from - and, a couple of months later, when I had given up on hearing anything, a package arrived with the Hardware Manual and the Programming Manual in it.  The paperwork just listed the manuals and at first I was a bit disappointed, but when I came to dispose of the packing I discovered the processor tucked in one of the sides - they'd obviously done it so I wouldn't have to pay the import tax, which was nice of them. It was a lovely white one (ceramic DIL package with a gold coloured top over the chip). [Possibly it was an engineering sample that someone had fished out of a drawer.] I built several computers with that chip - the first having a 128 x 8 RAM which was loaded from switches, with LEDs to show what the address and data buses were doing.

I spent many hours in my youth programming Z80 machine language on an S100 based Vector 1 computer. 

It is my favorite processor from that era simply because I learned so much from using it.

Better is a dodgy word in this context.

My favorite processor of that era was the Motorola 6809  - in many ways way ahead of it's time.

The 6502 was nice and in many ways much easier to code effectively than the Z80. The Z80 had all sorts of extra goodies which made it effective in a low cost systems kind of way in things like the Sinclair computer. An awful lot of Z80s never did any more than pretend to be Intel 8080s so the good bits were largely unused.

I did commercial designs based on 6502, 6802, 6805, 6809 and then 68HC11  - I always preferred the 6800 based architectures (of which the 6502 was a relation although made by different people).

MK

Yours and mine both. 

Assembly language is powerful, but sometimes tedious to do.

I enjoyed the projects I implemented, but technology has advanced far enough that you should not need to go there any more.

DAB

timswift "Better" is open to a wide set of interpretations.

The Z80 was had more CPU functions than the MOS Tech 6502 although the 6502 enjoyed more success at the time because of Apple, Acorn, and Atari (400, 800) usage.  The 6509 and 6510 were quite successful as part of the Commodore CBM and 64 computers respectively.  One wonders why MOS Tech did not conquer more of the world.  Probably set back significantly when IBM went with the Intel 8088.

DAB - At Datapoint Corp in the early 1980s, we used a 6MHz version ("Z80B") on the Datapoint model "2200" small business computer. I don't think that Zilog sold significant numbers of the "Z80H" which ran at the "break-neck" speed of 8MHz - maybe in the Bally game center machines?

Back in my Assembly Language programming days, my favorites were definitely the Z80 and the Intel UARTs and USARTs.

have observed that there is a lot more online information on the 6502 then the Z80

But with three times the transistors you would think the Z80 would have a advantage.  So I guess it's a pretty unfair comparison of processing power on my part.

Tim