Is anyone interested with the Z80?

Always good to be sceptical.

http://www.eetimes.com/document.asp?doc_id=1199863

https://www.researchgate.net/publication/3941932_013_mm_210_GHz_fT_SiGe_HBTs_-_expanding_the_horizons_of_SiGe_BiCMOS

https://books.google.com.au/books?id=sanRmmHKiMgC&printsec=frontcover&dq=isbn:1420066862&hl=en&sa=X&ved=0ahUKEwirgdiNga7…

As for speed if we built a Z80 with SiGe transistors it'd operate at 210GHz at that speed it becomes difficult to interconnect it with other components.

I kind of doubt that statement, sir. IBM claims SiGe speed records | EE Times  back in 2004 claims a 60 GHz speed.  But Z80 could operate faster if miniaturized to the degree the ARM is.

Clem

Yes, but how do you program it into a PSOC without an accompanying computer?

I haven't looked at it in detail but;

A typical ARM and Z80 has its RAM access speed based upon the period of its CPU Clock speed.

At 8MHz it's 125ns for both devices (and 250ns for ROM on an ARM).

As for clock cycles required for each instruction, they require about the same give or take a bit.

Things get relatively more complex on an ARM because many of its instructions require 3 operands.

e.g.

     ADC r1,r2,r3

Admittedly the Z80 has some limited 3 operand instructions with its index registers.

e.g.

     LD     A,(IX+d)

As for speed if we built a Z80 with SiGe transistors it'd operate at 210GHz at that speed it becomes difficult to interconnect it with other components.

Ah but have you looked at the ARM assembly language under the PSOC?

I suspect that it might be about as efficient as the  Z80 code, but significantly faster.

DAB

Hopefully this thread will pick up and get some folks wanting to play. Doesn't have to be a serious endeavor and maybe we can all learn(or re-learn).

It has been a few years since I played with these last, but may go drag some stuff out get started again. It's a nice break from the modern gear.......

Thanks for your input-will try it out!

As they say in Mortal Kombat - "Most Impressive!"

For NMOS CPUS I'd use higher value pull down resistors to cater for the one TTL (or 4 LSTTL) load limit.

If you put an 8 way dip switch in series with R1-R8 and a LED and current limiting resistor on the /HALT pin to VCC you can perform other programs.

Here are some examples to try;

11000011  C3  -> JP C3 C3 (continually loops at address C3C3 through C3C5

01110110  76 -> HALT     (Turns the HALT LED on - and Address LEDS still change value (for REFRESH))

00011000 18 18 -> JR (+18) similar to the NOP loop but in bigger steps.

11001001 C9 -> RET

11001101 CD -> CALL CDCD

11010011 DB DB -> IN A,(DB) (If you hook up a LED and current limiting resistor to the /IORQ line to VCC - it should blink)

10101010 10 10 -> DJNZ +10

This is the circuit I made boards for sometime back to get people playing with them:

More like a 16 bit counter but remember that the Z80 will pump out Refresh addresses on (M1) instruction fetch cycles.

The NMOS part has a lower speed minimum but the CMOS can be run as slow as one likes.

HI DAB

As you know I've used both.

The issue with using modern parts is that one needs a PC and also allows one to gloss over the fundamentals with everything provided on a platter.

Building a Z80 system forces one to at least assemble some hardware, learn machine code if desired, have many more hazards to navigate and develop a full sense of what is going on.

Here's a comparison of the different levels of programming learned.

PSOC C

main() {

     unsigned char a=0;

     for(;;) {

          write_LED(a);

          a = ~a;

     }

}

Z80 ASSEMBLY

LED_PORT     EQU     %01

START:

     XOR A

$1:

     OUT     (LED_PORT), A

     CPL

     JR     $1

Z80 HEX

AF D3 01 2F 18 FB

Have a great day!

Hi Monte.

I used Z80's back in the 1980's and used to have a bunch of them around, but with all the other processors now available, they are not something I would try to use today.  Too many support chips needed.

Given the Cypress PSOC, I would start there instead.

DAB