I studied the book on z80
ZXNet echo conference «hardware.zx»
From Petr Grushin → To All 1 December 2006
Hello hero
her> Show me the 4 byte op code, and for the 2 byte op codes M1
her> generated for EACH OPERATION CODE byte
There are also such, i.e. extended, although for extended CB, ED, DD and FD
another M1 cycle is produced. In normal mode, the command is retrieved in 4 seconds
tact, but then the picture changes. Whites, writing, reading - they also suck
bars.
The error on interrupt confirmation is ~50, but IORQ can be subtracted.
If we go further, then based on this we can determine in what modes
performance is measured by linking to a clock and counting useful and
lost cycles depending on who pulls the processor.
All circuits for debugging processor devices are built precisely on this principle. B
one, described in the "radio amateur", there is the possibility of step-by-step and cyclic
debugging devices on Z80. It's clear what's going on. In the future, you can hang
multichannel analyzer with a couple of dozen channels and already performs well
speed, enter the data into Excel, and analyze it there.
z.y. By the way, a phrase flashed here about eZ80 clocking, i.e. operations in one cycle
are they readable? Strange, that means I didn’t look at the datasheet properly.