ruud.baltissen_at_abp.nl
Date: 2002-08-30 11:05:31
Hallo Nicolas,
> http://www.geocities.com/researchtriangle/forum/8070/galblast.htm
Danke, and still valid.
> Rossmöller 8MHz Z80 accellerator module for the C128,
That rings a bell indeed. I have played with the idea of using a faster Z80
as well. But a week ago I found two token-ring "things" using a 10 MHz Z80.
My idea any way was letting the Z80 run at its maximum speed and doing an
asynchronous access. This means that it is coupled to the adress bus using
to two 74ALS573's: one buffers the data towards the C128 [1], the other
towards the Z80. PHI2 clocks both latches.
The moment the z80 initializes a transfer, a circuit puts the Z80 in
waitstate and waits for the positive edge of PHI2. The negative edge
releases the Z80 again. The 573 makes sure that the read data is still
available for the Z80.
[1] A 541 towards the C128 would do as well but it is more easy to piggyback
two 573's on top of each other. Another idea would be using a 645
biderectional buffer. But it has more controlpins which needs more logic
then those of the two 573's together. But using a GAL or CPLD.....
Another idea is manipulating the clocksignal for the Z80. IIRC the Z80 needs
up to 5 cycles/instruction and only one is needed to do the actual
datatransfer. Doing 4 when PH2 is (L) and the last one completely for PHI2 =
(H) could do the trick. But that means a 8 MHz Z80 in a 1 MHz system. But I
don't have much confidence in this idea, intuition, no hard fact to proove
it.
--
___
/ __|__
/ / |_/ Groetjes, Ruud
\ \__|_\
\___| http://Ruud.C64.org
Message was sent through the cbm-hackers mailing list
Archive generated by hypermail 2.1.4.