turbo-ZX

Immediately click on RESET, Alexander!

27 Oct 99 13:10, Alexander Kormishin wrote to Kirill Frolov:

KF>> You don’t have to be a rocket scientist to understand that
KF>> It’s more efficient to play with the processor clock frequency, and WAIT can be done
AK> Not at all. When you need to wait for access to memory, for theory it is the same as
AK> the processor slows down there - clocked or WAIT'om. In practice,
AK> it is easier to manage WAIT, which means it is more reliable and economical.

You forgot one thing - after reading the COP you will have 1/2 clock cycle after
removing WAIT, and after reading the data 1 clock cycle. Because If your circuit is capable of working at 1/2
of a clock cycle, then there is no point in losing
1/2 of a clock cycle when reading data. You can immediately bring the processor to the point after which it
will have to read data from the data bus and remove RD and MREQ and
slow down the clock frequency until the memory is ready. Gain --
1/2 bar on each reading. Only it may not exist if the memory
is synchronized with output to the screen. But there may be some use on the curb. Here you already need to look after two calls to the same
page of RAM and you can also save on half a clock cycle of the RAS signal.

KF>> to gain a little in speed... IMHO this is how you can do 7 MHz completely
KF>> without brakes on the handles. But why?
AK> Indeed, such feats are not needed.

They are necessary,but with crooked hands and petty logic it’s almost impossible.

KF>> Turbocharge computers assembled on h.z. whatever
AK> [...]
KF>> Leningrad will never work normally.

AK> It doesn’t have to. All these computers, neither the circuit nor the board, are intended for modifications. But the user cares little about these fears and
AK> he will still modify his machine if there is even the slightest
AK> hope of success.

There is little hope. It still won't work then.

Kirill Frolov. [ZX]