Changes
No 5V pin go to the pico. We use some 1 buffer, 1 level shifter, 3 multiplexers. We write to the CPC with 3.3V, we use 5V to the CPC only to drive some signal like A15 for C3 memory mode and IOReq to disable intern floppy.
We prefer to have somerthing between the Pico and the CPC, as even if it work now, there is no guarantee that the boards will work after some years of intensive usage.
It is really better to be carefull.
The board can decode any I/O port and of course RAM/ROM cycles. As it use a Pico2, its RAM is doubled vs Pico1 and it can access a 8MB PSRAM as well.
The PicoCPC is clocked at 366Mhz. So it is not as powerful as the [[RSF3]] clocked at 480Mhz. But it is more powerful than the [[M4 Board]] clocked at 168Mhz. [https://www.spinpoint.org/2015/10/23/cortex-m4-board/ Source]
<br>
== Multiplexer ==
The "Secret" of the PicoMEM/PicoCPC is its multiplexers. [https://www.cpcwiki.eu/forum/amstrad-cpc-hardware/picocpc/msg249997/#msg249997 Source]
It allow to limint the number of I/O needed and act as a Voltage converter at the same time.
The Pi Pico allow to have multiple PIO core, plus the main core working together in parallel,so the PIO manage the Multiplexer, reconstruct the Address, while CPU core detect and manage the Control signals, Read in advance the High part of the address to jupt to the correct code.
(IO MEM Read or Write, Bank number)
So, when the Main core is in the correct code section, it can read directly the full Address from the PIO.
The PicoCPC use only 12 I/O to manage all the address, Data and Control signals. We can add 2 pin only to control the multiplexer.
Then, the CPC bus is "manipulated" Via ROMDIS, RAMDIS, RD, A15, IOREQ (Pico Output)
With this way to do, it is possible to be almost as fast with multiplexing that without.
<br>
