130nm technology, wow. I don’t see how FOSS Z80 can work at 5V. It is likely 3.3V or even 2.5V and 100Mhz class processor.
Bill
You still have to keep 5 volt IO. I am sure you can push back from the bleeding edge
of techology a few generations back to get that.
I really don’t know what I’m talking about (I don’t even know the basics of chip fabrication), but out of curiosity I tried to google this. It seem’s there are several “5V native” components and the definition of “high voltage” (where degradation is expected) starts at 7.3V for NMOS and 8.1 for PMOS, I’m also under the impression that the genaral idea about speed is below 100 Mhz, more in the below-20MHz range, which should fit the Z80.
https://skywater-pdk.readthedocs.io/en/main/rules/device-details.html
Last time I did IC design was 0.35um 20 years ago, and it was already difficult to handle 5V. It looks like 130nm core is 1.8V and IO pads can do 5V. So a reasonable approach is an internal regulator to bring core voltage to 1.8V, then translate voltage to 5V IO. It is slow, but 1.8V is so fast, it can afford the propagation delays.
Bill
Does anyone know if Sharp still manufacture their licensed LH0080 clone? People seem to be selling new ones on Ebay.
I know they were developing a Z80 on glass about 20 years ago.
There are a bunch of clones in verilog or vhdl (eg T80). It’s a question of production. Gate arrays would probably be an option. That was how the Rabbit 2000 and 3000 were manufactured and where a sort of “super z180”.
I don’t think btw that the termination of the Z80 was about revenue. From the announcements it looks more like the fabs they used were discontinuing the process so they’d have had to rework it for a modern process which would not be trivial.