Re: How (does?) SuperCPU 64 control CPU port?

From: silverdr_at_srebrnysen.com
Date: Sun, 23 Apr 2023 10:36:00 +0000
Message-Id: <D29C05EC-70E0-46B0-AAAD-7C681E28C3DF_at_srebrnysen.com>
> On 23 Apr 2023, at 09:32, Maciej Witkowiak <ytm_at_elysium.pl> wrote:
> 
> Hi
> 
> I'm playing with a project that replaces 6510 chip in a C64 with a Teensy 4.1 board that emulates the CPU:
> 
> https://github.com/ytmytm/teensy64

đź‘Ť!

> The emulation can be a lot faster than a real 6510, so I plan to put in a SuperCPU compatibility mode eventually.
> The next logical step could be to drop the idea of replacing onboard 6510 and just put everything on a cartridge.

Not sure if it's so "logical". Speaking for myself I'd prefer having EXPANSION port open for other stuff, while CPU has its place on the mainboard

> This got me thinking: how does a real SCPU handle access to $01 port?
> 
> I guess that SuperCPU's CPLD can track the writes to $01

Correct me if I am wrong but AFAIR (?) $00 / $01 writes do not show on the bus, do they? Although I _think_ there were some tricks to catch them? Can anyone remind me how that works(/ed) if at all?

Sure the '816 could run in parallel in order to be able to pick them. Having in mind that it has a different timing (and behaviour) of some instructions that doesn't seem feasible though. Or am I missing something?

> But what about the tape - does that work with SCPU? I don't see any way of controlling tape from the cartridge port alone, it has to come from 6510 $01 port.

W/O H/W mod controlling tape's not going to happen ;-)

> Or perhaps SCPU works using a different principle and the onboard 6510 is sometimes running?

Tricky, at best… I can imagine that being theoretically possible but to me that's just one more reason why it's not so "logical" to put it into EXPANSION rather than CPU socket. The only – albeit important one – upside of cart-based approach for me is that many boards have CPU soldered-in
Received on 2023-04-23 13:00:22

Archive generated by hypermail 2.3.0.