At the very least the PIA and VIA datasheets are wrong about Voh; they are proper CMOS devices with totem poles that can drive rail to rail with an impedance of about 50 ohms (ie. at least as hard as 74HC). The exception is that on the -N devices, the GPIO pins are built as NMOS constructs, specifically for maximum compatibility with existing applications. On the -S devices, the GPIO pins are also strong CMOS totem poles, with the addition of "pin keeper" devices.

I think we also have documented cases of a W65C02S being simply plugged into an existing vintage machine with 74LS glue logic and NMOS memories, with the only "mod" required being to bend up the /VP pin which is a GND on the original NMOS devices. Technically one should also ensure RDY has an explicit pull-up on it. That would indicate that most if not all of its inputs are actually TTL compatible, even if they are specified as CMOS inputs, at least at modest clock speeds (NMOS 6502s went up to 3.5MHz maximum, most vintage machines ran them much slower than that for various reasons).

However, it is safest to assume that CMOS level inputs are required, especially if you are aiming for relatively high speeds, and to ensure that your clock signal is strong and clean with fast edges, and to make sure you have enough power supply bypassing near each IC's power pins.

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

I'm actually not concerned with the output stage of the WDC devices.

I need to know whether the input stage is "LVTTL" compatible or not. Being LVTTL would allow easy interfacing of 5V 65xx devices with 3.3V CMOS glue logic. It would be unfortunate, also for WDC, to have not thought of that.

_________________
Author of the GeckOS multitasking operating system, the usb65 stack, designer of the Micro-PET and many more 6502 content: http://6502.org/users/andre/

Again, can't you just run the WDC parts at 3.3V? They are specified for that, and that would save you an awful lot of hassle. That's what I meant by treating voltage level shifting as a distinct function, and putting it in only one place.

I could. But all of the "externally visible" I/O needs to have 5V: IEEE488. Userport. Tape Port. And those are directly connected to the VIA and PIA, and you cannot know which direction (in or out) some smart designer has used on which pin.

So, VIA/PIA need 5V -> CPU needs 5V -> I need a voltage converter CPLD -> WDC.

Fortunately it is only Phi2, and the data bus that needs that. Phi2 can work with the method I found in the Xilinx application note, the databus can handled by a 74HCT245 (TTL inputs can be driven by both 5V and 3.3V CMOS, and CMOS output to 5V from the rail is acceptable for the CPLD). So I think this is the least costly way. And in fact "in only one place"

RAM will probably then be 3.3V as well (as it's only connected to the CPLD) - 5V ROM needs TTL inputs, as upper address lines are driven by CPLD directly.
(But I may replace the parallel ROM with an SPI Flash at some point in time).

André

_________________
Author of the GeckOS multitasking operating system, the usb65 stack, designer of the Micro-PET and many more 6502 content: http://6502.org/users/andre/

Anyway, I have used the contact form on the WDC web site to ask about the datasheet inconsistency, will have to see if/when I get a response.

_________________
Author of the GeckOS multitasking operating system, the usb65 stack, designer of the Micro-PET and many more 6502 content: http://6502.org/users/andre/

This whole level-shifter from 3.3 to 5v and back again has been a "solved-thing" in the Raspberry Pi world for a good number of years now. One way is easy, but bi-directional isn't much harder - A quick check from my local hobby electronics store finds:

https://shop.pimoroni.com/products/spar ... ut-txb0104

but there are many many other devices.

-Gordon

_________________
--
Gordon Henderson.
See my Ruby 6502 and 65816 SBC projects here: https://projects.drogon.net/ruby/