Reliable source for NMOS Z80? Z0840004PSC really/alwways supports out(c),0?

[eto]

i wonder if this could be checked for and fixed in software

of course - but the point of using the undocumented opcode is that it is faster than using the documented opcodes.

I also don't think it's a big issue. The CPU does not die too often and I still can get NMOS CPUs. It's just annoying that I can't rely on any seller and I definitely have to check a CPU immediately when I receive it.

[1024MAK]

As well as Zilog, the NMOS Z80 was officially second sourced by other companies. Including but not limited to those from Mostek (MK3880), SGS-Thomson (Z8400) and Sharp (LH0080).

Toshiba made a CMOS-version, the TMPZ84C00, which may be the same design as the Zilog CMOS Z84C00.

The NEC NMOS version, the μPD780C-1 however was not an official second source, but a design that was manufactured after NEC reversed engineered a Z80. It is highly compatible.

In the USSR there were other unofficial clones.

Unfortunately, many Z80 NMOS chips that are available now are 'pulls' and many get cleaned up and remarked. So you don't know which manufacturers part you are getting or which technology it will be...

The same goes for some CMOS chips that are available from marketplaces where a 20MHz part may be a lower speed NMOS part.

The only way round this for NMOS parts, is to buy parts with older and realistic date codes (e.g. date codes starting with 80 through to 89) from suppliers that are not in the far east.

This problem is likely to get worse now since the production of new Zilog CMOS chips ended last year.

Mark