On 25/02/2021 23:17, Steve Walker wrote:
On 25/02/2021 22:50, Roger Hayter wrote:
On 25 Feb 2021 at 20:10:32 GMT, "GB" wrote:

The Humax remote stopped working, and I was just about to throw it away
and buy another when I thought I would google the problem. Who would
have thought that you can reboot a remote control?

Maybe, you guys all knew that, but it surprised me.

It doesn't totally surprise me, but it's a bit sad.Â* It should be
possible to
make a remote control without any operating system complex enough to
lock up.
What is the name for a system all of whose possible states are
defined?Â* That
is surely a reasonable ambition for a remote control?

Yes, there was certainly a chip back in the '90s (it might have been
called Viper) where every combination had been thoroughly tested and it
was known that there were no hardware design faults. I'm sure such chips

That was certainly the marketing spiel. But it all ended acrimoniously
and in tears. This paper predates the point where things started to go
wrong. But makes a clear distinction between what was formally proven to
be correct and what was asserted to be proven correct.

https://www.cl.cam.ac.uk/archive/mjc...ProofPaper.pdf

must exist today and, if so, the simple code required for something like
a remote should also be able to be exhaustively tested (automated
testing) and provide a system that cannot fault or lock up - ignoring
hardware failure.

It is often hardware failure or CPU glitches that catch you out. It is
hard for a CPU that finds its program counter memory mapped into ROM!
TI99xx series had all of its registers including the PC in ram.

--
Regards,
Martin Brown