>>>something external pulls it low [I2C clock works like this].  I was
suggesting
>>>my protocol in the context of people using a device like the 16C54 which has
>>>none of the above, talking to a PC printer port which has none of the above.
>>
>> Sorry again John but I'm really lost now!
>
>John, I think your 2-wire bidirectional protocol idea is very
>good. Asynchronous protocols are tricky, and I'm not sure I understand
>all details, but it certainly looks like good and robust
>engineering. Incidentally, I have seen a proprietary protocol invented
>by Sony for high speed communication between a PC and their
>DataDiscman, using similar principles. Keep up the good work!
>
cut cut......
>
>I don't want to pick on Parallax, because they make good products, and
>I think they have a good company policy in general. However, I have
>pointed out this problem to them and essentially received a short-cut
>answer like "Tough luck for you - We can't test all PCs." There is no
>question Parallax has been aware of the time dependence problem. A
>good protocol for high-speed communication with a PC must necessarily
>be time independent.
>
>Cheers,
>
> -- Martin
>
>

If anyone is seriously interested in the long reply to this please email me
direct and I will send a copy or upload to the pic list if there is a number
of people asking.

Regards

Jim





-----------------------------------------------------------------
NEWFOUND ELECTRONICS,  Now available WARP-3
Programs all PICs, 3 textool sockets. can reduce erase time 40-50%
ISP port,  fast, All options. 24Cxx & 93Cxx serial eeprom read/program
bonus offer.  Price: TBA approx $100US

email newfound@ne.com.au
------------------------------------------------------------------