sprintf/fprintf behaviour

[BLL]

Hi, I did use a scope on there to get the amplitude and the pulses were fine.
Since my last post, it suddenly put itself right. I can only presume it was a bad connection on the prototype board. I wiggled everything afterwards but it is still fine, so I am not sure!
The main thing is that it's counting the pulses from DO correctly.
Thanks all for your help.
Brian

[Ttelmah]

The key point is that the hardware will count what is there....

It may well be that the same bad connection was the core of your trouble. The point is that the pulse interrupt you were using, introduces about a 60 instruction 'overhead' every time it is called. If it only happened once in a single character (as it should), then your serial should still be acceptable. However happen more often than this, and the result will be garbage, which is what you were seeing.

[BLL]

You are probably right - I don't like prototype boards as I have had trouble before with them, but short of wasting a number of pcbs, one doesn't have much choice.
The timer1 method is much preferable to my interrupt method, but as said before, I didn't know you could do that, but I do now!
Thanks again for your help.

Regards

Brian