int_ext and time required to detect falling/rising edge

[evsource]

Assuming a 10mHz 16F88. Since I know PCM Programmer will ask (), I'm using PCM 4.032.

When using #int_ext, what would the constraints be to detect an incoming pulse (either rising or falling edge, whatever the interrupt was set up for). For example, detecting a falling edge. Would it matter how long the pulse stayed low? If there was a falling edge, but 1us later, the signal went high again, would it still fire the interrupt?

I know I could test it empirically, but for future reference of others (and to save me time testing it out), I thought it would be worth asking here. I did search the forum for the answer already.

[PCM programmer]

It's normally given at the end of this table, in the Electrical Specifications
section of the data sheet:

[FvM]

Some interrupts sources can operate also in sleep mode. In this case, the respective part of the interrupt logic can't use a system clock source and works asynchronously, reacting faster than an instruction cycle.

[Ttelmah]

You'll find the 'short' time (25nSec typically, - but not tested), is present on all chips that support 'clock switching'. Makes sense if you think about it, since otherwise the interrupt detection time would change when you switched clocks.

Best Wishes