I have to say I am more than eager to try out the new 28X2 and especially the big surprise, the 20X2. Two questions beforehand:
(1) the Basic manual states that the timing resolution of the pauseus command is 10 us at 4 MHz for the X1 parts, but 10us at 8 MHz for the X2 parts. So I assume the minimum achievable resolution on the 20X2 when running at 64 MHz is 1.25 us, correct? Now the manual does not indicate such a difference between X1 and X2 devices for the pulsout command. Is that a mistake, or is the pulsout resolution on the 20X2 really 10 us at 4 MHz and thus 0.625 us at 64 MHz?
(2) the manual says the hardware interrupts on the X2 parts is not a polled interrupt. Does that mean it will predictably call the interrupt routine with no more than 1 instruction cycle of jitter? The "old style" interrupts had a lot of jitter because the polling period is rather long, meaning that the reaction time to an interrupt could vary considerably (I do not care about the absolute reaction time, but just about the maximum variation). I tried hardware interrupts on a "real" PIC16F88 at 20 MHz and indeed in that case the interrupt routine got called with a jitter of just 0.2us.
Wolfgang
(1) the Basic manual states that the timing resolution of the pauseus command is 10 us at 4 MHz for the X1 parts, but 10us at 8 MHz for the X2 parts. So I assume the minimum achievable resolution on the 20X2 when running at 64 MHz is 1.25 us, correct? Now the manual does not indicate such a difference between X1 and X2 devices for the pulsout command. Is that a mistake, or is the pulsout resolution on the 20X2 really 10 us at 4 MHz and thus 0.625 us at 64 MHz?
(2) the manual says the hardware interrupts on the X2 parts is not a polled interrupt. Does that mean it will predictably call the interrupt routine with no more than 1 instruction cycle of jitter? The "old style" interrupts had a lot of jitter because the polling period is rather long, meaning that the reaction time to an interrupt could vary considerably (I do not care about the absolute reaction time, but just about the maximum variation). I tried hardware interrupts on a "real" PIC16F88 at 20 MHz and indeed in that case the interrupt routine got called with a jitter of just 0.2us.
Wolfgang