Ok, so I have been reading all the posts about tachometers, anemometers, and such trying to track down "simpler" ways of getting live speed measurements without bogging down the mcu too much. I have several projects kicking around in my head which more or less boil down to having a sensor that puts out frequency that changes a decent bit in live time (ie not a nice slow 2-3s sweep but rather 50ms changes). anyhow, I thought a tachometer would be similar in that you would have a frequency or pulse width change that you need to be able to measure in close to real time. anyhow, after reading all of ralph and other's struggles with the yC internals an idea started forming but i am not sure where exactly it is going or how it would get to what I/we could use.

so here is my thought. If you were to send a prescaled output from the yC, say 57.6Khz to the input of an decently high speed AND gate and then use other input to the AND for the output of the sensor/unknown frequency, the gate should output a third "frequency" that is related to the relationship of the two inputs. (Ok, I realize there are rise/fall times and such for the gate, but some of the modern 74HS can handle 57.6K-ish speeds just fine.. )This sounds kinda like extra work, but I am thinking that this would "spread out" the the input values and from what my simulation shows is that you would get "bands" of output based on the ratio and timing of the two frequencies. These bands would be easier and faster for the yC to pick up and output more live time. i haven't had a chance to test this out, as I have no AND gates at this time, but I would like to once I get some in. so what do you folks think? Is this something that might work or just re-inventing the wheel as a lopsided rectangle??

Seems pretty reasonable, I saw many designs using AND gates to "divide frequencies". However, instead of making uC output 57.6KHz, I'd use a 555 timer IC, to free a pin on uC.