I just started using my SRM for measuring standing starts and it doesn't measure power and cadence until after about one second after speed starts up. Is there a setting I can use that will make it measure that first second, is that an artifact of the autostart setting I use or the measurement method SRM uses? I have it set to record at half second intervals.

Unfortunately, I don't believe that there is anything you can do to overcome this limitation. Power is calculated as torque times angular velocity, with the former averaged over a complete pedal revolution and the latter based on the time required for that revolution to be completed. Thus, even if the crank is "awake" when you begin your effort, you won't get a power value until the end of the next full pedal revolution...as you've seen, this takes a while, especially if you're using a big gear (and have a puny maximal power, like me! ). The speed, OTOH, is recorded as increasing sooner, because the magnet on the wheel passes the reed switch on the fork more frequently (unless, of course, for some reason you're doing standing starts in a gear ratio of <1:1, e.g., on a mountain bike).

There is one trick I can pass along, which is this: if you set the recording interval to be significantly shorter than the time required to complete each pedal revolution (e.g., at 0.1 s intervals), you'll actually end up recording the data for each revolution, not simply as a function of time. This is so because the SRM always records the calculated power for full pedal revolutions...so if another full revolution hasn't occured since the last time data were written to memory, it will write the same data again. Sometimes using this trick can be of interest/value, although it will tend to fill up the memory very quickly.

Thanks for the complete answer. I was wondering about that because that sample SRM kilo file has the same problem but the sample pursuit file does not, but it appears as though they used a rolling start for the sample pursuit file.