It's easy enough to pull out a section of a regular max patch, and run it 100k times in succession as a benchmark. But I'm not sure how to do this kind of thing with MSP patches. Is there a more useful way to judge the system usage of bits of MSP patches than just looking at the CPU meter?

finding-bottlenecks-in-msp-code

Typically you could stick a portion into a poly patch and load it 200 times. Change one object and see the difference in cpu load. Still it's the cpu, but you get a better idea of the average load.

I would like to try using a profiler...Any suggestions for a good one for Windows? I'm about to give Microsoft's Xperf a go.

Finding bottlenecks in MSP code