After my post on HPCToolkit, I felt that I prefered QCacheGrind as a GUI to explore profiling results. So here is a gist with a Python script to convert XML HPCToolkit experiments to callgrind format: https://gist.github.com/mbrucher/6cad31e38beca770523b

For instance, this is a display of an Audio Toolkit test of L2 cache misses:

ATK L2 cache misses profile
ATK L2 cache misses profile

Enjoy!

I’d like to talk a little bit about the way a compressor and an expander can be written with the Audio Toolkit. Even if both effects use far less filters than the SD1 emulation, they still implement more than just one filter in the pipeline, contrary to a fixed delay line (another audio plugin that I released with the compressor and the expander).

Read More

Yes, because Cover Trees are sometimes too slow. In fact, I asked myself this question, not for the build time, but for the search time if the data has a structure. Imagine, what would happen if your data was more a less a regular grid? When I tried that, starting with a point at (0,0), then (1,0)… the first node (0,0) had references to all the last points (9,9), (9,8)… And I figured, it would be slower than a tree search. So I decided to give kd-trees a shot for this kind of search on a regular grid.

Read More

After presenting Valgrind as an emulation profiler, I will present Microsoft solution, Visual Studio Performance Tool. It is available in the Team Suite editions, and offers a sampling- and an instrumentation-based profiler. Of course, it is embedded in Visual Studio IDE and accessible from a solution.

Read More