No, not really. BenchmarkNET will compile new binaries and inject code into those. So it doesn’t support in-place measurement.
So…? You can load any binary into Rhino and execute them. This already might be enough to bypass the McNeel DRM. Again, I don’t know if this is discussion is really helpful or if it might work or not.
I see this topic much broader. If you want to know how parallel programming works, you don’t need to limit yourself to Rhino. If you understand these concepts, you can apply them anywhere. Again, my point is to point out that you need to identify the cause of a performance problem and not apply just any hack you find on the internet as a premature optimization.
Many of the code snippets in this forum wouldn’t pass any code review within a software company, because only few people really pay attention in explaining their code properly. If you run into a racing condition because of improper programming, people waste tons of hours in debugging, just because they can say “Well, at least my script runs in parallel and so 250 ms faster”. The same idiocracy you see when people code a simplistic app in 3 different programming languages just to squeeze out some milliseconds for some garbage-problem, and 2 years later someone finds a bug, but is not capable of fixing this, because that person just knows 2 out of these 3 languages…
Apart from that, the example shows, that with a proper selection of an algorithm, you can outperform any of these micro-optimizations.
And last but not least, many of the performance critical use-cases within the domain of Rhino do exist, because so many people do not pay attention to manage and to model their data with care.
Architectural plans with 500 layers, technical parts with thousands of unnessary heavy surfaces or just another differential-grown-Voronoi pattern are producing trillions of unnecessary computations every day. But instead of questioning this nonesense, people are even more eager to increase the big bulk of garbarge by buying a 100 core working station with two $10k graphic cards… Sure, maybe my last point is subjective, but I have seen this way too often
1 Like