GH performance improvement and comparison

I was just collaborating (Windows and Mac) on a script, when I realized some pretty different calculation speeds with the GH profiler…
So I quickly made a simple intersection script (find it attached below) to make some comparisons. There are good and bad news:

The bad: GH on Mac seems to be much slower than GH for Windows
The good: On both Windows and Mac GH in RH7 seems to be +25% faster!

Here are some numbers:

GH (RH6) on Mac: 2.0s
GH (RH6) on Win: 0.7s

GH (RH7) on Mac: 1.6s
RH (RH7) on Win: 0.5s

Is there something specific with multithreaded components maybe? Is there something that can be done on the Mac version to speed things up?

Otherwise keep up the good work guys and stay healthy!!!

Cheers, Rudi (8.3 KB)

What kind of numbers do you get if you turn off the preview? We are waiting for a service release to Mac OS that should help with performance for wire drawing. All we can really do at this point is wait until Apple ships this to start getting users to test things out.

Hey, thank you for the fast reply!
In my experience it didn’t change anything… maybe like 0.1s, but that may be because recomputing the definition gives slightly different computing times…

I wrote a reply too quickly and noticed that you were looking at the solve time on the components themselves. Display performance shouldn’t make a difference for that number, sorry.