Matrix Multiply slower on RT than PC?

[JeremyMarquis]

I was playing with large matrices and was testing execution speed on my laptop (the code is destined to run on RT). 

 

The code below tests using the Matrix Multiply function vs just straight array math.  Unsurprisingly the Matrix Multiply function executed faster.  Then when I ran it on RT both methods took the same amount of time? And Matrix Multiply (A x B.vi) takes LONGER on RT than PC?  Is there some windows-optimized code under the hood that is not available on RT?

 

I was hoping it would be faster since my laptop is several years old and this is NI's super duper 8135 PXI controller. I did deploy this as a startup app to ensure running from the IDE did not affect timing when running on the RT.

 

Edited November 17, 2014 by JeremyMarquis

[Darin]

I do not trust individual benchmarks of two operations in parallel.  Especially when RT gets involved.  Try serializing the two functions and see what the timing difference is then.

[crossrulz]

cross post on the dark side: http://forums.ni.com/t5/LabVIEW/Matrix-Multiply-slower-on-RT-than-PC/m-p/3053426#U3053426

[JeremyMarquis]

You are correct sir, I shouldn't have slapped that code together so fast.  The fun and danger of rapid prototyping in LabVIEW.

 

All results (both methods and both platforms) now report ~4.3ms.  I sure wish I could have that <2ms back...

 

Edited November 17, 2014 by JeremyMarquis