Slow performance of "Polar to Complex"

[D*]

The speed boost is only when using LabVIEW 64-bit. I'm using 2018 18.0f2 (64-bit).

[Ben]

---

@D* wrote:

Benchmark code attached.

Versus

---

Not showing us where the data is coming from makes me question if the is constant or structure folding going on. The array and the associated math could be getting precompiled into a constant...

Ben

[Bob_Schor]

---

@D* wrote:

@Bob : The Random number generation can be expensive.  In this case, it might even be comparable to what we are trying to measure.  Perhaps see how much of a performance boost if either :

---

Thanks for the thorough explanations!  I continue to "learn" in the Forums.

Bob Schor

[D*]

Ben, I've attached the benchmark code again.  Please run using 64-bit LabVIEW.

[alexderjuengere]

---

@GerdW wrote:

Hi,

for me PolarToComplex and SinCos are nearly the same speed:

[...]

(Intel i7-4700HQ)

---

@alexderjuengere wrote:

"Polar to complex" appears to be a bit slower than "Sine and Cosine"

60..65 ms in comparison to 55..60 ms

with LabView 2018 x32 on Windows 10 x64:

---

interesting, I am running on a Intel i5-5200

[GerdW]

Hi Alex,

one more run:

The VI is attached…

(Running LV2018 32bit, on Win8.1, i7-4700HQ)

[alexderjuengere]

---

@GerdW wrote:

Hi Alex,

one more run:

The VI is attached… Test polar to complex.vi ‏13 KB

---

I will run the .vi from above at home,

there I have an i7-4700HQ machine, labview x32, windows 10 x64

I'm currious, if I can reproduce your (Gerd's) result

[D* · Accepted solution]

You must run using Windows x64 AND LabVIEW 64-bit to see the difference.

Using Windows x64 and LabVIEW 32-bit does not show a difference. 

[Ben]

I did a quick Google and stumbled across some deep discussion involve the representation of Pi for a 64-bit machine along with some connection with the old 487 chip that used to be used for floating point math and the like.

So this could be a case where the Sin and Cos got a boost from 64-bit math while the polar did not get some needed attention.

For what it is worth...

Ben