Why does converting a SGL to SGL get rid of coersion dots?

[Ben]

" It must be a good one as I have never seen an instance where a DBL was misconverted into a SGL losing higher byte information."

I agree!

I wonder what it is.

Ben

[Ben]

I have been told I will get an answer to this riddle.

When I recieve same, I will post here.

Ben


Ben Rayner
Certified LabVIEW Developer
www.DSAutomation.com

[Ben]

Here is an update I recieved this AM.

"Ben,

I spoke with the LabVIEW PSE's today and explained the situation. I used
your example VI to illustrate the behavior and it was decided that
this required additional investigation. It was suggested that I
file this officially as a bug and then see what the owning developer
has to say about this behavior. I will put a link to your original
thread in the report with instructions to update the thread with
additional information or to notify me and I will do so. Either way,
I will let you know what I find out.

"

Sharing what I know,

Ben


Ben Rayner
Certified LabVIEW Developer
www.DSAutomation.com

[Ben]

I ran across this ini file setting and had to try it out on this example.

defprecision=Single

Sure enough, if you re-start LV with this setting then disconnect the lower input feeding the subtract node and then re-connect it,
BINGO
It is now a Single.

So...

any ideas what the bottom line is?

I'll post if the light suddenly goes on.

Ben

[Ben]

Well changing the setting and re-opening LV is not enough to fix this issue.

Since the wire has to be dis-connected and then re-connected to get rid of the coercion, this has to be an issue with what LV is doing at edit time.

So...


Ben

[Jeremy_C]

The algorithm for determining resulting data types gets much more complicated when using uninitialized shift registers. The algorithm used by the binary arithmetic functions is slightly different than the algorithm used by the compound arithmetic function. It was determined that the algorithm used by the compound arithmetic function is more robust than that used by the binary arithmetic functions. Modification to the algorithm used by the binary arithmetic functions is being considered for future versions of LabVIEW.