compiling to blackfin

[noahgrant]

wow, I'm really sorry to add a third message, but I forgot to mention that in test programs, along with the phase change for some signals, using an inverse FFT resulted in a loss of amplitude (from about 1 to on the order of 10^-5.  that's a lot!).  but when the test program was played on my computer, the phase change was not noticeable from the output signal, and neither was the magnitude change!  the output played at the same volume as in the input, despite the waveform chart on the front panel showing an amplitude five orders of magnitude less!  can you help explain that to me?  what can I do, will it matter coming out of the Blackfin and into an amplifier?  

Thank you, Kevin, yet again...

Noah 

[Kevin_S]

Hi noahgrant,

I'm a great resource with LabVIEW for Blackfin and the LabVIEW environment, but when it comes to a specific algorithm being implemented in LabVIEW, I'm at a bit of a loss.  Perhaps individuals from the community here can comment on your code and its usability (especially if there is a similar setup out there).

As far as determining phase lag of a signal being generated by the Blackfin processor, I would propose a simple test.  Use an oscilloscope or other measuring device to measure the input and output signals in order to put a specific time value to the lag.  (The lag would be dependent on the amount of processing code being run, and it would be unique to your code and perhaps even the signal being input.)

I don't know why your signal is different on the Blackfin versus your computer.  There is one test that you can perform.  Monitor the actual signal (in software) that is being sent to your speaker (on PC) or your output jack (on Blackfin).  Compare these values.  If they're the same, further time can be spent investigating the datasheet on the specific digital to analog converter being used by the Blackfin.  If they're different, further time can be spent investigating the difference between the analysis functions on your PC versus the Blackfin.

Kevin S.

Applications Engineer

National Instruments

[noahgrant]

thanks kevin!