While integrating FreeDV 700D I found a bug in FreeDV 1600 - it was
losing sync and inserting bit errors when there were timing slips due to
sample clock offsets.
Net effect was occasional bloops in the audio/loss of sync.
I've fixed the bug now, FreeDV GUI is solid when I decode files with
simulated sample clock offsets of +/- 10Hz (1000ppm).
This fix will find it's way into the FreeDV release later this month.
We should probably rebuild the SM1000 firmware too.
- David
On 03/05/18 06:08, Steve wrote:
I'm going to hazard a guess, that the dft() algorithm takes a big chunk
of the CPU.
It's nice because it doesn't have to be a power of 2, and works with the
2-dimensions.
For the DFT it takes 144 complex time samples and produces 19 complex
frequency samples.
The IDFT the opposite.
That's 144 x 19 = 2736 complex multiplications and additions each time
it is called.
KISS FFT is not a good candidate. FFTW probably is.
Steve
On Wed, May 2, 2018 at 4:18 AM, Alan Beard <bear...@unixservice.com.au
<mailto:bear...@unixservice.com.au>> wrote:
700D is a bit CPU hungry:
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Freetel-codec2 mailing list
Freetel-codec2@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freetel-codec2
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Freetel-codec2 mailing list
Freetel-codec2@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freetel-codec2
