Re: [SpectrumLabUsers] Decimation help required
- Short update:
The problem was indeed that the 'input preprocessor' did not divide the
sampling rate by the decimation ratio (for the following processing
stages). Fixed in the next build (V2.79 b3) which is currently being
uploaded to the known location.
Not solved yet: When changing the input decimation on the fly, SL
doesn't immediately adapt all subsequent stages to the new sampling rate.
- Wolf said:
> In fact, decimating the input by two should also divide the FFT binMy apologies. Wolf is of course correct. I claim a senior moment. Reducing the sample rate by 2 while keeping the same FFT size will result in half the bin width.
> width by two (not multiply by two).