1298Re: [SpectrumLabUsers] Re: 2.78 B26 scroll dropouts
- Feb 7, 2013Hello Tony,
Thanks for the image. The problem may be related to the time source itself, which was one of the changes between V2.75 and V2.78. The older versions retrieved data from a buffer (to feed the FFTs for the spectrogram) based on the number of samples, while the newer versions use the precise timestamps. If there's a gap in the timestamps, there's a gap in the spectrogram. This may be caused by the configurable 'time source', even though someone mentioned having tested all options but none of them cured it.
Anyway, if the soundcard's sampling rate is too far off, the timestamps (when based on the sampling clock) may be off too far after the program was running for some time. If the difference between the sample-based time, and the PC's system time exceeds a certain value the timing resets automatically (usually with an entry in the error log). I will try myself by deliberately entering a 'very wrong' sampling rate in the SR calibration table, and see what happens.
Am 07.02.2013 02:42, schrieb mux_folder2001:
Referring to file in folder "TonyC" called dropout.jpg:
I have uploaded a side by side of 2.78 and 2.75 listening to the same NDB. 2.78 shows a dropout at about the 44 Second tick mark while 2.75 does not. You can also see the timestamps pushed together. The beacon is TAD which isn't displaying too well because of the contrast being set too high.
- << Previous post in topic Next post in topic >>