Re: [Discuss-gnuradio] dummy block has huge impact on usability fir filter

[Martin Dvh]

Eric Blossom wrote:
> On Mon, Jan 17, 2005 at 01:32:48AM +0100, Martin Dvh wrote:
>
> > > Can you please update gnuradio-core 
> > > (gnuradio-core/src/lib/runtime/gr_single_threaded_scheduler.cc is the
> > > file that matters) and try your test with and without the dummy block?
> >
> > I am compiling the new version right now. I will let you now the results as 
> > soon as possible.
extensive results at:
http://www.olifantasia.com/projects/gnuradio/perftest/17012005/

file used to generate results:
http://www.olifantasia.com/projects/gnuradio/perftest/17012005/mdvh13_am_perftest_tofile.py

log of the session:
http://www.olifantasia.com/projects/gnuradio/perftest/17012005/mdvh13_am_perftest_tofile_17012005.log

generated output files:
http://www.olifantasia.com/projects/gnuradio/perftest/17012005/am_perftest_audiorateis32127_transbandis????_usedummyis????_out_float.raw

What I did was use my old driver which requires a high output_multiple.
I did an am_demodulation (like before) of a 5.0225 Mhz am modulated signal from 
a signal generator (modulation is a sinewave of I think 1 kHz)

If I use the dummy block the results are as follows.
Filter narrower as 2436.2 Hz gives a number of taps above 16015 and I have a 
silent output signal(like before).
Filter between 2436.2 and 4000 Hz does give output but is sounds strange. It is not a nice sine wave but sounds like the filter is resonating. 
If I stop the input it takes a while for the sound to fade away.
maybe this is normal for very narrow filters.
This behaviour I already get some time now but I think in the older cvs versions ( more then a month ago) the result was different. I am not 
sure about this. (I do have an older cvs version around so I may try to test this if I find the time)

with filters wider then 4000 Hz the output-signal starts to get normal.

If I do NOT use the dummy block I always get some output.
with trans_width wider then 4000 Hz the output starts to get normal.
below 4000 hz The sound is an intermittand sinewave.
It sounds like morse-code.
beep---------------beep-beep-beep----------------beep-beep-beep-------------------
The narrower the filter, the longer the silence between the beeps.

If I use a transition width of 300 Hz the beeps are dozens of seconds apart.

you can listen to all output files.
I made a float filesink (audio_rate is 32127)
http://www.olifantasia.com/projects/gnuradio/perftest/17012005/am_perftest_*.raw
If you use it as a float filesource and connect it to dsp you can hear all 
signals.
the audio_rate, transition bandwidth and wether or not dummy_block is used is 
in the filename.

you can read the number_of_taps of every test in the log:
http://www.olifantasia.com/projects/gnuradio/perftest/17012005/mdvh13_am_perftest_tofile_17012005.log

Greetings,
Martin