On those single board computers I don't bother writing to SD. I just route it through the network (assuming large enough bandwidth) to another graph or Python script that dumps it to a disk with faster storage bus.
You can use gr-zmqblocks for that. On Wed, Aug 6, 2014 at 11:51 PM, Marcus Müller <marcus.muel...@ettus.com> wrote: > The overflow is not a result of buffering; in fact, buffering helps. > > overflows happen because at any bottleneck in your processing chain, > samples start to stagnate, causing back pressure. When all buffers are > filled, there's nothing the rtl source can do but drop samples. That's > an overflow. > > A computationally intense filter might also lead to overflows, but here > it's because your CPU can't keep up with the sample rate, whereas with > the file sink, your storage can't sustain the write rates. > > Greetings, > Marcus > > On 06.08.2014 15:44, rejunte wrote: > > I tried what Vanush said and just connected RTL-SDR Source to a Null > Sink and > > got no overflows. > > > > I'm using a class 10 sd card, that must be the problem. > > > > I did some demodulation with no problem, but when I add a Low Pass > Filter, > > the overflow happens again. > > > > Would the filter do the same thing as the file sink and buffer the data > > causing the overflow? > > > > > > > > -- > > View this message in context: > http://gnuradio.4.n7.nabble.com/gnuradio-on-ubuntu-touch-tp49315p49799.html > > Sent from the GnuRadio mailing list archive at Nabble.com. > > > > _______________________________________________ > > Discuss-gnuradio mailing list > > Discuss-gnuradio@gnu.org > > https://lists.gnu.org/mailman/listinfo/discuss-gnuradio > > > > _______________________________________________ > Discuss-gnuradio mailing list > Discuss-gnuradio@gnu.org > https://lists.gnu.org/mailman/listinfo/discuss-gnuradio >
_______________________________________________ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
