On Sat, Nov 23, 2013 at 11:02 AM, Paul B. Huter <paul.b.hu...@gmail.com> wrote: > Thanks, Tom. Just to clarify, the 1Msps comes from the sample rate of 50M > and the decimation of 50, correct? Also, is this going to record 10 seconds > of "real-time" data? I was having issues with real-time, due to the large > data stream, so will this record data from t=0 to t=10, just not in > real-time? > > Paul B. Huter
If you're not running in real time, then you're dropping samples. The way these devices work is that they drop packets of samples, which can be quite a lot of data. So you'll store in your file sink 10 seconds worth of data, but if you're dropping samples, it won't be 10 consecutive seconds of data. But I just realized something. We've been trying to help you out constructing your filters and everything, but why are you setting your input sample rate to 50 MHz only to down-convert in software? The USRP will do that for you! You can ask the USRP to set the channel and sampling rate to what you want, then you just take that input and capture it. It's done it all for you! And since you're talking about just taking samples for a small amount of time, it's not like you're trying to use this concept to quickly change channels instantaneously. So why not let the hardware do the work for you? Tom > On Nov 23, 2013 9:54 AM, "Tom Rondeau" <t...@trondeau.com> wrote: >> >> On Sat, Nov 23, 2013 at 10:44 AM, Paul B. Huter <paul.b.hu...@gmail.com> >> wrote: >> > I ran my "record" setup for over 10 seconds, but when I run the data >> > file >> > back through my "playback" I only get about two seconds (if that) of >> > result. >> > What happened to all the data, and how can I change things next time to >> > get >> > all the data? >> > >> > My record setup is as follows: >> > >> > Rate: 50M >> > USRP Source >> > Low Pass Filter >> > Decimation: 50 >> > Cutoff Freq: 15M >> > Transition Width: 5M >> > File Sink >> > >> > Paul B. Huter >> >> >> Try this. You only want to capture for a certain length of time, which >> means a certain number of samples. So put a gr::blocks::head block >> between the filter and file sink. The number of items you'll want is >> going to be samp_rate (samps/sec) * time (secs) (and we know samp_rate >> here is 1 Msps). That will store exactly that many samples and then >> shut down the application. >> >> Tom _______________________________________________ Discuss-gnuradio mailing list Discuss-gnuradio@gnu.org https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
