On Mon, Nov 20, 2017 at 7:52 AM, Marcus Müller wrote:
> I was about to commit a PR VOLKizing nlog10 [1], but it struck me that
> my processor (imho, correctly) says that log(0) == nan and log(-10) ==
> nan.
>
> Whereas the old nlog10 test case asserts that it's log_10(x<=0) ==
In my case, the underflow issue was fixed by moving my bpsk transmit chains
from host to fabric. Basically the new flow graph only generates samples on
the PC. In your case it looks like your simply pulling samples from file,
right?
couple things to check: take a look at cpu usage when running
Hi Marcus,
Thanks for your advice.
Here are all the messages from gnu radio companion.
The symptoms are very easy to reproduce. The block I tested is simply
a connection between Signal Source (1KHz) and Audio sink, selecting the wxgui
Generating:
On 11/20/2017 03:07 PM, Glen Langston wrote:
Hello
I’m wondering if anyone else is having the same trouble with wxgui in
Gnuradio.
I’ve installed gnu radio on several MacBooks and ubuntu systems.
I’ve got an application we’ve been using for years that uses
wxgui
Now I’ve tried to upgrade
Hello
I’m wondering if anyone else is having the same trouble with wxgui in Gnuradio.
I’ve installed gnu radio on several MacBooks and ubuntu systems.
I’ve got an application we’ve been using for years that uses
wxgui
Now I’ve tried to upgrade to get access to a new SDR but the software is
That was it. I overlooked that setting your setup.
Hopefully the last issue. I'm still getting a lot of under flow errors when I
use the same radio (A/B ) for a rx and tx block. I have a WBX daughter board in
there so simultaneous rx-tx should not be a problem. As mentioned earlier, I
was
I've had this error before, can't recall the solution however one thing to
try is setting the DDC block select parameters. On the upper DDC block set
block select to 0 and on the lower DDC set the block select to 1.
(sometimes the auto-select function fails to work correctly I think).
On Mon, Nov
Hi Folks,
I was about to commit a PR VOLKizing nlog10 [1], but it struck me that
my processor (imho, correctly) says that log(0) == nan and log(-10) ==
nan.
Whereas the old nlog10 test case asserts that it's log_10(x<=0) == -18.
So, am I allowed to break that assertion for the greater
Hi Mohammad,
multiplying with a tone is, mathematically, a frequency shift! So,
maybe you'd want to multiply your input signal with your error
frequency?
You could also estimate the frequency of your tone (there's different
ways of doing that), and generate a complex tone (with the frequency