Uhd problem

Hi,

I’m a french student. I am working on UHD for the time-stamping on USRP2
with a WBX dboard for my final project.
I try to implement a specification on a usrp2.

I got a problem.

if someone can help me please:

when I use a switch with my usrp2 the system work but my bursts are
wrong.
(I tuned tx_timed_samples with a while loop to repeat burst with a tuned
delay between 2 bursts )

When I use my usrp2 directly on my host pc the burst are good but I can
send packet of 4449 samples max. if I send more samples, it give me an
error after 1 or 2 burst(Error: no usrp2 controle response) and with
burst under 4449 samples if I tune seconde or nseconde in the future, I
got the same problem. it can work with a delay of 100000000 ns in the
future.

I think I have a problem with my host Pc configuration (Ethernet or
somethink like that.)

Can someone help me to fixe this or explain me what i’m doing wrong.

( Sorry for my approximate english )

many thanks!

Mourad

hi nick,
thanks for your reply
i tried the same command line arguments you gave me but the problem
that there is always an under run error even if i change the args i
tried several combinations but none of them worked
any suggestions about that underrun problem

thanks,

osama

On 12/13/2011 09:02 AM, osama mohamed wrote:

hi nick, thanks for your reply i tried the same command line
arguments you gave me but the problem that there is always an under
run error even if i change the args i tried several combinations but
none of them worked any suggestions about that underrun problem

So, its entirely possible that you are seeing the result of the CPU
ramping up, or the kernel/usb drivers optimizing interrupts. Which means
that you may get some initial underflow with USRP1.

Some initial underflow is not an issue if you plan to use USRP1 for
continuous transmission. The problem would be continuous underflow,
which means your computer + application cannot produce data fast enough.

-Josh

Try dropping the sample rate even lower–1Msps?

Is it perhaps
possible that the device is plugged in to a USB-1.1 port, rather than a
USB-2.0 port?

On Tue, 13 Dec 2011 10:30:03 -0800, Nick F. wrote:

Like Josh said, one or two "U"s isn’t a big deal, but is it

underflowing constantly? That would be strange.

Are you looking at
the output on an oscilloscope?

Like Josh said, one or two "U"s isn’t a big deal, but is it
underflowing constantly? That would be strange.

Are you looking at the output on an oscilloscope?