[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] Here's a little beauty

From: Colby Boyer
Subject: Re: [Discuss-gnuradio] Here's a little beauty
Date: Tue, 5 Jul 2011 21:25:30 -0700

I can borrow my lab mates N210 and see what kind of performance I can get out of it on my T410 Thinkpad (i7 proc).

On Tue, Jul 5, 2011 at 9:12 PM, Marcus D. Leech <address@hidden> wrote:
> What sort of CPU are you using?
> --Colby
AMD Phenom II X6 1055T, with 6GB of 1333MT/s memory.   Rough ballpark
calculations show
 me that even a 4096-bin FFT shouldn't take more than about
0.45GFlop/sec at 25Msps, and the
 CPU is easily capable of at least 8GFlop/sec/core.  So I'm not sure
why it's baffing at 25Msps.

I've tried using both the on-mobo 1GiGE interface, and a PCI-resident
one.  Neither of those makes
 any difference to getting large numbers of 'O' at 25Msps.

If I decimate by 3 or more before the FFT (after vectorizing), I runs
OK, consuming about 40% of the
 total system CPU, and not producing any 'O'.  I could then,
theoretically, process the FFT output vector
 to extract only the magnitudes of the bins that correspond to my
channels of interest.  But
 decimating by 3 means that I'm losing sensitivity by a factor of
sqrt(3), which I'd rather not have
 to "swallow", the application is for radio astronomy, where
sensitivity is quite important.

Principal Investigator
Shirleys Bay Radio Astronomy Consortium

reply via email to

[Prev in Thread] Current Thread [Next in Thread]