|
From: | Andrew Ge |
Subject: | [Discuss-gnuradio] Transmitter's bizarre behavior on 64 bit workstation |
Date: | Mon, 18 Oct 2010 12:48:44 -0400 |
User-agent: | Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.2.9) Gecko/20100915 Thunderbird/3.1.4 |
Eric,I found a puzzling fact about running GNU Radio on the 64-bit Lenovo workstation D20.
Basically, the maximum packet size (payload at the transmitter side in the benchmark_tx.py) must be less than or equal to 221. Below this size, packets from the workstation can be correctly received by either a laptop or another workstation; above this size, no packet can be received correctly on either a laptop or another workstation. I was using 500kb data rate and GMSK modulations.
The same code works perfectly on a laptop as the transmitter, no matter the receiver side is on a laptop or a workstation.
At first, I thought it might be a result of the quad core (8 CPUs) architecture. So I disable 3 cores and only one core was left. However, I got the same result.
Also, the CPU usage for a transmitter doesn't make sense to me either. It was above 90% on the workstation (2 Intel Xeon X5550 cpus at 2.67 GHz). However it was only around 35% on the laptops (2 Intel T9550 CPUs at 2.66GHz).
I could not understand what might be the problem. Andrew
[Prev in Thread] | Current Thread | [Next in Thread] |