We've been watching on the system monitor of both machines
We just subscribed to the raw channels instead of the compressed and we're
seeing a transfer rate of up to 8MB/s on the system monitor instead of
1MB/s, so the problem has something to do with the compression
any ideas?
On Mon, Mar 7, 2011 at 5:27 PM, Ken Conley <
kwc@willowgarage.com> wrote:
> Have you taken the measurement using some other means? Please note from
> the documentation below that "rostopic bw" is not the best means of
> measuring this:
>
> """
> http://www.ros.org/wiki/rostopic#rostopic_bw
> NOTE: the bandwidth reported is the received bandwidth. If there are
> network connectivity issues, or if rostopic cannot keep up with the
> publisher, the reported number may be lower than the actual bandwidth.
> rostopic is implemented in Python, which cannot maintain as high throughput
> as roscpp <http://www.ros.org/wiki/roscpp>-based nodes.
> """
>
> - Ken
>
> On Mon, Mar 7, 2011 at 2:21 PM, Sean Anderson <
> sean.william.anderson@gmail.com> wrote:
>
>> Hi,
>>
>> My team is trying to use image_transport compression in order to transfer
>> large amounts of image data over our wireless network.
>>
>> We're using wireless N, but we have been unable to transfer for than 1
>> megabyte per sec over the network using ROS.
>>
>> When using "rostopic bw <channel>" in conjunction we have noticed that the
>> bandwidth rises to 2 megabyte per sec (we've seen a max of 4MB/s with
>> multiple tabs running the bw check)
>>
>> A colleague of mine noticed something about wpa_supplicant_node having a
>> wireless 11b max bandwidth limiter, is this what is limiting our transfer
>> rate? Does anyone know how to unlock the limit?
>>
>> Thanks so much,
>>
>> Sean Anderson
>>
>> _______________________________________________
>> ros-users mailing list
>> ros-users@code.ros.org
>> https://code.ros.org/mailman/listinfo/ros-users
>>
>>
>
(text/plain)