I didn't know it would work, that's the whole point of testing.  I suspected it would work because others have done similar things.  "Proving" it to work will only increase your disappointment when it doesn't, and waste precious time as you try to get it to match your expectations when you should be looking for a new approach.

No one can give you truly relevant calculations/formulas because only you know the specific requirements.

On Fri, Apr 1, 2011 at 8:18 AM, Chris.C <chris.coulter@ntlworld.com> wrote:

Thanks very much, those numbers do look good and will certainly help. But
before you ever started your project (I assume its a type of project your
doing) how did you know that it would perform as you wanted? With low CPU
usage etc?

Its all very good that I can go and say "well this person has done it, so it
has to be possible" But before ever writing your code what analysis did you
undertake to prove it could work? Or was it a case of get the components,
build and run it and then try and optimise it from there?

I hope you understand what I mean, basically I know its possible but how can
I prove it without having a working COM infront of me giving me solid data.


Steven-132 wrote:
> 600MHz is considered an "overclock" by TI, by default the COMs run at
> 500MHz.  I told you, I got all the numbers running software on a real COM.
>  I just ran top/bwm-ng to get the bandwidth/cpu usage while streaming
> frames
> over wifi.

View this message in context: http://old.nabble.com/Caspa-VL-and-Overo-Air-COM-tp31278377p31293402.html
Sent from the Gumstix mailing list archive at Nabble.com.

Create and publish websites with WebMatrix
Use the most popular FREE web apps or write code yourself;
WebMatrix provides all the features you need to develop and
publish your website. http://p.sf.net/sfu/ms-webmatrix-sf
gumstix-users mailing list