This morning when I was shutting my machine down from an overnight test of running openCV to display a data stream from one of my cameras, I noticed that my frame rates nearly doubled in the video window.

This happens for about 3-5 seconds just before the os goes blank, but there is a very noticeable difference in video Fps (or processing with python..).

While the machine was shutting down i was wondering what might be some of these larg processes that are closing, and how much i even really need them during python runtime.

now im trying to figure out what some of these processes are which are not mission critical to running python or opencv and shut them down.

I was wondering however what you guys would recommend, up to now I felt that the frame rate i was getting was limited by the hardware. From what I noticed this morning now makes me think that there's a good bit of hidden processing power running in the background that could potentially be tapped.

Im using a VIA P910-10Q mobo + Ubuntu 12.10

Thanks for any advise

1 Answer 1


Try running a lighter DE and see if that reproduces higher frame rates.

I suggest you test a lightweight Window manager like awesome or xmonad

sudo apt-get install awesome

login to awesome instead of unity. Awesome is really minimalistic and the default setup is... well... minimal. All the settings and configs are in ~/.config/awesome/ the main one being rc.lua.

However for testing purposes you shouldn't care. Simply press Super-Enter to spawn a terminal and run your process from there. The Super key is the Windows key, we just try really hard to avoid that word.

Running a different DE is a lazy "solution" to your problem.

If it doesn't help you will have to look at all the running processes and make guesses to which one creates the bottleneck. run htop or top to get a nice overview.

run ps -aux | grep python to get a list of all python processes.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.