Comments on: Implementing a simple circular buffer http://blogs.mathworks.com/videos/2009/05/08/implementing-a-simple-circular-buffer/ Doug Hull is a proud MathWorker who is on a mission to help you with MATLAB. Sun, 22 Nov 2009 23:32:29 +0000 http://wordpress.org/?v=2.3.1 By: dhull http://blogs.mathworks.com/videos/2009/05/08/implementing-a-simple-circular-buffer/#comment-1433 dhull Mon, 11 May 2009 13:37:48 +0000 http://blogs.mathworks.com/videos/2009/05/08/implementing-a-simple-circular-buffer/#comment-1433 I profiled this with a really large amount of data, more iterations and longer buffer. That part of the video is on the "cutting room floor". Basically, there is a negligible performance hit. To be a real test, it would need to be done in context to see if this architecture is a problem. This video is in keeping with one of the central themes of my programming style: "Avoid premature optimization". If I were to try and avoid the churn that this circular buffer goes through, the complexity and readability would suffer. -Doug I profiled this with a really large amount of data, more iterations and longer buffer. That part of the video is on the “cutting room floor”. Basically, there is a negligible performance hit. To be a real test, it would need to be done in context to see if this architecture is a problem.

This video is in keeping with one of the central themes of my programming style: “Avoid premature optimization”. If I were to try and avoid the churn that this circular buffer goes through, the complexity and readability would suffer.

-Doug

]]> By: Daniel Armyr http://blogs.mathworks.com/videos/2009/05/08/implementing-a-simple-circular-buffer/#comment-1432 Daniel Armyr Mon, 11 May 2009 07:23:29 +0000 http://blogs.mathworks.com/videos/2009/05/08/implementing-a-simple-circular-buffer/#comment-1432 Like Petter, I wonder if this should be considdered an efficient implementation or only an easy one? I use a few rinbuffers here and there and they are all implemented like this, because it is easy. However, the line buffer = [newValue buffer(1:end-1)] gives the optimizer in me the shivers. Is that just me? Like Petter, I wonder if this should be considdered an efficient implementation or only an easy one?

I use a few rinbuffers here and there and they are all implemented like this, because it is easy. However, the line buffer = [newValue buffer(1:end-1)] gives the optimizer in me the shivers. Is that just me?

]]> By: Petter http://blogs.mathworks.com/videos/2009/05/08/implementing-a-simple-circular-buffer/#comment-1431 Petter Sat, 09 May 2009 14:22:19 +0000 http://blogs.mathworks.com/videos/2009/05/08/implementing-a-simple-circular-buffer/#comment-1431 Wouldn't that solution force the entire array to be reshuffled every time a new value is added? Wouldn’t that solution force the entire array to be reshuffled every time a new value is added?

]]>