> Notes: > Latency time bins are in 1,2,5 sequence over 5 decades > Y axis is displayed on log scale by default >
An on-line histogram is great! But I feel the histogram could reveal more if there were more bins. For example my test a few days ago shows about three features: a main peak at 0, with side-peaks at +/- 10us, and very rare events at +/- 20us http://www.anderswallin.net/wp-content/uploads/2012/12/latency_histogram.png (showing both positive and negative values is probably debatable, abs(jitter) might contain all the relevant information) I made these by logging the instantaneous latency to a file (sampler+halsampler) and plotting off-line with a python script. I think it should be possible to grab the contents of the FIFO on-the-fly, not write the values to file at all, and accumulate+display the histogram 'live'. Anders ------------------------------------------------------------------------------ LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial Remotely access PCs and mobile devices and provide instant support Improve your efficiency, and focus on delivering more value-add services Discover what IT Professionals Know. Rescue delivers http://p.sf.net/sfu/logmein_12329d2d _______________________________________________ Emc-developers mailing list Emc-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-developers
