On 8/29/2013 2:34 AM, Anders Wallin wrote: > How difficult would it be to automatically gather information about the > system (rt-patch, versions, etc.) along with the jitter values and format > them into a standard message (email, whatever). > The website could then have a database which is auto-updated whenever it > receives a correctly formatted email? > > for preempt-rt, there are already large databases out there. Does anyone > know how well the linuxcnc latency-test correlates with the standard > preempt-rt values quoted? > https://www.osadl.org/Latency-plots.latency-plots.0.html > > > Anders > > On Thu, Aug 29, 2013 at 9:19 AM, Chris Morley > <chrisinnana...@hotmail.com>wrote: > >> I am thinking we are going to need to add a column for >> type of realtime patch with UB coming along soon. >> >> I am also thinking that the two columns showing max interval are less >> useful. >> removing them would give more room for other text and also help people >> to concentrate on what is important - jitter. >> >> Even at that, servo jitter is less useful as everywhere we always quote >> base jitter. >> We should probably address that somehow in the manual... >> >> I will change the wiki, removing the max interval and adding realtime >> patch. >> Anyone think this is the wrong thing to do? >> >> >> Chris M >> >
Chris: I always wanted to improve that table but I ran out of round-toits even before I finished some other Wiki cleanup work I began earlier. After Charles put up some numbers from his experiments with RT-PREEMPT, I started thinking about how to merge them and my own with the existing table. At the time, I was thinking there was so much entries already in the existing RTAI table which were unlikely to be redone that I'd try a new table for systems evaluated going forward. In the end I did nothing at all, so my advice to you is, ignore my thinking out loud and "just do it!" It's dead simple to try ideas on the Wiki. Just edit the existing page, preview it, and abandon your edits if you don't like the result. Anders: I like the idea of scripting the transmission and acquisition of the test results. It fits with previous noodling on this list about a standard system information-gathering and reporting scheme which, among other things, would improve our ability to help people debug their installation. Are you volunteering? <grin> As for intercomparison of the results of the different methods of determining jitter---our latency-test, the latency tests which come are often included in the realtime suites, and the osadl test---I'd love for someone knowledgeable to craft yet another Wiki page. I know I'm not that person. Regards, Kent ------------------------------------------------------------------------------ Learn the latest--Visual Studio 2012, SharePoint 2013, SQL 2012, more! Discover the easy way to master current and previous Microsoft technologies and advance your career. Get an incredible 1,500+ hours of step-by-step tutorial videos with LearnDevNow. Subscribe today and save! http://pubads.g.doubleclick.net/gampad/clk?id=58040911&iu=/4140/ostg.clktrk _______________________________________________ Emc-developers mailing list Emc-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-developers
