walt <[EMAIL PROTECTED]> posted [EMAIL PROTECTED], excerpted below, on Thu, 07 Aug 2008 01:01:01 +0000:
> I don't know what the 53397 articles in 11.3 seconds is about, but > grepping thru pan's source code should tell us that -- tomorrow ;o) That's the number of articles pan loads from disk, that it has pre- threaded from previous sessions. Old-pan used to dynamically rethread every time (tho it has a similar log entry as it did so), while new-pan stores the existing threaded state and only has to dynamically thread new overviews/headers as they come in. Or at least that's what it's /supposed/ to be, AFAIK. I've suspected for awhile that there's a race condition in the timing, such that it doesn't actually start timing until many are already read-in from disk. That because the uncached and cached timings are too close to each other. So it's not necessarily actually timing the read-in from disk, but the actual load into pan memory from system memory -- if that's the proper way to put it, I'm not sure I'm using the proper terminology there. But maybe that's better anyway, from a pan performance measuring viewpoint. -- Duncan - List replies preferred. No HTML msgs. "Every nonfree program has a lord, a master -- and if you use the program, he is your master." Richard Stallman _______________________________________________ Pan-users mailing list Pan-users@nongnu.org http://lists.nongnu.org/mailman/listinfo/pan-users
