>> >> >> > For what it's worth, I have much the same issues with Firefox. Steady > growth and increase in memory usage that only goes away with a > restart. And no, 4 GB is plenty. I remember when Mac OS ran in 8 MB on > a 68040 and still displayed the web just fine. Heck, generally faster > :-). > > === > Wanna know what's odd? I have a significant increase in responsiveness > by keeping virtual memory turned OFF for as long as possible. I don't > run com.apple.dynamic-pager.plist at system start -- I have a watchdog > script that keeps an eye on the available memory, and runs it when > things are tight. > > Firefox becomes a lot more responsive with dynamic paging turned off. > > And, there's only three programs I use that even trigger VM on my 1 GB > G4. The other two are games. > > 4 GB is plenty of memory. You should not need that much. > Firefox has had the behaviour of suddenly freeze for a more or less amount of time for as long as I remember. Safari has not. I usually choose Safari as an example of how a browser slows down proportionally to the load as one could expect. Firefox has an abstract point were things goes wrong and it often not just freeze it's own window but can get a death grip on the entire ui for current user. This is not in my view entierly a FF problem since I do not see any good reason to why OS X would let the graphics get freezed by such an application as Firefox. So, this is rather the charcteristics of Firefox than a bug for me. Bugs are in order of important something that's going to devs who crush them one by one. This one seems to apart from stay aliva also pass every javascript test or browser combat in every such articles I have read. It is ridiculious to core a performance winner which can't handle heavy loads. I belive the test procedures are defined so things like this too easely slips through unnoticed. The best way to catch these bugs must be to load the browsers with all kinds of heavy scripted sites from the web as a user could do. The test should not strive to do the same over and over, the aim is to trigger a bug.
Now that Safari has similar behaviour I hope it lead to not only a fix for Safari but also a check to see why it is possible to freeze the entire gui that is rendered! I also hope it leads to a better insight in the nature of the test suits used by different testers, escpecially the computer media! I use 2 GB of RAM on one machine and 1.5 GB on another. Perhaps I could turn this around and say that sys admins, devs etcetera being responsibel for the IT in some way should see it as obligatory to test stuff on low end machines. Always have over kill makes you somewhat loose touch with real world use. That is if you have a more diverse user base than just 12 core maxed out RAM MacPro users! // John Stalberg_______________________________________________ MacOSX-talk mailing list [email protected] http://www.omnigroup.com/mailman/listinfo/macosx-talk
