Andi Vajda wrote: > > … the UI/data interaction model that was chosen. Any action in the UI can > (and does) have lots (really, _lots_) of ramifications all the way down to > the lowest level of the core. > > This is one of the things the ongoing rearchitecture effort is attempting > to address. Blaming Python for Chandler's performance woes is barking up > the wrong tree. >
More than anything, I'm curious. Not complaining about anything in particular. Just seeing how Chandler 1.x architecture behaves in a 'tight' situation such as this, sharing work-arounds. The flash drive + symlink approach is working well for me. The PowerPC versus Intel debate is well-trodden, I shouldn't encourage it here :) Rob Stearn wrote: > > … if you're observing greatly increased speeds in PPC than in Intel, it > isn't memory bandwidth thats causing it. > > From a troubleshooting POV, I would suspect: > > Disk defragmentation (yes it can happen on Macs too) > > Remaining Disk space > > An indication a fresh install is needed, multiple upgrades > > 10.3>10.4>10.5 especially bridging architectures leaves a lot of stuff you > don't need. > > Something very wrong in the quality of the Python distro thats provided > > Another process taking up CPU. > > All or some of the above. > > From my experience PPC machines running Chandler are accompanied with the > soft whistle of fans. > In my cases, I guess, in order of significance: 1. (most significant) layout of files on disk 2. great size of catalog of HFS+ volume, with RAM limited to 2 GB. Possible, I guess, that Intel and PowerPC architectures have different approaches to use of /.hotfiles.btree and that 'end' of a disk. Within the very limited and highly fragmented free space, fragmentation of swap files seems to make surprisingly little difference. I'll review this topic in a few weeks, or sooner, when I get a new laptop :) Regards Graham -- View this message in context: http://n2.nabble.com/Performance%3A-comparing-Intel-and-PowerPC-versions-of-Chandler-desktop-1.0.3-RC1-on-Mac-OS-X-10.5.6-tp2348910p2372593.html Sent from the Chandler users mailing list archive at Nabble.com. _______________________________________________ [email protected] mailing list unsubscribe here: http://lists.osafoundation.org/mailman/listinfo/chandler-users Chandler wiki: http://chandlerproject.org/wikihome
