For those running v7 the new 7.1.48 client is an improvement.
Joe Coffland said:FAHClient V7.1.48 released (6th Open-Beta)
Postby jcoffland » 15 Feb 2012, 21:45
Beta Testers,
This release hopefully wraps up the final features and bug fixes before we begin to recommend the v7 client as the preferred client on the front page.
Image
Documentation
Installation and user guides can be found here:
FAHControl -> The Graphical User Interface (GUI) what controls the Slots.
FAHViewer -> It shows the protein being folded, if applicable.
Pictorial Installation Guide (Windows) -> A detailed pictorial guide on the V7 installation.
Installation Guide (Windows) -> A brief guide on Windows installation.
Installation Guide (Linux) -> A guide for Linux installation.
Installation Guide (OSX) -> A guide for OSX installation that is in progress.
Client Remote Interface -> Documentation for 3rd party developers.
Main Page -> Main page of the V7.
Getting Help
Aside from the documentation the best place to get help is in this forum. If you do have a problem post a message. There are many knowledgeable people ready and willing to help. Keep in mind, we greatly appreciate thorough reports delivered by patient people who can keep a cool head even when things go wrong.
Bugs/Tickets
Open Tickets Ordered by Milestone and Priority
Active Tickets by Change Time
Note: Some tickets may be closed because they are fixed in an upcoming alpha release but are not yet fixed in the beta release.
Change Logs:
FAHClient:
v7.1.48:
Added code to the Windows installer to stop the service.
Fixed Windows default theme.
v7.1.47:
Fix Error popup: gpu-index has no default. #802
Added PPD calculation. #408
Ignore ETA calculations that are triggered during folding core startup.
Fix OSX data directory permissions.
Don't use estimated progress to decide when to download new WUs.
v7.1.46:
Integrated caxalot's OSX install script changes.
Run as user nobody in /Library/Application Support/FAHClient on OSX.
Fixed windows installer copyright. #832
Retry windows install if client running.
Attempt to fix builds for OSX < 10.6. #572
v7.1.45:
Improved WU error handling, retry and recovery.
Added 'Z' to times to indicate UTC for ISO 8601 time format.
Update viewer eta and progress information more often.
Second attempt at FAHCoreWrapper '-lifeline' usage
Removed Windows installer check for previous install. #825, #726
Fixed debian configuration questions. #749
Don't allow progress estimate to go over 100%. #395
Don't build OSX app for client. Instead install to /usr/bin.
Use corrrect user home directory in OSX. #826
v7.1.44:
Cause FAHCoreWrapper to automatically exit if client dies. #794
Improved ETA/TPF/PPD estimation. #395
Update GPU index allocation after slot delete or modify. #788
Add Debian dependency on libssl.so.0.9.8. #791
GPU white list updates. #778
FAHControl:
v7.1.48:
Added some tooltips.
Fixed windows default theme.
v7.1.47:
Added UTC time to status bar. #647
Fixed missing system info. #834
Fixed false 'Inactive' systray message. #526
Display total PPD for all clients and all slots. #408
Removed Unit ID from Work Unit status.
v7.1.46:
Default local client autostart to false.
Integrated caxalot's OSX install script changes.
Fixed themes for Linux. #819
Hide theme prefernece in OSX, causes crashes in current gtk.
Really fix, Increment number of CPUs by 1 instead of 2. #804
v7.1.45:
Fixed OSX icons file. (calxalot)
Increment number of CPUs by 1 instead of 2. #804
v7.1.44:
Fixed floating buttons on log tab. #789
Changed 'Queue ID' -> 'Work Queue ID'. #790
Changed '1' -> '01' in folding slot id. #790
FAHViewer:
v7.1.46:
Integrated caxalot's OSX launcher script.
Statically link libGLEW to avoid 1.5/1.6 dep problems. #792
v7.1.45:
Use more fine grained WU progress estimated by the clinet. #808
Fixed bug which caused viewer to become jerky after client reconnect.
Fixed view/client eta/progress synchronization problems. #811
Check OSX app signature to ???. #821
Additional Notes:
OSX Support:
We have continued to work on the OSX release and are trying to resolve a number of problems. See report:31 for a list of OSX specific open defect tickets. Calxalot has been helping with this. However, there are some significant problems still outstanding. Most importantly it currently only works on OSX 10.6. There are some likely solvable issues with building applications on 10.6 that will run on 10.5. and 10.4 which may be resolved soon. 10.7 poses a much bigger challenge. Several things in Gtk2.0, the graphics toolkit we use for FAHControl, are badly broken in OSX 10.7. It also appears that Gtk2.0 is no longer being supported very well due to the release of Gtk3.0. Moving to Gtk3.0 is a major undertaking and it's unclear whether that would solve the problems with running on 10.7 anyway.
To sum things up, support for OSX 10.4 and 10.5 should be available soon but it's unclear when we will be able to support FAHControl on 10.7. A console only F@H client should work on 10.7 but that is unlikely to be popular. If you are interested in running F@H on OSX we'd love to have your feedback.
Also, thanks to calxalot, we now have an OSX uninstaller which can be found on the downloads page.
Better Support for New GPUs:
We have implemented a system for supporting new GPUs with out requiring a client upgrade. This is useful for people with new GPUs. We can now add support for GPUs by adding them to a whitelist. The client comes with a built in whitelist which can be updated manually by installing the last GPUs.txt file in your run folder. This will eventually be automated but the current manual system maybe useful to those who want to get the latest and greatest GPUs up and running.
The latest GPUs.txt can be found here: https://fah-web.stanford.edu/file-relea ... c/GPUs.txt Edit: Place it in the data directory and restart your client. For example: C:\Users\PantherX\AppData\Roaming\FAHClient
To get new GPUs added to the list make a post in the this forum with the GPU's PCI vendor and device ID requesting the change. Simply editing the GPUs.txt yourself will allow you to setup a GPU slot with an unsupported GPU but the assignment server will not assign a work unit unless it also has a matching GPUs.txt.
Improved ETA/TPF/PPD Estimation:
The work unit completion time, time per frame and points per day estimations have been much improved. If you notice problems with these please let us know. In general, it should be much better than in previous releases. Ticket #828 describes a known but low-priority problem.
Conclusion:
If all goes well we will publish this release to the front page. If you think there are any compelling reasons not to do so please let us know but please keep in mind that we are currently in a feature freeze.
Happy hunting,
Joseph Coffland
Folding@home Developer
Cauldron Development LLC