Is there a way i can back up my tasks etc to avoid this happening again
Unless you need to tinker about with the BOINC software itself there's no need to back your tasks up. That said, for future reference, SETI tasks are stored in the
..\BOINC\Data\projects\setiathome.berkeley.edu directory (off the top of me head).
I apologise for you losing the tasks you had when you installed the optimised apps - I couldn't remember if they were valid or not with those apps; obviously they weren't
About the stash of WUs halz mentioned... when SETI is back up after its routine maintenance outage, go into your account and, under Preferences, change the 'Connect about every' (or similar) option to however many days' worth of WUs you wish to download and have available to crunch at any given time. A setting of at least 3-4 days is preferable in case of unplanned project downtime, though I have mine set to 7 days just to make sure
The other thing that you might want to use is the ReSchedule tool that you can find on
this page. (You'll need to create an account on the lunatics site to download it.) This converts VLAR (Very Low Angle Range) GPU WUs into CPU WUs (VLAR WUs take aaaaaaaages on GPUs, but run normally on CPUs). I think if you downloaded the Unified Installer the GPU app it included automatically kills any VLAR WUs it comes across, but this is no use to the science being done and is detrimental in terms of Berkeley's (and your!) bandwidth. ReSchedule allows such WUs to run on your CPU instead of being killed, which is much better for everyone. It also allows CPU WUs to be converted to GPU WUs, which is useful if you ever run out of GPU tasks but have a clutch of (non-VLAR) CPU WUs.
EDIT: I have no idea if it works in Vista or Win 7 - I've only ever run it in XP.