Hypernova, I bet you didn't realise crunching for science could be this much fun

Just to help you set up your 'mega flush' here is the average time between when the first result is sent for a workunit and the time that it is assimilated:

app_name	hours
faah	        49.0
hcmd2	        56.9
hcc1	        75.4
hpf2	        460.3

Note that since I'm talking about assimilation, it is very helpful for HPF2. For the other projects assimilation occurs a few minutes after validation.

If you keep your nerves and crunch that 9 day cache off-line, then before reconnecting do that switch to the ZR project, I forecast you could be doing 5 million plus. Simply multiply the average credit per hour per your My Grid page for your the target sciences and multiply with cores and hours per day.

If you had the Large-BOINCTasks utility, you can monitor up to 300 crunchers from one place. Set it to filter [condense] the Upload results and it will show the total days/hours completed on the go so the saliva will start running in abundance as the numbers grow bigger.

Note that the uploads per result are rather sizeable, so be patient and sometimes there will be upload back-offs due grid load and do set the caches back to normal before connecting again else you end up with another X days worth of cache.

The keeper.

PS: Don't know which client you run, but until very recent versions the accidental hit on that Project Update button will open the line for 5 minutes and flush whatever there is complete. It does without sharing your feelings :D

In trying to get all of my projects up to 5 years, I switched to a combination of HCMD and HPF2. Over the last week or so, I've only seen a handful of errors on HPF2.
Two days ago, I upgraded all but one machine to the latest BOINC recommended version (6.10.56) and have yet to see an error on the new version.
It might be worth a try to upgrade a planet or two and give it a test.