Is this thread dead? Because it seems to be either ALOT of people crunching here lately or there aren't alot of tasks available right now. I mean it seems every day or so a project with lots of tasks all of a sudden has none for several days to a week or more.

I don't see how changing deadlines does anything. It won't make tasks complete faster. If someone is running only SRBase this does nothing. If there a mix of projects then the client will adjust, once, if needed. But then will continue with the same project weighting.

If anything, someone with a large queue will just have tasks fail to complete before the deadline, be sent back to server, redistributed and end up at the back of someone else's queue.

It will be faster. Instead of waiting 3d to resend it will only take 1d.

When would you typically bring a base onto SRbase for testing?

Range 1-25k is not for BOINC.

For some of the smaller test groups (k's remaining>10,000; n=2.5K), could a script do batches of prime tests?

For example, the script would package ~350 tests into a single work unit and the whole work unit would run for the full duration of the k (~9 minutes at AMD Ryzen 9 3900 3.28 GHz - 4 cores with PRST).

Each k would be tested within the range until primality or failure.

I could prepackage these if necessary. There are 120 bases with k's remaining>10,000 n=2.5K. One is all the way up to 2.2 Million k's.

I think a script that would package the smaller groups into work units would be beneficial.

You will have million of tests, this range isnt something for boinc. Split the tests with llr and use stoponprimedk. Its faster.