http://setiweb.ssl.berkeley.edu/beta/workunit.php?wuid=6825221 is just one of several my phone has gone through today. The phone is finishing them in ~90 seconds, regular computers in 1/3 that time.

They're all coming up with 30 triplets.

Is this just one of those things that comes up in the data once in a while, or is there something wrong with the WUs and the way they were split?

---

David
signature sent back to alpha testing

http://setiweb.ssl.berkeley.edu/beta/workunit.php?wuid=6825221 is just one of several my phone has gone through today. The phone is finishing them in ~90 seconds, regular computers in 1/3 that time.

They're all coming up with 30 triplets.

Is this just one of those things that comes up in the data once in a while, or is there something wrong with the WUs and the way they were split?

The one you linked has AR 2.729266, and the triplet_thresh is 5.0709376. That's enough lower than the usual ~7.24 triplet_thresh for that AR to cause the quick overflow. I had a couple yesterday[1] at similar AR with triplet_thresh at 4.738186 which also overflowed.

The pulse_thresh in those WUs is also lower than normal, and the numerical difference is very nearly the same 2.17 for your case and 2.515 for my case. For reference, here's samples of the normal thresholds:

AR        triplet_thresh   pulse_thresh
-------   --------------   ------------
< 0.225   9.738            20.734
  0.4     9.149            20.149
  0.8     8.462            19.462
  1.6     7.772            18.772
  3.2     7.080            18.080
  6.4     6.388            17.388
  10+     5.943            16.943

[1]Example WU 6817455

                                                                  Joe