Every few days boinc runs a benchmark on your system to figure out approximately how much work your system does. So, my guess is that boinc recently ran a benchmark on your system and it thinks it can do more work then last time it ran the benchmark.

Using new benchmark
225.9/6.36 = 35.5 pts per hr

using old benchmark
176.7/5.73 = 30.8 pts per hr
199.8/6.48 = 30.8 pts per hr

So now all of your WUs will ask for 15% more credit per hr of work. :)

Kate,

This calculation only applies if the two wingman did an unequal number of jobs/attempts/positions.

As for Bearcat prepping to get steamed up, he's the one at the wheel feeding his own oven ... in the know what he did between the 29th and the 2nd to that client to suddenly have 15% bigger claim due a change of benchmark, as anhhai correctly "did compute!". Him torquing the CPU cycles (OCing) could have done that, but another dozen things only Bearcat knows could have done it too. Vaguely remember a discussion recently of someone finding his/her CPU to not stick to max speed :P

--//--

On this particular box, its a dual dual core 2 ghz xeon woodcrest with 4 gb memory. This one has not had any throttling issues as my westmere box has. Zero OC or anything else. Shouldnt have changed by much but since i run clean water and CEP2 one at a time. My guess will be since CEP2 writes allot to disk, CEP2 is slowing down boinc in general. This box does nothing but crunch 24/7. Already found granted points similar to my post. Thanks for the replys.

Nanoprobe, its about efficiency. In this day and age, waste is expensive. if i determine a project is wasting cycles, like CEP2 does IMHO, then its time to move on to something that doesnt. been here long enough to know between getting the bugs out to something thats wasteful.

No replies needed anymore here. I will move on to something else!

nanoprobe, WUs/hr is not a good barometer because WU size vary so much, but the work still needs to be done. Pts is a good barometer. Let me give you an example. One of my PC will get only around 8 pts per hr crunching HFCC and HPF2. However, it get 15 pts per hr for everything else (except CEP2). This tells me that I should not run HFCC or HPF2 on it. For other machines most project can run fairly well on it. But if I want to optimize, I look at how efficient my system compares to others by the way of pts.


Edit: this is assuming the same OS. Because Linux efficiency is different from windows, you have to compare linuxes boxes to linuxes boxes. windows boxes to windows boxes

nanoprobe, I said assuming the same OS. We all know that linux is better at doing work. Comparing pts between Linux machines and windows machines is a worthless exercise and just gets people made at how wrong it is.
Ranking is also a very misleading. You can do several HCC WU or 1 CEP2 WU. But it doesn't matter, crunch whatever you like. Do what makes you happy.