longer tasks yet the same credits awarded for the GRPS#5 cpu tasks
29 Dec 2020 13:01:26 UTC
Topic 224339
(moderation:
I've noticed in the last few days the app version has changed and I'm now getting much longer tasks than just a week ago yet the credit its are still the same....is this just my machine or do they need adjusting? I am running the Gamma-ray pulsar search #5 v1.08 () windows_intelx86 tasks.
22 Dec 2020 task:
| LATeah1075F_1208.0_1016740_0.0_0 | 512628111 | 22 Dec 2020 13:34:14 UTC | 22 Dec 2020 18:39:01 UTC | Completed and validated | 16,391 | 16,324 |
693 |
28 Dec task:
| LATeah1076F_72.0_2346_-2.4e-11_0 | 514603113 | 28 Dec 2020 21:44:48 UTC | 29 Dec 2020 11:28:42 UTC | Completed and validated | 44,234 | 44,107 | 693 |
You can see the name of the task has changed as well as the time it's taking to finish them, nothing else has changed on my pc and I'm not crunching any different projects between the 2 dates, yet the tasks are taking 4 times as long but the credits remain at 693.
mikey wrote:I've noticed in
)
may not be a good comparison but on one of my raspberry pis the run time/cpu time has remained about the same - 26000/26000 without any change over the past several weeks. i too experienced a name change but it did not effect the timing.
EDIT: running Ubuntu Linux
Einstein uses fixed rewards
)
Einstein uses fixed rewards for tasks, no matter how long they take to crunch.
Keith Myers wrote: Einstein
)
Hmmmm
robl wrote: mikey
)
That's not good...I hope something changes for one of us!!!
mikey wrote:That's not
)
A couple of things need to be understood with this whole thread.
The app hasn't changed! What has changed (which it does do quite regularly) is the data file being analysed.
Mikey quotes figures for tasks that use LATeah1075F.dat and LATeah1076F.dat. The second one is just a new file - the next one in a long running series of data files.
I looked at one of robl's hosts where the run times are around 26,000 to 27,000 secs for many. They all used the LATeah1075F.dat data file. The next tasks to run are all for the new data LATeah1076F.dat so robl will soon be able to 'enjoy' the longer run times as well :-).
This is nothing new - it's been happening for years, although I have to rely on memory since I don't crunch CPU tasks any more. Occasionally, there will be 'short running' data files so enjoy them while they last :-). The longer running variety used to be (by far) the more common variety and certainly the type on which the original credit award was based. The figures quoted in this thread are consistent with the notion that LATeah1075F.dat was a data file of the 'short running' variety.
If there are more 'short running' files available these days, then you're getting quite a bonus :-).
Cheers,
Gary.
Gary Roberts wrote: mikey
)
Thank you very much, I was tearing my hair out trying to figure out what changed but since it's normal I will proceed towards my goal.
Thank you everyone for you help!!
Gary Roberts wrote: I looked
)
Yep, I am getting some of those new WUS on other Piz and I am noticing my times are also doubling.
robl wrote: Gary Roberts
)
Yeah it figures.