Since the latest executable update I'm finding on all Windows devices that the ETA is incorrectly calculated and all tasks seem to end at around 64%.

Did somebody mean to write 0x64 in the code but wrote 64 instead? :-P

Otherwise, have I got a setting wrong or something?

---

I took a look at the three machines you have associated with your profile and two are not doing any new work but the work they returned even up to early this morning were fine. The last machine also seems to be working fine and is still actively crunching and returning good results. Can you please provide a link to the machine(s) you are having problems with?

And if by ETA you mean Estimated Run Time - then you just need to let them run for a while and eventually they will normalize - it takes BOINC manager a while to normalize the times to your system.

Thanks for the reply and sorry to be so slow getting back to you.

I'm not seeing any issues with task completion and I appear to be getting credit no worries, but on all three machines when I run DENIS it estimates somewhere around an hour for a task but then only counts down very slowly. The tasks do appear to take the right amount of total time to complete, but because the progress of the task seems to end at ~65% the time estimate never reaches zero and the task simply drops from around 20 minutes remaining to zero at the end.

Again, that's the only issue I'm seeing so if it's simply that you can't fit the task's progress to a linear counter then I'm not worried, but this didn't happen with previous versions of the tasks so it got me thinking what might have changed. The architecture of one machine has changed significantly but the other two have not so I don't think that's it.

Anyway, not a problem and it won't stop me crunching those heart simulations! :-)