As an additional comment to this thread started by JIM's message 52604:-

For over a year I have recorded my credits on an excel spreadsheet. The figures below support the original estimates of under allocation of credits. I have 3 PCs running 24/7 and the first four daily results show credits prior to me including WAH 2 model types and the second four when all machines were running WAH 2 models predominantly.

The difference is slightly under-stated because in the interval I had to replace a failed PC with a faster model.

Score Difference

07/09/15 17,715,079 26,294
08/09/15 17,741,556 26,477
09/09/15 17,763,513 21,957
10/09/15 17,788,352 24,839
Sub-total 99,567
16/09/15 17,868,946 7,887
17/09/15 17,877,331 8,385
18/09/15 17,884,787 7,456
19/09/15 17,891,952 7,165
Sub-total 30,893


I now tend to micromanage by temporarily suspending WAH 2 models when others are available!

---

Apologies, the formatting did not hold up in my post. The block reads "Date Total Difference"

---

It's evan worse than I figured. Compensation is less than 1/3. Thanks for putting it into numbers. Credit wise the WAH2 tasks are dogs, and slow ones at that.

---

This problem is being looked at as of about now.

Looks like this has been sorted,

These tasks may even become the ones to go for now!

We have now amended this to increase the credit by a factor of 4.5 for WAH2. You should see this reflected in credit tomorrow.
If you feel that this factor should be different please let us know and we will amend accordingly.

Thank you for alerting us to this.

Best wishes,
Sarah

I think this is likely to drop a little to accurately reflect the value of the work done. And before anyone asks, yes it will be retrospective.

Edit, increase is going to be by a factor of about 3.8.

Ok, credits maybe will be adjusted, but what about another details.
These WAH2 are true monsters - approx. 3 GB of disk space to process, approx. 1 GB to upload (uploads are hitting me very hard, because I access Internet using GSM connection with 10 GB/month limit)... And what we get from the project? No graphics, no simulation or modelling graphs on the task's page... Absolutely nothing, almost frozen progress bar only.

---

Graphics would be nice. - none of the recent model types fro Linux have had them and that is all I run these days. (the beta testing site does have graphics on some tasks but the ones I have run recently won't go to Linux over on the main site. The only time I look at the graphics if when I want to see if there is a problem with a task. - There have been some in the past that instead of erroring out go into a loop.

In the past some tasks have been split into three (if I remember correctly) because of their large size and the very long time taken (on the computers of the day) - It was not unusual for tasks to take months to finish! The problem with that approach is a great increase in the number of errors.

If I only had a GSM connection to the internet I must admit, I would not run CPDN tasks on my machines.

I do think it would be useful if information about upload size could be included on the project pages for the individual tasks and will suggest this on the mailing list I am part of by virtue of being a beta tester. Incidentally, the beta tasks don't acrue any credits at all!

Looks like the credit has just updated. My RAC has just jumped from 25,909 to RAC: 41,999.

thank you to all the:

staff,
scientists,
volunteer developers,
volunteer beta testers,
volunteer moderators,
volunteer crunchers,

As Dr. David Anderson - (PhD - Computer scientist - Developer of BOINC) wrote:
there is no more worthwhile scientific goal than investigating the fate of Earth
and CPDN has made critical contributions to this investigation.

Best Wishes to All,
Byron

Byron Leigh Hatch wrote:

Looks like the credit has just updated...

I see that, too. Thanks to the staff and admins for getting us fixed up!

---

Good work guys! I see that the credit correction has been made. My RAC, that had dropped by about half while I was running the WAH2 models, has now almost returned to pre-WAH2 levels. This makes me feel a lot better about the prospect of crunching more WAH2�s.

---

metalius

I said earlier in this thread that these new models were bigger and slower, because of the higher resolution. This higher resolution is what the researchers have been wanting for years, because it makes the simulations more realistic.
This is probably how the project is moving now, at least for some model types.

It may mean that a lot of people will have to drop out of the project because of slow net speeds and limited accounts, but also because their hardware is not up to it, and/or they're running lots of projects, with cpdn only getting a short time each day.

My earlier post is here.

WAH2 credits have been multiplied with 3.824.

I said earlier...

Les!
I know, I remember...
1. I understand and I accept project requirements.
2. I do not accept blind crunching by such enormous requirements.
P.S. I hope, my point of view is now understandable (sorry for my poor English).

---