Advanced search

Forums : General Topics : if camb_legacy tasks are basically useless, then why...?
Message board moderation

To post messages, you must log in.

AuthorMessage
Thund3rb1rd

Send message
Joined: 16 May 08
Posts: 7
Credit: 2,317,209
RAC: 1,533
Message 21947 - Posted: 7 Oct 2018, 23:06:56 UTC

I've read several posts lately that state the camb_legacy tasks are basically useless to the C@H project. If that's true, then why is C@H even messing with them? Why not kill them off completely? I receive both camb_legacy and camb_boinc2docker tasks but for some time now, I receive far more camb_legacy tasks than VM tasks. Which leads inevitably to questioning the value of volunteer computing for C@H at all?
ID: 21947 · Report as offensive     Reply Quote
Jonathan

Send message
Joined: 27 Sep 17
Posts: 144
Credit: 7,284,489
RAC: 12,750
Message 21950 - Posted: 8 Oct 2018, 11:46:22 UTC - in response to Message 21947.  

If you want to turn off the legacy workunits, you can do that by editing your project preferences.
ID: 21950 · Report as offensive     Reply Quote
pls

Send message
Joined: 5 Nov 10
Posts: 3
Credit: 603,960
RAC: 0
Message 22045 - Posted: 12 Jan 2019, 6:42:26 UTC

I'm sure you realize that you didn't answer the question that was asked.

Can we try that again, please. I'd also like to know.
ID: 22045 · Report as offensive     Reply Quote
Jim1348

Send message
Joined: 17 Nov 14
Posts: 100
Credit: 4,242,353
RAC: 3,900
Message 22046 - Posted: 12 Jan 2019, 16:11:48 UTC - in response to Message 21947.  

Good question, even though I do not do legacy.

But I am not clear what good we are doing on BOINC2Docker either. Are the results being used currently? Are there any new publications? Or have all the basic questions been answered? There have not been any updates for a while.
ID: 22046 · Report as offensive     Reply Quote
Jonathan

Send message
Joined: 27 Sep 17
Posts: 144
Credit: 7,284,489
RAC: 12,750
Message 22047 - Posted: 13 Jan 2019, 12:25:28 UTC - in response to Message 22046.  

Have you checked out the FAQ section, "What science is being done with Cosmology@Home?"
That has a few answers.
ID: 22047 · Report as offensive     Reply Quote
Jim1348

Send message
Joined: 17 Nov 14
Posts: 100
Credit: 4,242,353
RAC: 3,900
Message 22048 - Posted: 13 Jan 2019, 13:25:40 UTC - in response to Message 22047.  
Last modified: 13 Jan 2019, 13:26:04 UTC

Thanks, I have seen it. But it only gives a general overview of the type of science done. They are using our PCs to calibrate the PICO code (whatever that is). It is all a very nice use of machine learning.

But how calibrated do you have to get? Are any scientist still accessing their program, or has all the useful science already been done? Are we wasting our time and computer power?

It would be interesting to know.
ID: 22048 · Report as offensive     Reply Quote
Jonathan

Send message
Joined: 27 Sep 17
Posts: 144
Credit: 7,284,489
RAC: 12,750
Message 22049 - Posted: 13 Jan 2019, 18:40:34 UTC - in response to Message 22048.  

There is some info off of Marius's webpage. Link below.
https://cosmicmar.com/research/

PICO is explained a bit at link below.
https://sites.google.com/a/ucdavis.edu/pico/

Marius is about the only one I know of that might be able to explain or answer some of your questions but it is a bit rare to see him post. I really haven't seen anyone else from the project post and I don't know if Marius has any 'helpers' or it is pretty much just him running this at the moment. Hopefully, he will post sometime.
ID: 22049 · Report as offensive     Reply Quote
Jim1348

Send message
Joined: 17 Nov 14
Posts: 100
Credit: 4,242,353
RAC: 3,900
Message 22050 - Posted: 13 Jan 2019, 20:14:36 UTC - in response to Message 22049.  

Those are quite well written for the non-experts, and give us an idea of how are PCs are being used. But we need to get (at least) annual updates.

    How much have the estimations improved?
    What has that helped him to accomplish?


We spend a lot of time and computer power trying to get the answers. But what are we getting the answers to these days?

ID: 22050 · Report as offensive     Reply Quote
Jim1348

Send message
Joined: 17 Nov 14
Posts: 100
Credit: 4,242,353
RAC: 3,900
Message 22052 - Posted: 15 Jan 2019, 18:22:28 UTC - in response to Message 22050.  

I think these publications which cite PICO are more what I am looking for.
http://adsabs.harvard.edu/cgi-bin/nph-ref_query?bibcode=2007ApJ...654....2F&refs=CITATIONS&db_key=AST

It would help if Marius would pick out a significant one each year and explained it to us.
ID: 22052 · Report as offensive     Reply Quote

Forums : General Topics : if camb_legacy tasks are basically useless, then why...?