Difference Between Projects?
3 Oct 2011 22:29:11 UTC
Topic 195990
(moderation:
I am currently running Einstein, Cosmology, and Milky Way @home. I see that Einstein has discovered some new pulsars, and that is great! That got me wondering what is the difference between these three projects. Do they complement each other; do different things, or do they compete; do the same thing? Which one is more likely to succeed? How do I decide if one project is more worthy of my computer time than another? They all give good credits although Einstein takes longer, but I know the amount of credit is not the most important consideration. Would be interested to read other people's thoughts.
Difference Between Projects?
)
The projects are rather different IMHO. I don't know about Cosmology@Home, but MilkyWay@Home is an astrophysical simulation experiment: it crunches data from a model to make predictions of the evolution of galaxies which can help to understand the underlying mechanisms of nature. So "success" here is defined more in the outcome of the whole simulation, not in the result of a single WU. It also tries rather innovative simulation techniques as far as I know.
Einstein@Home is more like a classical "needle in the haystack" distributed search where the overall goal is to find a signal (or exclude the presence of a signal within certain limits) in a big heap of data from real detectors. There are several sub-searches going on with different heaps of data generated by different experiments: laser interferometric gravitational wave detectors, radio telescopes, and a gamma ray detector in space. If a signal is found, you can pinpoint the users who returned the most significant results for the signal (at least for the GW and radio telescope searches, not sure about the processing of the gamma ray stuff). And even if no signal is found, that is still a result in the sense as we know how coarse the sieve is that we are using to go thru the data. If we find nothing, any existing signal must be weaker than the weakest signal detectable by out "sieve". That's called an "upper limit" on the strength of the signal.
This is impossible to answer I guess...It's like picking your favorite football team or your favorite songwriter. Best of all: you don't even have to make a choice, BOINC was written specifically to support more than one project at a time. Apart from not "putting all your eggs in a single science basket" :-), this also helps when one project is down for maintenance or other reasons.
Some people like projects that are supported by big, established institutions, others tend to prefer smaller projects with less secure funding, and in both cases for perfectly legitimate reasons. Just make your own choice. Or pick all of them.
HB
Cosmology tries to find a
)
Cosmology tries to find a model that describes the universe.
DSKAG Austria Research Team: [LINK]http://www.research.dskag.at[/LINK]
RE: How do I decide if one
)
Well, we are delighted to be chosen by people to donate their computing power to ! ;-)
But of course, for various reasons E@H ( or any project ) may not fulfill contributor's expectations or capacities. Hence the BOINC scheme allows for a great deal of fine tuning of the work mix for a given client.
As for the intellectual content of a project and it's "worth", well that's one of those value judgments and quite individual. I've seen quite a range of opinions from purely competitive ( credits ) through to wonderment ( gee whizz factor ), none of which are necessarily mutually exclusive. Naturally the scientific value is independent of whatever motivation lies behind the donation - to twist an old phrase 'the sun shines equally upon the just and the unjust'.
So while for me credits are no more than a useful technical measure of progress, I won't belittle those that think otherwise. Their contribution is no more or less valuable. Indeed I prefer to consider a scientific enterprise like E@H as a great way of unifying people, thus opposing the fraught modern trend of making endless distinctions between us. Certainly the pulsar discoveries have demonstrated the vast cross section of people who are contributors. :-)
Cheers, Mike.
I have made this letter longer than usual because I lack the time to make it shorter ...
... and my other CPU is a Ryzen 5950X :-) Blaise Pascal