LCLS: 50 year old plasma model overthrown

Point out news stories, on the net or in mainstream media, related to polywell fusion.

Moderators: tonybarry, MSimon

Post Reply

Axil
Posts: 935
Joined: Fri Jan 02, 2009 6:34 am

Re: LCLS: 50 year old plasma model overthrown

Post by Axil »

Instead of x-rays, I would like to see research done with spark discharge. The effects of intense electron screening might be a powerful actor is the fusion process as it has been recognized for many years by the astrophysics community in low energy fusion of matter is space.

happyjack27
Posts: 1439
Joined: Wed Jul 14, 2010 5:27 pm

Post by happyjack27 »

"Scientists use complicated algorithms that may include millions of lines of code to simulate the behavior of superheated matter and build better models of how fusion works."

there is absolutely no reason there needs to be that many lines of code, i don't care how complex the simulation is. that is just bad coding. and a lot of it.

it reminds me of brook's law ( http://www.catb.org/jargon/html/B/Brookss-Law.html ): the expected advantage from splitting development work among N programmers is O(N) (that is, proportional to N), but the complexity and communications cost associated with coordinating and then merging their work is O(N^2) (that is, proportional to the square of N).

Skipjack
Posts: 6805
Joined: Sun Sep 28, 2008 2:29 pm

Post by Skipjack »

I think that the million lines of code is BS.

kcdodd
Posts: 722
Joined: Tue Jun 03, 2008 3:36 am
Location: Austin, TX

Post by kcdodd »

I don't think that's an unreasonably number. You may be assuming that represents a single monumental simulation, but of course there are numerous models for different aspects of plasma physics, plus all the code to interconnect and analyze everything, and some run on supercomputers etc. Consider linux kernal at like 15million lines.
Carter

bennmann
Posts: 241
Joined: Sun May 17, 2009 5:56 pm
Location: Southeast US

Post by bennmann »

Hardware kernels, serious business.

happyjack27
Posts: 1439
Joined: Wed Jul 14, 2010 5:27 pm

Post by happyjack27 »

kcdodd wrote:I don't think that's an unreasonably number. You may be assuming that represents a single monumental simulation, but of course there are numerous models for different aspects of plasma physics, plus all the code to interconnect and analyze everything, and some run on supercomputers etc. Consider linux kernal at like 15million lines.
most of the code in a linux kernel is actually all of the drivers and what not. since they use a mono-lithic kernel (meaning device drivers are run at cpu level 0, and thus have to be compiled into the kernel). the "true" kernel is actually pretty tiny.

Post Reply