Re: Please help me in understanding ssor preconditioner
From: George Beatal (Gbeat24_at_aol.com)
Date: 03/29/05
- Next message: H. S.: "Segmentation Fault in StatInit()"
- Previous message: Zdislav V. Kovarik: "Re: MATLAB shortcoming?"
- In reply to: Victor Eijkhout: "Re: Please help me in understanding ssor preconditioner"
- Messages sorted by: [ date ] [ thread ]
Date: 29 Mar 2005 12:24:11 -0800
(Victor Eijkhout) wrote in message news:<1gu4qwj.vih9k6xksxlN%see.sig@for.addy>...
> You must have made an error somewhere. For instance "a22" should be "a22
> + a21 a11^{-1} a12". same for 3 and 4.
>
Ah, thanks for the correction, Victor. To get this result I have to
scan the rows and columns for the second matrix multiplication in
reverse order. If that is correct maybe that was my error. I did read
about a backward second sweep in one paper but I didn't understand it
at the time. Well, I'll try that and see if I can make it work.
> Btw, that formula should be (D+L) D^{-1} (D+L)^t : transpose, not
> inverse. You wrote that correctly higher up. Maybe that's your mistake?
>
That was a typo on my part. Sorry!
George
- Next message: H. S.: "Segmentation Fault in StatInit()"
- Previous message: Zdislav V. Kovarik: "Re: MATLAB shortcoming?"
- In reply to: Victor Eijkhout: "Re: Please help me in understanding ssor preconditioner"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
