|Re: [eigen] fft module status|
[ Thread Index |
| More lists.tuxfamily.org/eigen Archives
- To: eigen@xxxxxxxxxxxxxxxxxxx
- Subject: Re: [eigen] fft module status
- From: Rohit Garg <rpg.314@xxxxxxxxx>
- Date: Tue, 1 Dec 2009 22:33:45 +0530
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :from:date:message-id:subject:to:content-type :content-transfer-encoding; bh=4pn8Qh2gX2s7H5VPutmuPkklEBAkllBZ2T5RdVHRcVA=; b=gp6pBD8KdnCcWdyo8ZMfLnVpGlenvPcdV/xhghHEvR742g2kRCs5Q/bt7Q6bYGxEQ2 2ItiCg6th9NjckU2ExMRh4OtGPvi6mmkH/o+qi2M6QGBJmC3WMl8RlM4dqJYdG3MN+0d BtghW9pCe2j2MIFPPpwmUrhJv/cHND8CgkfD4=
- Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :content-type:content-transfer-encoding; b=fFT/0E3SMStYHSRVXqzUKg8ce+lSkP3XHpzUkR2Wanx29v7ngzK8Cw8M0h+ITqij8c Lr4u7hjCHX0VmqqmiXzgxsxywWXNtj9RZq7ahjAeXt2uTSbHYRj8pjBdUPvBlTvkKm+O RTsN6r8/01sBQ46287bOU0Md++47hROFUqw/k=
I'd say go for 2D FFT first
On Tue, Dec 1, 2009 at 9:46 PM, Mark Borgerding <mark@xxxxxxxxxxxxxx> wrote:
> You are correct in your assessment. The MKL backend and 2-D FFTs are still
> I updated the status page and put a link to the EigenFFT wiki page to
> clarify this.
> Please feel free to opine (here or on the wiki discussion page) which ranks
> higher in priority: MKL backend or multi-dimensional FFTs.
> FYI, If you want to use MKL as a backend before I get around to writing it
> (probably within a month), you can make the intel MKL fftw3xc interface that
> comes with it. Then tell Eigen::FFT that it should use FFTW as a backend.
> This would allow the advantages of MKL with the current state of the code.
> -- Mark
> Thomas Capricelli wrote:
>> I'm a bit confused with respect to the FFT module. The status page on the
>> wiki seems to say that all but tests and documentation is done, but i can't
>> find support for dimension more than 1, and i can't find anything related to
>> the MLK backend..
>> Mark, can you provide some more information ? I have looked at your clone,
>> but there doesn't seem to be anything more there. Could it be that some
>> commits were not merged in ?
>> best regards,
Department of Physics
Indian Institute of Technology