DSPRelated.com
Forums
   Forums    More Forums    comp.dsp  

Simple convolution and fft

Started by Les Cargill ●March 20, 2012
Reply by Les Cargill ●March 23, 20122012-03-23
Fred Marshall wrote:

> On 3/23/2012 10:50 AM, Les Cargill wrote:
>> Fred Marshall wrote:
>>> On 3/22/2012 10:30 AM, Les Cargill wrote:
>>>> That is what fftw calculated as the FFT of the vector { 100, 200, 100 }
>>>> And indeed - the -200 values confuse me no end.
>>>
>>> Maybe we should consider that "it's what Les got when he used FFTW" ..
>>> :-)
>>>
>>> Fred
>>
>>
>> Ach - found that one. It agrees with the results form Octave
>>
>> --
>> Les Cargill
>
> You mean that Octave is wrong too???
>
> Fred



Nah, my FFTW code was wrong. Octave agrees with the fixed
version.

--
Les Cargill
Reply by Les Cargill ●March 24, 20122012-03-24
Fred Marshall wrote:

> On 3/23/2012 10:50 AM, Les Cargill wrote:
>> Fred Marshall wrote:
>>> On 3/22/2012 10:30 AM, Les Cargill wrote:
>>>> That is what fftw calculated as the FFT of the vector { 100, 200, 100 }
>>>> And indeed - the -200 values confuse me no end.
>>>
>>> Maybe we should consider that "it's what Les got when he used FFTW" ..
>>> :-)
>>>
>>> Fred
>>
>>
>> Ach - found that one. It agrees with the results form Octave
>>
>> --
>> Les Cargill
>
> You mean that Octave is wrong too???
>
> Fred



Don't know why I didn't think of this before: - you can
bring up the .m files for most (all?) of the library functions
from Octave. Like duh!

--
Les Cargill