Advice on using FFTW

[CBFalconer]

.... snip ...

Were you honestly trying to help the OP with your response? Did you
seriously expect him to post the full standard C source and
documentation for FFTW either here or someplace where we can access
it? Did you seriously expect anyone here to take the time to become
enough of an expert on it to help?

It seemed blatantly obvious to me that the OP's only sensible course
of action would be to ask elsewhere. If you didn't know anything
about FFTW, you didn't need to say anything.

No, but I was trying to be clear about what would make the enquiry
topical, with a slight dose of satirical humor. I guess the whole
thing was not an outstanding success. This seems to be the fate of
my humour far too often these days.

 

[Flash Gordon]

Malcolm McLean wrote, On 28/12/07 07:32:

Firstly, C experts who post here want to get something out of the group.

Yes. Specifically a lot of them want to learn more about C and/or help
other people learn more about C.

If you know C, but don't know about FFTs, then it really is worth
advancing your knowledge a notch. The FFT is a fundamental technique
that is used in a wide variety of applications,

It is fundamental to a large number of applications but is completely
irrelevant to a *vast* amount of programming. I've done exactly two
tasks where FFTs were involved. FFTs are *not* fundamental to computing,
they are fundamental to a subset of the tasks done in computing.

it not like a Windows
API call which MS will deprecate next week.
Irrelevant.

Secondly, we shouldn't draw topicality so tightly that algorithms cannot
be discussed at all, though a post on FFTs rather than on C interfaces
to FFTs belongs elsewhere. In this case, the OP seems to know what an
FFT is, and how it can be used. His problem is getting a C program to
call a function he understands at the mathematical level. Really clc is
the right group for this.

So given a choice between a group where people who use the library and
thus know its interface are likely to post and one where most people do
not use the package, you think the best place to post is the one where
people do not have the expertise. Interesting.

Given a definition of the interface I've no problem with discussion any
C issues, it's just that there are far more things that can go wrong
which are nothing to do with C than things which are. Also I prefer
people get the help they need which generally means asking where a
library is known when asking about using a library.

 

[James Kuyper]

....
Probably CBFalconer was expecting a simple example of some code trying
to use an API.

No, he was explicitly asking for the source code of the package itself,
not a use-case. This was heavy-handed attempt at humor, obvious to
anyone with a sufficient contextual knowledge to recognize that fact.
However, the OP was extremely unlikely to be a member of the group of
people who possessed sufficient contextual knowledge, which made it a
very poor way to answer the OP's question.

 

[Malcolm McLean]

Malcolm McLean wrote, On 28/12/07 07:32:
It is fundamental to a large number of applications but is completely
irrelevant to a *vast* amount of programming. I've done exactly two tasks
where FFTs were involved. FFTs are *not* fundamental to computing, they
are fundamental to a subset of the tasks done in computing.

Is multiplying two integers fundamental enough for you? That's just one of
the many applications of FFTs. When you write z = x * y; the chip might well
execute an FFT in hardware.

So given a choice between a group where people who use the library and
thus know its interface are likely to post and one where most people do
not use the package, you think the best place to post is the one where
people do not have the expertise. Interesting.

Given a definition of the interface I've no problem with discussion any C
issues, it's just that there are far more things that can go wrong which
are nothing to do with C than things which are. Also I prefer people get
the help they need which generally means asking where a library is known
when asking about using a library.

I don't know where FFTW is discussed, but I expect they don't have much
patience with "how do I load a bitmap, get the data into the right format,
and call the transform". They'd be right to see this as a C rather than a
frequency transform issue.

 

[Flash Gordon]

Malcolm McLean wrote, On 29/12/07 07:54:

Is multiplying two integers fundamental enough for you? That's just one
of the many applications of FFTs. When you write z = x * y; the chip
might well execute an FFT in hardware.

Irrelevant, the multiply could be done by some other means (and is on at
least some processors). You, the programmer, do not even need to know
FFTs exist in order to multiply two numbers. If the only way you could
write a "hello world" program involved the *programmer* knowing about
FFTs you might have a valid point, but the fact is that you only need
knowledge about them for a subset of programming and you need to know
nothing about FFTs (not even their existence) for a *vast* amount of
programming.

I don't know where FFTW is discussed,

Others do and posted the information.

but I expect they don't have much
patience with "how do I load a bitmap, get the data into the right
format, and call the transform". They'd be right to see this as a C
rather than a frequency transform issue.

No, they would be right to see it as a LIBRARY issue, which is what I
have said repeatedly.

 

[David Thompson]

<snip> But if I was doing a one-off job, or
doing a homework problem for a graduate class, or something like that, I
would definitely opt for easy over correct or fast.

Easy over fast, agree; over flexible, ditto; over portable or robust,
OK; but I hope you didn't really mean easy over correct. <G?>

- formerly david.thompson1 || achar(64) || worldnet.att.net