tutorial.texi

@node  Tutorial, Other Important Topics, Introduction, Top
@chapter Tutorial
@menu
* Complex One-Dimensional DFTs::
* Complex Multi-Dimensional DFTs::
* One-Dimensional DFTs of Real Data::
* Multi-Dimensional DFTs of Real Data::
* More DFTs of Real Data::
@end menu

This chapter describes the basic usage of FFTW, i.e., how to compute
@cindex basic interface
the Fourier transform of a single array.  This chapter tells the
truth, but not the @emph{whole} truth. Specifically, FFTW implements
additional routines and flags that are not documented here, although
in many cases we try to indicate where added capabilities exist.  For
more complete information, see @ref{FFTW Reference}.  (Note that you
need to compile and install FFTW before you can use it in a program.
For the details of the installation, see @ref{Installation and
Customization}.)

We recommend that you read this tutorial in order.@footnote{You can
read the tutorial in bit-reversed order after computing your first
transform.}  At the least, read the first section (@pxref{Complex
One-Dimensional DFTs}) before reading any of the others, even if your
main interest lies in one of the other transform types.

Users of FFTW version 2 and earlier may also want to read @ref{Upgrading
from FFTW version 2}.

@c ------------------------------------------------------------
@node Complex One-Dimensional DFTs, Complex Multi-Dimensional DFTs, Tutorial, Tutorial
@section Complex One-Dimensional DFTs

@quotation
Plan: To bother about the best method of accomplishing an accidental result.
[Ambrose Bierce, @cite{The Enlarged Devil's Dictionary}.]
@cindex Devil
@end quotation

@iftex
@medskip
@end iftex

The basic usage of FFTW to compute a one-dimensional DFT of size
@code{N} is simple, and it typically looks something like this code:

@example
#include <fftw3.h>
...
@{
    fftw_complex *in, *out;
    fftw_plan p;
    ...
    in = (fftw_complex*) fftw_malloc(sizeof(fftw_complex) * N);
    out = (fftw_complex*) fftw_malloc(sizeof(fftw_complex) * N);
    p = fftw_plan_dft_1d(N, in, out, FFTW_FORWARD, FFTW_ESTIMATE);
    ...
    fftw_execute(p); /* @r{repeat as needed} */
    ...
    fftw_destroy_plan(p);
    fftw_free(in); fftw_free(out);
@}
@end example

You must link this code with the @code{fftw3} library.  On Unix systems,
link with @code{-lfftw3 -lm}.

The example code first allocates the input and output arrays.  You can
allocate them in any way that you like, but we recommend using
@code{fftw_malloc}, which behaves like
@findex fftw_malloc
@code{malloc} except that it properly aligns the array when SIMD
instructions (such as SSE and Altivec) are available (@pxref{SIMD
alignment and fftw_malloc}). [Alternatively, we provide a convenient wrapper function @code{fftw_alloc_complex(N)} which has the same effect.]
@findex fftw_alloc_complex
@cindex SIMD


The data is an array of type @code{fftw_complex}, which is by default a
@code{double[2]} composed of the real (@code{in[i][0]}) and imaginary
(@code{in[i][1]}) parts of a complex number.
@tindex fftw_complex

The next step is to create a @dfn{plan}, which is an object
@cindex plan
that contains all the data that FFTW needs to compute the FFT. 
This function creates the plan:

@example
fftw_plan fftw_plan_dft_1d(int n, fftw_complex *in, fftw_complex *out,
                           int sign, unsigned flags);
@end example
@findex fftw_plan_dft_1d
@tindex fftw_plan

The first argument, @code{n}, is the size of the transform you are
trying to compute.  The size @code{n} can be any positive integer, but
sizes that are products of small factors are transformed most
efficiently (although prime sizes still use an @Onlogn{} algorithm).

The next two arguments are pointers to the input and output arrays of
the transform.  These pointers can be equal, indicating an
@dfn{in-place} transform.
@cindex in-place


The fourth argument, @code{sign}, can be either @code{FFTW_FORWARD}
(@code{-1}) or @code{FFTW_BACKWARD} (@code{+1}),
@ctindex FFTW_FORWARD
@ctindex FFTW_BACKWARD
and indicates the direction of the transform you are interested in;
technically, it is the sign of the exponent in the transform.  

The @code{flags} argument is usually either @code{FFTW_MEASURE} or
@cindex flags
@code{FFTW_ESTIMATE}.  @code{FFTW_MEASURE} instructs FFTW to run
@ctindex FFTW_MEASURE
and measure the execution time of several FFTs in order to find the
best way to compute the transform of size @code{n}.  This process takes
some time (usually a few seconds), depending on your machine and on
the size of the transform.  @code{FFTW_ESTIMATE}, on the contrary,
does not run any computation and just builds a
@ctindex FFTW_ESTIMATE
reasonable plan that is probably sub-optimal.  In short, if your
program performs many transforms of the same size and initialization
time is not important, use @code{FFTW_MEASURE}; otherwise use the
estimate.  

@emph{You must create the plan before initializing the input}, because
@code{FFTW_MEASURE} overwrites the @code{in}/@code{out} arrays.
(Technically, @code{FFTW_ESTIMATE} does not touch your arrays, but you
should always create plans first just to be sure.)

Once the plan has been created, you can use it as many times as you
like for transforms on the specified @code{in}/@code{out} arrays,
computing the actual transforms via @code{fftw_execute(plan)}:
@example
void fftw_execute(const fftw_plan plan);
@end example
@findex fftw_execute

The DFT results are stored in-order in the array @code{out}, with the
zero-frequency (DC) component in @code{out[0]}.
@cindex frequency
If @code{in != out}, the transform is @dfn{out-of-place} and the input
array @code{in} is not modified.  Otherwise, the input array is
overwritten with the transform.

@cindex execute
If you want to transform a @emph{different} array of the same size, you
can create a new plan with @code{fftw_plan_dft_1d} and FFTW
automatically reuses the information from the previous plan, if
possible.  Alternatively, with the ``guru'' interface you can apply a
given plan to a different array, if you are careful.
@xref{FFTW Reference}.

When you are done with the plan, you deallocate it by calling
@code{fftw_destroy_plan(plan)}:
@example
void fftw_destroy_plan(fftw_plan plan);
@end example
@findex fftw_destroy_plan
If you allocate an array with @code{fftw_malloc()} you must deallocate
it with @code{fftw_free()}.  Do not use @code{free()} or, heaven
forbid, @code{delete}.
@findex fftw_free

FFTW computes an @emph{unnormalized} DFT.  Thus, computing a forward
followed by a backward transform (or vice versa) results in the original
array scaled by @code{n}.  For the definition of the DFT, see @ref{What
FFTW Really Computes}.
@cindex DFT
@cindex normalization


If you have a C compiler, such as @code{gcc}, that supports the
C99 standard, and you @code{#include <complex.h>} @emph{before}
@code{<fftw3.h>}, then @code{fftw_complex} is the native
double-precision complex type and you can manipulate it with ordinary
arithmetic.  Otherwise, FFTW defines its own complex type, which is
bit-compatible with the C99 complex type. @xref{Complex numbers}.
(The C++ @code{<complex>} template class may also be usable via a
typecast.)
@cindex C++

To use single or long-double precision versions of FFTW, replace the
@code{fftw_} prefix by @code{fftwf_} or @code{fftwl_} and link with
@code{-lfftw3f} or @code{-lfftw3l}, but use the @emph{same}
@code{<fftw3.h>} header file.
@cindex precision


Many more flags exist besides @code{FFTW_MEASURE} and
@code{FFTW_ESTIMATE}.  For example, use @code{FFTW_PATIENT} if you're
willing to wait even longer for a possibly even faster plan (@pxref{FFTW
Reference}).
@ctindex FFTW_PATIENT
You can also save plans for future use, as described by @ref{Words of
Wisdom-Saving Plans}.

@c ------------------------------------------------------------
@node Complex Multi-Dimensional DFTs, One-Dimensional DFTs of Real Data, Complex One-Dimensional DFTs, Tutorial
@section Complex Multi-Dimensional DFTs

Multi-dimensional transforms work much the same way as one-dimensional
transforms: you allocate arrays of @code{fftw_complex} (preferably
using @code{fftw_malloc}), create an @code{fftw_plan}, execute it as
many times as you want with @code{fftw_execute(plan)}, and clean up
with @code{fftw_destroy_plan(plan)} (and @code{fftw_free}).  

FFTW provides two routines for creating plans for 2d and 3d transforms,
and one routine for creating plans of arbitrary dimensionality.
The 2d and 3d routines have the following signature:
@example
fftw_plan fftw_plan_dft_2d(int n0, int n1,
                           fftw_complex *in, fftw_complex *out,
                           int sign, unsigned flags);
fftw_plan fftw_plan_dft_3d(int n0, int n1, int n2,
                           fftw_complex *in, fftw_complex *out,
                           int sign, unsigned flags);
@end example
@findex fftw_plan_dft_2d
@findex fftw_plan_dft_3d

These routines create plans for @code{n0} by @code{n1} two-dimensional
(2d) transforms and @code{n0} by @code{n1} by @code{n2} 3d transforms,
respectively.  All of these transforms operate on contiguous arrays in
the C-standard @dfn{row-major} order, so that the last dimension has the
fastest-varying index in the array.  This layout is described further in
@ref{Multi-dimensional Array Format}.

FFTW can also compute transforms of higher dimensionality.  In order to
avoid confusion between the various meanings of the the word
``dimension'', we use the term @emph{rank}
@cindex rank
to denote the number of independent indices in an array.@footnote{The
term ``rank'' is commonly used in the APL, FORTRAN, and Common Lisp
traditions, although it is not so common in the C@tie{}world.}  For
example, we say that a 2d transform has rank@tie{}2, a 3d transform has
rank@tie{}3, and so on.  You can plan transforms of arbitrary rank by
means of the following function:

@example
fftw_plan fftw_plan_dft(int rank, const int *n,
                        fftw_complex *in, fftw_complex *out,
                        int sign, unsigned flags);
@end example
@findex fftw_plan_dft

Here, @code{n} is a pointer to an array @code{n[rank]} denoting an
@code{n[0]} by @code{n[1]} by @dots{} by @code{n[rank-1]} transform.
Thus, for example, the call
@example
fftw_plan_dft_2d(n0, n1, in, out, sign, flags);
@end example
is equivalent to the following code fragment:
@example
int n[2];
n[0] = n0;
n[1] = n1;
fftw_plan_dft(2, n, in, out, sign, flags);
@end example
@code{fftw_plan_dft} is not restricted to 2d and 3d transforms,
however, but it can plan transforms of arbitrary rank.

You may have noticed that all the planner routines described so far
have overlapping functionality.  For example, you can plan a 1d or 2d
transform by using @code{fftw_plan_dft} with a @code{rank} of @code{1}
or @code{2}, or even by calling @code{fftw_plan_dft_3d} with @code{n0}
and/or @code{n1} equal to @code{1} (with no loss in efficiency).  This
pattern continues, and FFTW's planning routines in general form a
``partial order,'' sequences of
@cindex partial order
interfaces with strictly increasing generality but correspondingly
greater complexity.

@code{fftw_plan_dft} is the most general complex-DFT routine that we
describe in this tutorial, but there are also the advanced and guru interfaces,
@cindex advanced interface
@cindex guru interface 
which allow one to efficiently combine multiple/strided transforms
into a single FFTW plan, transform a subset of a larger
multi-dimensional array, and/or to handle more general complex-number
formats.  For more information, see @ref{FFTW Reference}.

@c ------------------------------------------------------------
@node One-Dimensional DFTs of Real Data, Multi-Dimensional DFTs of Real Data, Complex Multi-Dimensional DFTs, Tutorial
@section One-Dimensional DFTs of Real Data

In many practical applications, the input data @code{in[i]} are purely
real numbers, in which case the DFT output satisfies the ``Hermitian''
@cindex Hermitian
redundancy: @code{out[i]} is the conjugate of @code{out[n-i]}.  It is
possible to take advantage of these circumstances in order to achieve
roughly a factor of two improvement in both speed and memory usage.

In exchange for these speed and space advantages, the user sacrifices
some of the simplicity of FFTW's complex transforms. First of all, the
input and output arrays are of @emph{different sizes and types}: the
input is @code{n} real numbers, while the output is @code{n/2+1}
complex numbers (the non-redundant outputs); this also requires slight
``padding'' of the input array for
@cindex padding
in-place transforms.  Second, the inverse transform (complex to real)
has the side-effect of @emph{overwriting its input array}, by default.
Neither of these inconveniences should pose a serious problem for
users, but it is important to be aware of them.

The routines to perform real-data transforms are almost the same as
those for complex transforms: you allocate arrays of @code{double}
and/or @code{fftw_complex} (preferably using @code{fftw_malloc} or
@code{fftw_alloc_complex}), create an @code{fftw_plan}, execute it as
many times as you want with @code{fftw_execute(plan)}, and clean up
with @code{fftw_destroy_plan(plan)} (and @code{fftw_free}).  The only
differences are that the input (or output) is of type @code{double}
and there are new routines to create the plan.  In one dimension:

@example
fftw_plan fftw_plan_dft_r2c_1d(int n, double *in, fftw_complex *out,
                               unsigned flags);
fftw_plan fftw_plan_dft_c2r_1d(int n, fftw_complex *in, double *out,
                               unsigned flags);
@end example
@findex fftw_plan_dft_r2c_1d
@findex fftw_plan_dft_c2r_1d

for the real input to complex-Hermitian output (@dfn{r2c}) and
complex-Hermitian input to real output (@dfn{c2r}) transforms.
@cindex r2c
@cindex c2r
Unlike the complex DFT planner, there is no @code{sign} argument.
Instead, r2c DFTs are always @code{FFTW_FORWARD} and c2r DFTs are
always @code{FFTW_BACKWARD}.
@ctindex FFTW_FORWARD
@ctindex FFTW_BACKWARD
(For single/long-double precision
@code{fftwf} and @code{fftwl}, @code{double} should be replaced by
@code{float} and @code{long double}, respectively.)
@cindex precision


Here, @code{n} is the ``logical'' size of the DFT, not necessarily the
physical size of the array.  In particular, the real (@code{double})
array has @code{n} elements, while the complex (@code{fftw_complex})
array has @code{n/2+1} elements (where the division is rounded down).
For an in-place transform,
@cindex in-place
@code{in} and @code{out} are aliased to the same array, which must be
big enough to hold both; so, the real array would actually have
@code{2*(n/2+1)} elements, where the elements beyond the first
@code{n} are unused padding.  (Note that this is very different from
the concept of ``zero-padding'' a transform to a larger length, which
changes the logical size of the DFT by actually adding new input
data.)  The @math{k}th element of the complex array is exactly the
same as the @math{k}th element of the corresponding complex DFT.  All
positive @code{n} are supported; products of small factors are most
efficient, but an @Onlogn algorithm is used even for prime sizes.

As noted above, the c2r transform destroys its input array even for
out-of-place transforms.  This can be prevented, if necessary, by
including @code{FFTW_PRESERVE_INPUT} in the @code{flags}, with
unfortunately some sacrifice in performance.
@cindex flags
@ctindex FFTW_PRESERVE_INPUT
This flag is also not currently supported for multi-dimensional real
DFTs (next section).

Readers familiar with DFTs of real data will recall that the 0th (the
``DC'') and @code{n/2}-th (the ``Nyquist'' frequency, when @code{n} is
even) elements of the complex output are purely real.  Some
implementations therefore store the Nyquist element where the DC
imaginary part would go, in order to make the input and output arrays
the same size.  Such packing, however, does not generalize well to
multi-dimensional transforms, and the space savings are miniscule in
any case; FFTW does not support it.

An alternative interface for one-dimensional r2c and c2r DFTs can be
found in the @samp{r2r} interface (@pxref{The Halfcomplex-format
DFT}), with ``halfcomplex''-format output that @emph{is} the same size
(and type) as the input array.
@cindex halfcomplex format
That interface, although it is not very useful for multi-dimensional
transforms, may sometimes yield better performance.

@c ------------------------------------------------------------
@node Multi-Dimensional DFTs of Real Data, More DFTs of Real Data, One-Dimensional DFTs of Real Data, Tutorial
@section Multi-Dimensional DFTs of Real Data

Multi-dimensional DFTs of real data use the following planner routines:

@example
fftw_plan fftw_plan_dft_r2c_2d(int n0, int n1,
                               double *in, fftw_complex *out,
                               unsigned flags);
fftw_plan fftw_plan_dft_r2c_3d(int n0, int n1, int n2,
                               double *in, fftw_complex *out,
                               unsigned flags);
fftw_plan fftw_plan_dft_r2c(int rank, const int *n,
                            double *in, fftw_complex *out,
                            unsigned flags);
@end example
@findex fftw_plan_dft_r2c_2d
@findex fftw_plan_dft_r2c_3d
@findex fftw_plan_dft_r2c

as well as the corresponding @code{c2r} routines with the input/output
types swapped.  These routines work similarly to their complex
analogues, except for the fact that here the complex output array is cut
roughly in half and the real array requires padding for in-place
transforms (as in 1d, above).

As before, @code{n} is the logical size of the array, and the
consequences of this on the the format of the complex arrays deserve
careful attention.
@cindex r2c/c2r multi-dimensional array format
Suppose that the real data has dimensions @ndims (in row-major order).
Then, after an r2c transform, the output is an @ndimshalf array of
@code{fftw_complex} values in row-major order, corresponding to slightly
over half of the output of the corresponding complex DFT.  (The division
is rounded down.)  The ordering of the data is otherwise exactly the
same as in the complex-DFT case.

For out-of-place transforms, this is the end of the story: the real
data is stored as a row-major array of size @ndims and the complex
data is stored as a row-major array of size @ndimshalf{}.

For in-place transforms, however, extra padding of the real-data array
is necessary because the complex array is larger than the real array,
and the two arrays share the same memory locations.  Thus, for
in-place transforms, the final dimension of the real-data array must
be padded with extra values to accommodate the size of the complex
data---two values if the last dimension is even and one if it is odd.
@cindex padding
That is, the last dimension of the real data must physically contain
@tex
$2 (n_{d-1}/2+1)$
@end tex
@ifinfo
2 * (n[d-1]/2+1)
@end ifinfo
@html
2 * (n<sub>d-1</sub>/2+1)
@end html
@code{double} values (exactly enough to hold the complex data).
This physical array size does not, however, change the @emph{logical}
array size---only
@tex
$n_{d-1}$
@end tex
@ifinfo
n[d-1]
@end ifinfo
@html
n<sub>d-1</sub>
@end html
values are actually stored in the last dimension, and
@tex
$n_{d-1}$
@end tex
@ifinfo
n[d-1]
@end ifinfo
@html
n<sub>d-1</sub>
@end html
is the last dimension passed to the plan-creation routine.

For example, consider the transform of a two-dimensional real array of
size @code{n0} by @code{n1}.  The output of the r2c transform is a
two-dimensional complex array of size @code{n0} by @code{n1/2+1}, where
the @code{y} dimension has been cut nearly in half because of
redundancies in the output.  Because @code{fftw_complex} is twice the
size of @code{double}, the output array is slightly bigger than the
input array.  Thus, if we want to compute the transform in place, we
must @emph{pad} the input array so that it is of size @code{n0} by
@code{2*(n1/2+1)}.  If @code{n1} is even, then there are two padding
elements at the end of each row (which need not be initialized, as they
are only used for output).

@ifhtml
The following illustration depicts the input and output arrays just
described, for both the out-of-place and in-place transforms (with the
arrows indicating consecutive memory locations):
@image{rfftwnd-for-html}
@end ifhtml
@ifnotinfo
@ifnothtml
@float Figure,fig:rfftwnd
@center @image{rfftwnd}
@caption{Illustration of the data layout for a 2d @code{nx} by @code{ny}
real-to-complex transform.}
@end float
@ref{fig:rfftwnd} depicts the input and output arrays just
described, for both the out-of-place and in-place transforms (with the
arrows indicating consecutive memory locations):
@end ifnothtml
@end ifnotinfo

These transforms are unnormalized, so an r2c followed by a c2r
transform (or vice versa) will result in the original data scaled by
the number of real data elements---that is, the product of the
(logical) dimensions of the real data.
@cindex normalization


(Because the last dimension is treated specially, if it is equal to
@code{1} the transform is @emph{not} equivalent to a lower-dimensional
r2c/c2r transform.  In that case, the last complex dimension also has
size @code{1} (@code{=1/2+1}), and no advantage is gained over the
complex transforms.)

@c ------------------------------------------------------------
@node More DFTs of Real Data,  , Multi-Dimensional DFTs of Real Data, Tutorial
@section More DFTs of Real Data
@menu
* The Halfcomplex-format DFT::
* Real even/odd DFTs (cosine/sine transforms)::
* The Discrete Hartley Transform::
@end menu

FFTW supports several other transform types via a unified @dfn{r2r}
(real-to-real) interface,
@cindex r2r
so called because it takes a real (@code{double}) array and outputs a
real array of the same size.  These r2r transforms currently fall into
three categories: DFTs of real input and complex-Hermitian output in
halfcomplex format, DFTs of real input with even/odd symmetry
(a.k.a. discrete cosine/sine transforms, DCTs/DSTs), and discrete
Hartley transforms (DHTs), all described in more detail by the
following sections.

The r2r transforms follow the by now familiar interface of creating an
@code{fftw_plan}, executing it with @code{fftw_execute(plan)}, and
destroying it with @code{fftw_destroy_plan(plan)}.  Furthermore, all
r2r transforms share the same planner interface:

@example
fftw_plan fftw_plan_r2r_1d(int n, double *in, double *out,
                           fftw_r2r_kind kind, unsigned flags);
fftw_plan fftw_plan_r2r_2d(int n0, int n1, double *in, double *out,
                           fftw_r2r_kind kind0, fftw_r2r_kind kind1,
                           unsigned flags);
fftw_plan fftw_plan_r2r_3d(int n0, int n1, int n2,
                           double *in, double *out,
                           fftw_r2r_kind kind0,
                           fftw_r2r_kind kind1,
                           fftw_r2r_kind kind2,
                           unsigned flags);
fftw_plan fftw_plan_r2r(int rank, const int *n, double *in, double *out,
                        const fftw_r2r_kind *kind, unsigned flags);
@end example
@findex fftw_plan_r2r_1d
@findex fftw_plan_r2r_2d
@findex fftw_plan_r2r_3d
@findex fftw_plan_r2r

Just as for the complex DFT, these plan 1d/2d/3d/multi-dimensional
transforms for contiguous arrays in row-major order, transforming (real)
input to output of the same size, where @code{n} specifies the
@emph{physical} dimensions of the arrays.  All positive @code{n} are
supported (with the exception of @code{n=1} for the @code{FFTW_REDFT00}
kind, noted in the real-even subsection below); products of small
factors are most efficient (factorizing @code{n-1} and @code{n+1} for
@code{FFTW_REDFT00} and @code{FFTW_RODFT00} kinds, described below), but
an @Onlogn algorithm is used even for prime sizes.

Each dimension has a @dfn{kind} parameter, of type
@code{fftw_r2r_kind}, specifying the kind of r2r transform to be used
for that dimension.
@cindex kind (r2r)
@tindex fftw_r2r_kind
(In the case of @code{fftw_plan_r2r}, this is an array @code{kind[rank]}
where @code{kind[i]} is the transform kind for the dimension
@code{n[i]}.)  The kind can be one of a set of predefined constants,
defined in the following subsections.

In other words, FFTW computes the separable product of the specified
r2r transforms over each dimension, which can be used e.g. for partial
differential equations with mixed boundary conditions.  (For some r2r
kinds, notably the halfcomplex DFT and the DHT, such a separable
product is somewhat problematic in more than one dimension, however,
as is described below.)

In the current version of FFTW, all r2r transforms except for the
halfcomplex type are computed via pre- or post-processing of
halfcomplex transforms, and they are therefore not as fast as they
could be.  Since most other general DCT/DST codes employ a similar
algorithm, however, FFTW's implementation should provide at least
competitive performance.

@c =========>
@node The Halfcomplex-format DFT, Real even/odd DFTs (cosine/sine transforms), More DFTs of Real Data, More DFTs of Real Data
@subsection The Halfcomplex-format DFT

An r2r kind of @code{FFTW_R2HC} (@dfn{r2hc}) corresponds to an r2c DFT
@ctindex FFTW_R2HC
@cindex r2c
@cindex r2hc
(@pxref{One-Dimensional DFTs of Real Data}) but with ``halfcomplex''
format output, and may sometimes be faster and/or more convenient than
the latter.
@cindex halfcomplex format
The inverse @dfn{hc2r} transform is of kind @code{FFTW_HC2R}.
@ctindex FFTW_HC2R
@cindex hc2r
This consists of the non-redundant half of the complex output for a 1d
real-input DFT of size @code{n}, stored as a sequence of @code{n} real
numbers (@code{double}) in the format:

@tex
$$
r_0, r_1, r_2, \ldots, r_{n/2}, i_{(n+1)/2-1}, \ldots, i_2, i_1
$$
@end tex
@ifinfo
r0, r1, r2, r(n/2), i((n+1)/2-1), ..., i2, i1
@end ifinfo
@html
<p align=center>
r<sub>0</sub>, r<sub>1</sub>, r<sub>2</sub>, ..., r<sub>n/2</sub>, i<sub>(n+1)/2-1</sub>, ..., i<sub>2</sub>, i<sub>1</sub>
</p>
@end html

Here,
@ifinfo
rk
@end ifinfo
@tex
$r_k$
@end tex
@html
r<sub>k</sub>
@end html
is the real part of the @math{k}th output, and
@ifinfo
ik
@end ifinfo
@tex
$i_k$
@end tex
@html
i<sub>k</sub>
@end html
is the imaginary part.  (Division by 2 is rounded down.) For a
halfcomplex array @code{hc[n]}, the @math{k}th component thus has its
real part in @code{hc[k]} and its imaginary part in @code{hc[n-k]}, with
the exception of @code{k} @code{==} @code{0} or @code{n/2} (the latter
only if @code{n} is even)---in these two cases, the imaginary part is
zero due to symmetries of the real-input DFT, and is not stored.
Thus, the r2hc transform of @code{n} real values is a halfcomplex array of
length @code{n}, and vice versa for hc2r.
@cindex normalization


Aside from the differing format, the output of
@code{FFTW_R2HC}/@code{FFTW_HC2R} is otherwise exactly the same as for
the corresponding 1d r2c/c2r transform
(i.e. @code{FFTW_FORWARD}/@code{FFTW_BACKWARD} transforms, respectively).
Recall that these transforms are unnormalized, so r2hc followed by hc2r
will result in the original data multiplied by @code{n}.  Furthermore,
like the c2r transform, an out-of-place hc2r transform will
@emph{destroy its input} array.

Although these halfcomplex transforms can be used with the
multi-dimensional r2r interface, the interpretation of such a separable
product of transforms along each dimension is problematic.  For example,
consider a two-dimensional @code{n0} by @code{n1}, r2hc by r2hc
transform planned by @code{fftw_plan_r2r_2d(n0, n1, in, out, FFTW_R2HC,
FFTW_R2HC, FFTW_MEASURE)}.  Conceptually, FFTW first transforms the rows
(of size @code{n1}) to produce halfcomplex rows, and then transforms the
columns (of size @code{n0}).  Half of these column transforms, however,
are of imaginary parts, and should therefore be multiplied by @math{i}
and combined with the r2hc transforms of the real columns to produce the
2d DFT amplitudes; FFTW's r2r transform does @emph{not} perform this
combination for you.  Thus, if a multi-dimensional real-input/output DFT
is required, we recommend using the ordinary r2c/c2r
interface (@pxref{Multi-Dimensional DFTs of Real Data}).

@c =========>
@node Real even/odd DFTs (cosine/sine transforms), The Discrete Hartley Transform, The Halfcomplex-format DFT, More DFTs of Real Data
@subsection Real even/odd DFTs (cosine/sine transforms)

The Fourier transform of a real-even function @math{f(-x) = f(x)} is
real-even, and @math{i} times the Fourier transform of a real-odd
function @math{f(-x) = -f(x)} is real-odd.  Similar results hold for a
discrete Fourier transform, and thus for these symmetries the need for
complex inputs/outputs is entirely eliminated.  Moreover, one gains a
factor of two in speed/space from the fact that the data are real, and
an additional factor of two from the even/odd symmetry: only the
non-redundant (first) half of the array need be stored.  The result is
the real-even DFT (@dfn{REDFT}) and the real-odd DFT (@dfn{RODFT}), also
known as the discrete cosine and sine transforms (@dfn{DCT} and
@dfn{DST}), respectively.
@cindex real-even DFT
@cindex REDFT
@cindex real-odd DFT
@cindex RODFT
@cindex discrete cosine transform
@cindex DCT
@cindex discrete sine transform
@cindex DST


(In this section, we describe the 1d transforms; multi-dimensional
transforms are just a separable product of these transforms operating
along each dimension.)

Because of the discrete sampling, one has an additional choice: is the
data even/odd around a sampling point, or around the point halfway
between two samples?  The latter corresponds to @emph{shifting} the
samples by @emph{half} an interval, and gives rise to several transform
variants denoted by REDFT@math{ab} and RODFT@math{ab}: @math{a} and
@math{b} are @math{0} or @math{1}, and indicate whether the input
(@math{a}) and/or output (@math{b}) are shifted by half a sample
(@math{1} means it is shifted).  These are also known as types I-IV of
the DCT and DST, and all four types are supported by FFTW's r2r
interface.@footnote{There are also type V-VIII transforms, which
correspond to a logical DFT of @emph{odd} size @math{N}, independent of
whether the physical size @code{n} is odd, but we do not support these
variants.}

The r2r kinds for the various REDFT and RODFT types supported by FFTW,
along with the boundary conditions at both ends of the @emph{input}
array (@code{n} real numbers @code{in[j=0..n-1]}), are:

@itemize @bullet

@item
@code{FFTW_REDFT00} (DCT-I): even around @math{j=0} and even around @math{j=n-1}.
@ctindex FFTW_REDFT00

@item
@code{FFTW_REDFT10} (DCT-II, ``the'' DCT): even around @math{j=-0.5} and even around @math{j=n-0.5}.
@ctindex FFTW_REDFT10

@item
@code{FFTW_REDFT01} (DCT-III, ``the'' IDCT): even around @math{j=0} and odd around @math{j=n}.
@ctindex FFTW_REDFT01
@cindex IDCT

@item
@code{FFTW_REDFT11} (DCT-IV): even around @math{j=-0.5} and odd around @math{j=n-0.5}.
@ctindex FFTW_REDFT11

@item
@code{FFTW_RODFT00} (DST-I): odd around @math{j=-1} and odd around @math{j=n}.
@ctindex FFTW_RODFT00

@item
@code{FFTW_RODFT10} (DST-II): odd around @math{j=-0.5} and odd around @math{j=n-0.5}.
@ctindex FFTW_RODFT10

@item
@code{FFTW_RODFT01} (DST-III): odd around @math{j=-1} and even around @math{j=n-1}.
@ctindex FFTW_RODFT01

@item
@code{FFTW_RODFT11} (DST-IV): odd around @math{j=-0.5} and even around @math{j=n-0.5}.
@ctindex FFTW_RODFT11

@end itemize

Note that these symmetries apply to the ``logical'' array being
transformed; @strong{there are no constraints on your physical input
data}.  So, for example, if you specify a size-5 REDFT00 (DCT-I) of the
data @math{abcde}, it corresponds to the DFT of the logical even array
@math{abcdedcb} of size 8.  A size-4 REDFT10 (DCT-II) of the data
@math{abcd} corresponds to the size-8 logical DFT of the even array
@math{abcddcba}, shifted by half a sample.

All of these transforms are invertible.  The inverse of R*DFT00 is
R*DFT00; of R*DFT10 is R*DFT01 and vice versa (these are often called
simply ``the'' DCT and IDCT, respectively); and of R*DFT11 is R*DFT11.
However, the transforms computed by FFTW are unnormalized, exactly
like the corresponding real and complex DFTs, so computing a transform
followed by its inverse yields the original array scaled by @math{N},
where @math{N} is the @emph{logical} DFT size.  For REDFT00,
@math{N=2(n-1)}; for RODFT00, @math{N=2(n+1)}; otherwise, @math{N=2n}.
@cindex normalization
@cindex IDCT


Note that the boundary conditions of the transform output array are
given by the input boundary conditions of the inverse transform.
Thus, the above transforms are all inequivalent in terms of
input/output boundary conditions, even neglecting the 0.5 shift
difference.

FFTW is most efficient when @math{N} is a product of small factors; note
that this @emph{differs} from the factorization of the physical size
@code{n} for REDFT00 and RODFT00!  There is another oddity: @code{n=1}
REDFT00 transforms correspond to @math{N=0}, and so are @emph{not
defined} (the planner will return @code{NULL}).  Otherwise, any positive
@code{n} is supported.

For the precise mathematical definitions of these transforms as used by
FFTW, see @ref{What FFTW Really Computes}.  (For people accustomed to
the DCT/DST, FFTW's definitions have a coefficient of @math{2} in front
of the cos/sin functions so that they correspond precisely to an
even/odd DFT of size @math{N}.  Some authors also include additional
multiplicative factors of 
@ifinfo
sqrt(2)
@end ifinfo
@html
&radic;2
@end html
@tex
$\sqrt{2}$
@end tex
for selected inputs and outputs; this makes
the transform orthogonal, but sacrifices the direct equivalence to a
symmetric DFT.)

@subsubheading Which type do you need?

Since the required flavor of even/odd DFT depends upon your problem,
you are the best judge of this choice, but we can make a few comments
on relative efficiency to help you in your selection.  In particular,
R*DFT01 and R*DFT10 tend to be slightly faster than R*DFT11
(especially for odd sizes), while the R*DFT00 transforms are sometimes
significantly slower (especially for even sizes).@footnote{R*DFT00 is
sometimes slower in FFTW because we discovered that the standard
algorithm for computing this by a pre/post-processed real DFT---the
algorithm used in FFTPACK, Numerical Recipes, and other sources for
decades now---has serious numerical problems: it already loses several
decimal places of accuracy for 16k sizes.  There seem to be only two
alternatives in the literature that do not suffer similarly: a
recursive decomposition into smaller DCTs, which would require a large
set of codelets for efficiency and generality, or sacrificing a factor of 
@tex
$\sim 2$
@end tex
@ifnottex
2
@end ifnottex
in speed to use a real DFT of twice the size.  We currently
employ the latter technique for general @math{n}, as well as a limited
form of the former method: a split-radix decomposition when @math{n}
is odd (@math{N} a multiple of 4).  For @math{N} containing many
factors of 2, the split-radix method seems to recover most of the
speed of the standard algorithm without the accuracy tradeoff.}

Thus, if only the boundary conditions on the transform inputs are
specified, we generally recommend R*DFT10 over R*DFT00 and R*DFT01 over
R*DFT11 (unless the half-sample shift or the self-inverse property is
significant for your problem).

If performance is important to you and you are using only small sizes
(say @math{n<200}), e.g. for multi-dimensional transforms, then you
might consider generating hard-coded transforms of those sizes and types
that you are interested in (@pxref{Generating your own code}).

We are interested in hearing what types of symmetric transforms you find
most useful.

@c =========>
@node The Discrete Hartley Transform,  , Real even/odd DFTs (cosine/sine transforms), More DFTs of Real Data
@subsection The Discrete Hartley Transform

If you are planning to use the DHT because you've heard that it is
``faster'' than the DFT (FFT), @strong{stop here}.  The DHT is not
faster than the DFT.  That story is an old but enduring misconception
that was debunked in 1987.

The discrete Hartley transform (DHT) is an invertible linear transform
closely related to the DFT.  In the DFT, one multiplies each input by
@math{cos - i * sin} (a complex exponential), whereas in the DHT each
input is multiplied by simply @math{cos + sin}.  Thus, the DHT
transforms @code{n} real numbers to @code{n} real numbers, and has the
convenient property of being its own inverse.  In FFTW, a DHT (of any
positive @code{n}) can be specified by an r2r kind of @code{FFTW_DHT}.
@ctindex FFTW_DHT
@cindex discrete Hartley transform
@cindex DHT

Like the DFT, in FFTW the DHT is unnormalized, so computing a DHT of
size @code{n} followed by another DHT of the same size will result in
the original array multiplied by @code{n}.
@cindex normalization

The DHT was originally proposed as a more efficient alternative to the
DFT for real data, but it was subsequently shown that a specialized DFT
(such as FFTW's r2hc or r2c transforms) could be just as fast.  In FFTW,
the DHT is actually computed by post-processing an r2hc transform, so
there is ordinarily no reason to prefer it from a performance
perspective.@footnote{We provide the DHT mainly as a byproduct of some
internal algorithms. FFTW computes a real input/output DFT of
@emph{prime} size by re-expressing it as a DHT plus post/pre-processing
and then using Rader's prime-DFT algorithm adapted to the DHT.}
However, we have heard rumors that the DHT might be the most appropriate
transform in its own right for certain applications, and we would be
very interested to hear from anyone who finds it useful.

If @code{FFTW_DHT} is specified for multiple dimensions of a
multi-dimensional transform, FFTW computes the separable product of 1d
DHTs along each dimension.  Unfortunately, this is not quite the same
thing as a true multi-dimensional DHT; you can compute the latter, if
necessary, with at most @code{rank-1} post-processing passes
[see e.g. H. Hao and R. N. Bracewell, @i{Proc. IEEE} @b{75}, 264--266 (1987)].

For the precise mathematical definition of the DHT as used by FFTW, see
@ref{What FFTW Really Computes}.