------=_Part_1647_5324833.1352964921902
Content-Type: text/plain; charset=ISO-8859-1

Hello.

Now that C++ has a good numerical library for random number generation, I
wonder what will come next concerning numerical utilities. More precisely,
from my point of view, there are three types of operations that are used by
a wide range of topics :
- statistics :
- interpolation
- fft (fast fourier transform)
It would be great to have standardized numerical algorithms for these
operations.

Do you now if some of these have already been discussed and may be included
in the next major version ?

Thank you very much.
Vincent Reverdy

--




------=_Part_1647_5324833.1352964921902
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Hello.<br><br>Now that C++ has a good numerical library for random number g=
eneration, I wonder what will come next concerning numerical utilities. Mor=
e precisely, from my point of view, there are three types of operations tha=
t are used by a wide range of topics :<br>- statistics : <br>- interpolatio=
n<br>- fft (fast fourier transform)<br>It would be great to have standardiz=
ed numerical algorithms for these operations.<br><br>Do you now if some of =
these have already been discussed and may be included in the next major ver=
sion ?<br><br>Thank you very much.<br>Vincent Reverdy<br>

<p></p>

-- <br />
&nbsp;<br />
&nbsp;<br />
&nbsp;<br />

------=_Part_1647_5324833.1352964921902--

.

On Thu, Nov 15, 2012 at 6:35 PM, VinceRev <vince.rev@gmail.com> wrote:
> Hello.
>
> Now that C++ has a good numerical library for random number generation, I
> wonder what will come next concerning numerical utilities. More precisely,
> from my point of view, there are three types of operations that are used by
> a wide range of topics :
> - statistics :
> - interpolation
> - fft (fast fourier transform)
> It would be great to have standardized numerical algorithms for these
> operations.
>
> Do you now if some of these have already been discussed and may be included
> in the next major version ?
>

I don't know of any but I would really love to have these too.

--
Dean Michael Berris | Software Engineer
Google

--




.

------=_Part_342_25697439.1352966351841
Content-Type: text/plain; charset=ISO-8859-1

It's wonderful to want things, but they don't just happen. Random number
generation happened because someone wrote Boost.Random, and something not
entirely unlike that was accepted into C++11.

If you want this to happen, you're going to need to at the very least write
a real proposal, preferably with some proof of concept for the interface.

On Wednesday, November 14, 2012 11:35:22 PM UTC-8, VinceRev wrote:
>
> Hello.
>
> Now that C++ has a good numerical library for random number generation, I
> wonder what will come next concerning numerical utilities. More precisely,
> from my point of view, there are three types of operations that are used by
> a wide range of topics :
> - statistics :
> - interpolation
> - fft (fast fourier transform)
> It would be great to have standardized numerical algorithms for these
> operations.
>
> Do you now if some of these have already been discussed and may be
> included in the next major version ?
>
> Thank you very much.
> Vincent Reverdy
>

--




------=_Part_342_25697439.1352966351841
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

It's wonderful to want things, but they don't just happen. Random number ge=
neration happened because someone wrote Boost.Random, and something not ent=
irely unlike that was accepted into C++11.<br><br>If you want this to happe=
n, you're going to need to at the very least write a real proposal, prefera=
bly with some proof of concept for the interface.<br><br>On Wednesday, Nove=
mber 14, 2012 11:35:22 PM UTC-8, VinceRev wrote:<blockquote class=3D"gmail_=
quote" style=3D"margin: 0;margin-left: 0.8ex;border-left: 1px #ccc solid;pa=
dding-left: 1ex;">Hello.<br><br>Now that C++ has a good numerical library f=
or random number generation, I wonder what will come next concerning numeri=
cal utilities. More precisely, from my point of view, there are three types=
 of operations that are used by a wide range of topics :<br>- statistics : =
<br>- interpolation<br>- fft (fast fourier transform)<br>It would be great =
to have standardized numerical algorithms for these operations.<br><br>Do y=
ou now if some of these have already been discussed and may be included in =
the next major version ?<br><br>Thank you very much.<br>Vincent Reverdy<br>=
</blockquote>

<p></p>

-- <br />
&nbsp;<br />
&nbsp;<br />
&nbsp;<br />

------=_Part_342_25697439.1352966351841--

.

--047d7b6049baea44cb04ce8494d1
Content-Type: text/plain; charset=ISO-8859-1

AFAIK, effective implementation of fft requires techniques hardly
compatible with c++ standard library approach, e.g., code generation. For
details see fftw as one of the fastest implementations out there.

And after all, does it make sense to have fft without other DSP algorithms
like windowing functions, filters, resampling, decimation etc.?
On Nov 14, 2012 11:35 PM, "VinceRev" <vince.rev@gmail.com> wrote:

> Hello.
>
> Now that C++ has a good numerical library for random number generation, I
> wonder what will come next concerning numerical utilities. More precisely,
> from my point of view, there are three types of operations that are used by
> a wide range of topics :
> - statistics :
> - interpolation
> - fft (fast fourier transform)
> It would be great to have standardized numerical algorithms for these
> operations.
>
> Do you now if some of these have already been discussed and may be
> included in the next major version ?
>
> Thank you very much.
> Vincent Reverdy
>
> --
>
>
>
>

--




--047d7b6049baea44cb04ce8494d1
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<p dir=3D"ltr">AFAIK, effective implementation of fft requires techniques h=
ardly compatible with c++ standard library approach, e.g., code generation.=
 For details see fftw as one of the fastest implementations out there.</p>

<p dir=3D"ltr">And after all, does it make sense to have fft without other =
DSP algorithms like windowing functions, filters, resampling, decimation et=
c.?</p>
<div class=3D"gmail_quote">On Nov 14, 2012 11:35 PM, &quot;VinceRev&quot; &=
lt;<a href=3D"mailto:vince.rev@gmail.com">vince.rev@gmail.com</a>&gt; wrote=
:<br type=3D"attribution"><blockquote class=3D"gmail_quote" style=3D"margin=
:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Hello.<br><br>Now that C++ has a good numerical library for random number g=
eneration, I wonder what will come next concerning numerical utilities. Mor=
e precisely, from my point of view, there are three types of operations tha=
t are used by a wide range of topics :<br>
- statistics : <br>- interpolation<br>- fft (fast fourier transform)<br>It =
would be great to have standardized numerical algorithms for these operatio=
ns.<br><br>Do you now if some of these have already been discussed and may =
be included in the next major version ?<br>
<br>Thank you very much.<br>Vincent Reverdy<br>

<p></p>

-- <br>
=A0<br>
=A0<br>
=A0<br>
</blockquote></div>

<p></p>

-- <br />
&nbsp;<br />
&nbsp;<br />
&nbsp;<br />

--047d7b6049baea44cb04ce8494d1--

.

--f46d0401692391345404ce88bd71
Content-Type: text/plain; charset=ISO-8859-1

I think statistics is probably more basic that specific algorithms as FFT.

In any case, there is a study group on Numerics (SG6). That's the right
place where any proposal should go in first place.


--
  J. Daniel


On Thu, Nov 15, 2012 at 2:38 AM, Alexander Kornienko <alexfh@google.com>wrote:

> AFAIK, effective implementation of fft requires techniques hardly
> compatible with c++ standard library approach, e.g., code generation. For
> details see fftw as one of the fastest implementations out there.
>
> And after all, does it make sense to have fft without other DSP algorithms
> like windowing functions, filters, resampling, decimation etc.?
> On Nov 14, 2012 11:35 PM, "VinceRev" <vince.rev@gmail.com> wrote:
>
>> Hello.
>>
>> Now that C++ has a good numerical library for random number generation, I
>> wonder what will come next concerning numerical utilities. More precisely,
>> from my point of view, there are three types of operations that are used by
>> a wide range of topics :
>> - statistics :
>> - interpolation
>> - fft (fast fourier transform)
>> It would be great to have standardized numerical algorithms for these
>> operations.
>>
>> Do you now if some of these have already been discussed and may be
>> included in the next major version ?
>>
>> Thank you very much.
>> Vincent Reverdy
>>
>> --
>>
>>
>>
>>
>  --
>
>
>
>

--




--f46d0401692391345404ce88bd71
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

I think statistics is probably more basic that specific algorithms as FFT.<=
div><br></div><div>In any case, there is a study group on Numerics (SG6). T=
hat&#39;s the right place where any proposal should go in first place.</div=
>

<div><br></div><div><br></div><div>--</div><div>=A0 J. Daniel</div><div cla=
ss=3D"gmail_extra"><br><br><div class=3D"gmail_quote">On Thu, Nov 15, 2012 =
at 2:38 AM, Alexander Kornienko <span dir=3D"ltr">&lt;<a href=3D"mailto:ale=
xfh@google.com" target=3D"_blank">alexfh@google.com</a>&gt;</span> wrote:<b=
r>

<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><p dir=3D"ltr">AFAIK, effective implementati=
on of fft requires techniques hardly compatible with c++ standard library a=
pproach, e.g., code generation. For details see fftw as one of the fastest =
implementations out there.</p>



<p dir=3D"ltr">And after all, does it make sense to have fft without other =
DSP algorithms like windowing functions, filters, resampling, decimation et=
c.?</p><div><div class=3D"h5">
<div class=3D"gmail_quote">On Nov 14, 2012 11:35 PM, &quot;VinceRev&quot; &=
lt;<a href=3D"mailto:vince.rev@gmail.com" target=3D"_blank">vince.rev@gmail=
..com</a>&gt; wrote:<br type=3D"attribution"><blockquote class=3D"gmail_quot=
e" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">


Hello.<br><br>Now that C++ has a good numerical library for random number g=
eneration, I wonder what will come next concerning numerical utilities. Mor=
e precisely, from my point of view, there are three types of operations tha=
t are used by a wide range of topics :<br>


- statistics : <br>- interpolation<br>- fft (fast fourier transform)<br>It =
would be great to have standardized numerical algorithms for these operatio=
ns.<br><br>Do you now if some of these have already been discussed and may =
be included in the next major version ?<br>


<br>Thank you very much.<br>Vincent Reverdy<br>

<p></p>

-- <br>
=A0<br>
=A0<br>
=A0<br>
</blockquote></div>

<p></p></div></div>

-- <br>
=A0<br>
=A0<br>
=A0<br>
</blockquote></div><br><br clear=3D"all"><div><br></div><br>
</div>

<p></p>

-- <br />
&nbsp;<br />
&nbsp;<br />
&nbsp;<br />

--f46d0401692391345404ce88bd71--

.

On 11/15/12, J. Daniel Garcia <josedaniel.garcia@uc3m.es> wrote:
> I think statistics is probably more basic that specific algorithms as FFT.
>
> In any case, there is a study group on Numerics (SG6). That's the right
> place where any proposal should go in first place.

Note that at present we are concentrating on number types and related
functions.  The intent is to get the common 'vocabulary' types in place
so that we can get the algorithmic libraries working together.

--
Lawrence Crowl

--




.