Good FFT routines for sound card?

[Joerg]

Hello Folks,

What would you recommend as a good FFT program for use with laptop
internal sound cards?

Ideally it would come from a reliable source such as a university,
meaning not too many bugs and no virus. It should have a somewhat
practical user interface. I am not a programmer who could easily compile
something using a collection of routines such as the FFTW files from
MIT. I do have some programs that came with engineering books I bought
but these are DOS and can't handle sound card access easily anymore.
They are from the days when you piped external ADC stuff into the PC via
the parallel port. Free would be nice, of course ;-)

So far I found this one from Rutgers:
http://duncan.rutgers.edu/physicsfreewares.htm

Regards, Joerg

http://www.analogconsultants.com

 

[martin griffith]

On Wed, 27 Jul 2005 19:24:01 GMT, in sci.electronics.design Joerg
<notthisjoergsch@removethispacbell.net> wrote:

Hello Folks,

What would you recommend as a good FFT program for use with laptop
internal sound cards?

Ideally it would come from a reliable source such as a university,
meaning not too many bugs and no virus. It should have a somewhat
practical user interface. I am not a programmer who could easily compile
something using a collection of routines such as the FFTW files from
MIT. I do have some programs that came with engineering books I bought
but these are DOS and can't handle sound card access easily anymore.
They are from the days when you piped external ADC stuff into the PC via
the parallel port. Free would be nice, of course ;-)

So far I found this one from Rutgers:
http://duncan.rutgers.edu/physicsfreewares.htm

Regards, Joerg

http://www.analogconsultants.com
I think Daqarta is still around, very DOSsy, but have a word with Bob

Masta at daqarta.


martin

 

[Joerg]

Hello Martin,

I think Daqarta is still around, very DOSsy, but have a word with Bob
Masta at daqarta.

It won't run under Win2000 or NT and doesn't support laptop sound cards.
Other than that it looks like a great SW, seems worth the registration
fee. Have to keep my eyes peeled for the new Windows version.

Regards, Joerg

http://www.analogconsultants.com

 

[martin griffith]

On Wed, 27 Jul 2005 20:12:31 GMT, in sci.electronics.design Joerg
<notthisjoergsch@removethispacbell.net> wrote:

Hello Martin,

I think Daqarta is still around, very DOSsy, but have a word with Bob
Masta at daqarta.

It won't run under Win2000 or NT and doesn't support laptop sound cards.
Other than that it looks like a great SW, seems worth the registration
fee. Have to keep my eyes peeled for the new Windows version.

Regards, Joerg

http://www.analogconsultants.com
Hiya Joerg,

Havent been to Daqarta site in some time, he was promising windows gui
a Long Time Ago......
maybe Sound Technology, basically Darqarta with a fancy GUI. You buy
the modules you want. Might have a url somewhere,

shuffle ancient hard drives into action
hi de ho, dum dee dum, where is Slartibardfast when you need him?

nope,

sorry


martin

 

[martin griffith]

On Thu, 28 Jul 2005 00:14:16 +0200, in sci.electronics.design martin
griffith <martingriffith@XXyahoo.co.uk> wrote:

On Wed, 27 Jul 2005 20:12:31 GMT, in sci.electronics.design Joerg
notthisjoergsch@removethispacbell.net> wrote:

Hello Martin,

I think Daqarta is still around, very DOSsy, but have a word with Bob
Masta at daqarta.

It won't run under Win2000 or NT and doesn't support laptop sound cards.
Other than that it looks like a great SW, seems worth the registration
fee. Have to keep my eyes peeled for the new Windows version.

Regards, Joerg

http://www.analogconsultants.com
Hiya Joerg,
Havent been to Daqarta site in some time, he was promising windows gui
a Long Time Ago......
maybe Sound Technology, basically Darqarta with a fancy GUI. You buy
the modules you want. Might have a url somewhere,

shuffle ancient hard drives into action
hi de ho, dum dee dum, where is Slartibardfast when you need him?

nope,

sorry


martin
http://www.spectraplus.com/

martin

 

[Joerg]

Hello Martin,

http://www.spectraplus.com/

Interesting. The only weird thing is that there is an order sheet but
they do not seem to disclose any pricing. Really strange.

Regards, Joerg

http://www.analogconsultants.com

 

Find a copy of gram.exe

Steve Roberts

 

[YD]

On Wed, 27 Jul 2005 19:24:01 GMT, Joerg
<notthisjoergsch@removethispacbell.net> wrote:

Hello Folks,

What would you recommend as a good FFT program for use with laptop
internal sound cards?

Ideally it would come from a reliable source such as a university,
meaning not too many bugs and no virus. It should have a somewhat
practical user interface. I am not a programmer who could easily compile
something using a collection of routines such as the FFTW files from
MIT. I do have some programs that came with engineering books I bought
but these are DOS and can't handle sound card access easily anymore.
They are from the days when you piped external ADC stuff into the PC via
the parallel port. Free would be nice, of course ;-)

So far I found this one from Rutgers:
http://duncan.rutgers.edu/physicsfreewares.htm

Regards, Joerg

http://www.analogconsultants.com

If what you want is a readymade program I recommend the freeware gram
5.01. The author's site seems to be down but do a search for
gram50.zip and you'll find several places for download.

- YD.

--
Remove HAT if replying by mail.

 

[Joerg]

Hello Steve,

Thanks, I'll try it. I heard this program name before in the ham radio
community but thought it was an offline tool. From what I found via
Google it seems to also run realtime.

Regards, Joerg

http://www.analogconsultants.com

 

[Joerg]

Hello YD,

If what you want is a readymade program I recommend the freeware gram
5.01. The author's site seems to be down but do a search for
gram50.zip and you'll find several places for download.

On advice from Steve I found one zip file at a German site. Don't know
its version yet but hopefully will have time to try it on the weekend.

Long term I want to do more, like testing filter routines. But then I'd
have to migrate from the leather seat back to individual files and C
routines (which I have).

Regards, Joerg

http://www.analogconsultants.com

 

[FChoquette]

Hi,

If it is only for testing purpose, not to compile in an executable,
Matlab may help you just fine.

AUDIORECORDER(Fs, NBITS, NCHANS)

will record on-the-fly, using any sample rate supported by your card.
By default, it will use Windows default input device but you can
override this. Resulting data can be in many bit depth (8, 16, 24) and
mono or stereo. Data is returned as an integer array.

You can then FFT the recorded array and do anything you want with it.

Last time I used this function, I made an automatic piano tuner with it
Have fun!


Francois Choquette


Joerg wrote:

Hello Folks,

What would you recommend as a good FFT program for use with laptop
internal sound cards?

Ideally it would come from a reliable source such as a university,
meaning not too many bugs and no virus. It should have a somewhat
practical user interface. I am not a programmer who could easily compile
something using a collection of routines such as the FFTW files from
MIT. I do have some programs that came with engineering books I bought
but these are DOS and can't handle sound card access easily anymore.
They are from the days when you piped external ADC stuff into the PC via
the parallel port. Free would be nice, of course ;-)

So far I found this one from Rutgers:
http://duncan.rutgers.edu/physicsfreewares.htm

Regards, Joerg

http://www.analogconsultants.com

 

[Joerg]

Hello Francois,

If it is only for testing purpose, not to compile in an executable,
Matlab may help you just fine.

AUDIORECORDER(Fs, NBITS, NCHANS)

will record on-the-fly, using any sample rate supported by your card.
By default, it will use Windows default input device but you can
override this. Resulting data can be in many bit depth (8, 16, 24) and
mono or stereo. Data is returned as an integer array.

You can then FFT the recorded array and do anything you want with it.

That is a great way to do it. However, except for academic use Matlab is
expensive IIRC and I believe they want yearly fees.

Last time I used this function, I made an automatic piano tuner with it
Have fun!

I did a kludge to tune our piano. Wrote a small calc routine in a spread
sheet that also allowed me to add the small correction factor for
tempering. This spit out a frequency table for all strings. I could have
somehow fed that into the generator via GPIB but just printed it out and
dialed it in while at the piano.

The problem is that with our temperature extremes this old piano doesn't
hold tune for too long. But it does sound like one of those good old
saloon pianos.

Regards, Joerg

http://www.analogconsultants.com

 

[Boris Mohar]

On Wed, 27 Jul 2005 19:24:01 GMT, Joerg
<notthisjoergsch@removethispacbell.net> wrote:

Hello Folks,

What would you recommend as a good FFT program for use with laptop
internal sound cards?

So far I found this one from Rutgers:
http://duncan.rutgers.edu/physicsfreewares.htm

Regards, Joerg

http://www.analogconsultants.com

Try these free

http://www.arachnoid.com/
http://www.qsl.net/dl4yhf/spectra1.html



Regards,

Boris Mohar

Got Knock? - see:
Viatrack Printed Circuit Designs (among other things) http://www.viatrack.ca

 

[Joerg]

Hello Boris,

http://www.arachnoid.com/

This one appears to be Linux only.

http://www.qsl.net/dl4yhf/spectra1.html

That looks very good. Thank you. I will try it out. From reading that
site I happened upon another FFT program here:
http://www.weaksignals.com/

Regards, Joerg

http://www.analogconsultants.com

 

[Jamie]

Joerg wrote:

Hello Boris,

http://www.arachnoid.com/


This one appears to be Linux only.

http://www.qsl.net/dl4yhf/spectra1.html


That looks very good. Thank you. I will try it out. From reading that
site I happened upon another FFT program here:
http://www.weaksignals.com/

Regards, Joerg

http://www.analogconsultants.com
i have Pascal code for FFT if you wish to look at it.

--
Real Programmers Do things like this.
http://webpages.charter.net/jamie_5

 

[Joerg]

Hello Jamie,

i have Pascal code for FFT if you wish to look at it.

Thank you but I can't read Pascal. A wee bit of C and assembler but I am
not really a SW specialist. However, at some point I'll have to dive in
to try something out. Then the problems are mostly not so much the C
routines where I have a ton of "canned" versions that came with books.
The real problems is talking to the sound card and all kinds of Windows
interfaces. Maybe I'll do that in DOS ;-)

One issue is the miserably shallow documentation of "modern" computers.
On mine they only tell you that the sound comes in via an Analog Devices
chip but not which one. That pretty much "sums up" the HW docs for
sound. It is pathetic.

Regards, Joerg

http://www.analogconsultants.com

 

[Jamie]

Joerg wrote:

Hello Jamie,

i have Pascal code for FFT if you wish to look at it.


Thank you but I can't read Pascal. A wee bit of C and assembler but I am
not really a SW specialist. However, at some point I'll have to dive in
to try something out. Then the problems are mostly not so much the C
routines where I have a ton of "canned" versions that came with books.
The real problems is talking to the sound card and all kinds of Windows
interfaces. Maybe I'll do that in DOS ;-)

One issue is the miserably shallow documentation of "modern" computers.
On mine they only tell you that the sound comes in via an Analog Devices
chip but not which one. That pretty much "sums up" the HW docs for
sound. It is pathetic.

Regards, Joerg

http://www.analogconsultants.com
i have a good suggestion for you.

find your self a copy of Delphi.
there are lots of code out there for putting the
sound card in record mode to capture the PCM bits.
i can even supply code for you so that you can
adapt to windows coding.
my URL below will show an app that uses the sound
card to decode and encode Slow Scan Tv. it using
FFT coding as an option but defaults to zero crossing
on initial install.
i find for rather clean tones, i get better
resolution on detect with out chewing up CPU time.

FFT has a little problem with precision that can
be helped out using larger Float numbers at the cost
of CPU time.


--
Real Programmers Do things like this.
http://webpages.charter.net/jamie_5

 

[Joerg]

Hello Jamie,

i have a good suggestion for you.
find your self a copy of Delphi.
there are lots of code out there for putting the
sound card in record mode to capture the PCM bits.

Thanks, but I don't want to become an expert in code. Just trying to FFT
something. So far I found that the Rutgers program is pretty bare bones
but performs the best (fastest).

Later I may have to try to do FFT in an MSP430. I am not looking forward
to that as it will be like cramming 15 people into a VW Beetle (has been
done...). Also, it's not going to be the MSP430 on steroids, no HW
multiplier here :-(

Regards, Joerg

http://www.analogconsultants.com