PDA

View Full Version : 24/96 Audacity


irishcrazy2005
2005-08-09, 05:50 PM
Oops!

Five
2005-08-09, 06:29 PM
this isn't the sort of info you will find on this board.

you could try http://audacity.sourceforge.net/ you just have to set a few things to make it display all the way up to 22500kHz in the spectral display (in the options).

You could also seek out an old shareware version of CEP2.0 there's a link posted in the first thread in the lossy or lossless forum here. when that prog expires you are still allowed to use the spectral view and frequency analysis altho pretty much everything else won't work after it expires, unlike Adobe Audition where nothing works after it expires.

irishcrazy2005
2005-08-09, 07:08 PM
this isn't the sort of info you will find on this board.

you could try http://audacity.sourceforge.net/ you just have to set a few things to make it display all the way up to 22500kHz in the spectral display (in the options).

You could also seek out an old shareware version of CEP2.0 there's a link posted in the first thread in the lossy or lossless forum here. when that prog expires you are still allowed to use the spectral view and frequency analysis altho pretty much everything else won't work after it expires, unlike Adobe Audition where nothing works after it expires.

Thanks for the info. I want to be clear that I wasn't trying to do anything illegal, so sorry if what I asked was!

-Phil

irishcrazy2005
2005-08-09, 07:29 PM
Ok, so now I am using Audacity and I have a question. I have two versions of the same show. Same source, one is just the original 24/96 seed and the other is the 16/44.1 seed that the taper also released. When opening these in Audacity and zooming into a 2-sec section, the waveforms look exactly the same. Why is this, and where should I be seeing differences?

-Phil

irishcrazy2005
2005-08-09, 07:38 PM
Ahha! Now I've figured it out. If you look at the FFT analysis, the 24-bit sources gets lower and higher in frequencies! I should just retitle this thread, "Thread where I talk to myself."

-Phil

irishcrazy2005
2005-08-10, 02:59 PM
Ok, now i am confused about some things. I understand the Nyquist theorem saying that the sampling frequency must be twice the bandwidth of the source, but I am confused by this spectrum. What are these higher frequencies in this FFT of a 24-bit, 96kHz source? Are these really acoustic frequencies being picked up by the mics, or are they just higher harmonics or something? I am confused. I didn't think that mics could pick up frequencies this high. What is the sharp cut-off at about 24kHz? Also, if I lower the FFT size, the higher frequencies all but dissappear. Help! I am so confused.....

-Phil

http://img343.imageshack.us/img343/7343/24960nx.jpg (http://imageshack.us)

prof_peabody
2005-08-10, 04:14 PM
Without much info, and my audio physics fading rapidly, it looks like the source had a hard (rectangle) high-pass filter (e.g. not tappered). This kind of filtering is great fr producing all kinds of aliasing. It looks like the energy peak at 12 kHz is generating a nice harmonic at 24 and 36 kHz.

On a side note, i wonder about the usefulness of recording 96 kHz with some mics. Seems overkill to me; 24 bit I can understand.

prof_peabody
2005-08-10, 04:20 PM
Ok, things are starting to move in my head. It looks like someone wasted a lot of bits recording at 96 kHz. Somewhere in the set-up something is filtering the signal just above 22 kHz. The rectangle filter is introducing a fair amount of noise just below 22 kHz. If you wante to clean up the recording, I'd probably run a tapered filter on the high end to filter out that crap.