It's whatever your sampling rate is. If you resample to 96 then it will go to 48khz. The lowest frequency, I believe, is determined by the height of the window, so it is not likely to go below 10-20hz.
Resample? height of the window? I don't understand.. the frequency range goes from left to right and I expect the "level" of each frequency it displays is calculated either against some kind of maximum or an arbitrary value, but I mean I'm interested to know what the very left-most column represents (guessing 20Hz) and the right most (guessing 20kHz), but I was after definite numbers.
I don't know what resampling has to do with it, as I thought that just increases the accuracy of the reproduction of the music, very little to no impact on the volume(s) of each frequency... (other than the fact that if u had a "music" file with NO samples, then technically you'd get no volume at all
Originally Posted by Publius It's whatever your sampling rate is. If you resample to 96 then it will go to 48khz. The lowest frequency, I believe, is determined by the height of the window, so it is not likely to go below 10-20hz.
/me goes off and checks.
Hahah, you're right. That's kinda lazy. My guess is that they just take the output after the DSPs but do not keep track to scale the frequency range if it resamples.
If you resample to 96Khz, then the frequency range that can be reproduced increases to 48Khz. Even though resampling will not introduce any frequencies above the original bandwidth (well, if you use a good interpolator), they are just probably doing the FFT on the output from the DSPs. So when you play back at your regular CD sampling rate, the highest is 22KHz but if you resample, then it's going to be one-half the new sampling rate.
As for the low end, they probably go down to DC. The difference along the frequency axis of the graph between 0 and 20 Hz (which is the lowest that redbook will go) is going to be very small so whether it goes down into the tens of Hertz or all the way to DC level is probably not going to be noticed by the viewer.
You're right Born2bwire.. didn't think of that (viewer not being able to tell the difference anyway), as long as it goes to 20Hzish and isn't just showing down to the minimum adjustable band on the equalizer plugin (my original suspicion), I don't mind.
One thing this thread had confused me about though, is resampling. The number of samples means the number of elements per time unit used to store (or in this case recreate) an audio signal right? So I would expect resampling to mean taking the original signal, and say u were gonna resample from, asy, 16bit to 32bit, you'd be taking a bunch of extra samples by using the avg between every point u were originally given, supposedly creating a smoother signal and theoretically a better final output. Now.. to me it sounded like you guys were saying this resamlping was having some kind of effect on the EQ??
I don't get is why resampling would effect the range of frequencies shown on a display (or reproduced by the soundcard, for that matter).
Cos isn't the number of samples to do with accuracy recreating a set predetermined frequency spectrum (WRT audio anyway)?
In ideal resampling, yes, you would change the frequency of the samples from say 44.1KHz to 96KHz and at the same time you would not introduce any new frequency information. However, the bandwith of a 96Khz sampled signal is 0-48KHz. So some interpolators will introduce higher frequencies than in the original depending on how they are going about it, but this is beside the point. The point is that the new frequency range has enlarged and when Foobar does not consider the original signal's bandwidth when it samples for the visualization. So when they take the frequency content using FFT after the DSP's, they see a bandwidth of 48KHz and not the original 22.05KHz.
No wonder the bottom half of my EQ looks kinda normal and then the treble half has heaps of low level output, makes it look like everything has lots of bass. Though I don't have resampling turned on so I dunno why it's skewed that way. hmmmm