New analyzer...

[Pink Noise]

I'm having problems with my ML1 and the people at NTI America seem to think it's normal behavior...

Well, I have to say that what they are saying sounds about correct to me too as far as the results go, but there are different things going on than has been suggested - possibly than the NTI rep realises, judging from what he's actually said. Unfortunately I can only tell you about the AL1, and it does rather more, and different, measurements. But as far as I can tell, it does them absolutely correctly, and I don't see why the ML1 would be any different from this POV. First, the actual definitions:

Pink Noise (aka 1/F noise) - the power spectral density is proportional to the reciprocal of the frequency - hence the aka. This means that there is equal energy in all octaves. The energy level falls with higher octaves at -3dB/octave. Because the number of Hz/octave doubles each time you go up one, there are more individual frequencies to disperse the energy in, so each displayed one has less in it.

White Noise has a flat power spectral density, so there is equal power in any band, at any centre frequency. Since there are more possible bands the higher we go, then the more energy there is in an octave's worth.[/list:u]
FFT's look at a large number of equal width bands across the spectrum, so an FFT of white noise gives a level display - which is what you would see in Audition. But if you use the Real Time Analysis (RTA) function, this displays in 1/3 octave bands of equal width (but with each one having more energy in it as they get higher), so the result is displayed differently. With the RTA, white noise accordingly slopes up at 3dB/octave. So, this has nothing to do with anything except the display mode, which is probably one you are not used to. And it really is as simple as that.

And it follows then, that the RTA of Pink noise will give you a level displayed result, and the FFT of Pink Noise will slope downwards.

This is exactly what the AL1 displays - I checked. As far as response times are concerned, this is also a function of what happens when you do real-time or FFT analysis - the lower the frequency, the longer it takes to compute, because the results come in more slowly! That's just the way it is, and there's nothing anybody can do about it. The display might be lagging, but the information is essentially correct when it gets there, I think you'll find.

This reaction delay is also the reason that a lot of digital filters are rather less good at LF than they are at HF, certainly in real-time.

Those are the definitions that I'm familiar with, but this technician seemed to think I was incorrect.

But if you use the Real Time Analysis (RTA) function, this displays in 1/3 octave bands of equal width (but with each one having more energy in it as they get higher), so the result is displayed differently.

This is why I suggested that I might see these results in a 1/1 octave analysis, but shouldn't I see some sort of slope with third octave readings? Also, I forgot to mention, when I use the microphone to analyze pink noise through my studio monitors or through a club PA system (that I set up, of course), it slopes as I would expect pink noise to. I'm fine with it being one way or the other, but give me both and I'm confused...

Those are the definitions that I'm familiar with, but this technician seemed to think I was incorrect.

Well this acoustician (who actually knows what he's talking about) is telling you that the above definitions are correct, and that if this is what you understand, then so are you! As I said before, this is simply a display issue.

This is why I suggested that I might see these results in a 1/1 octave analysis, but shouldn't I see some sort of slope with third octave readings?

You'd see the same slope, only at 3x the resolution. Noise doesn't know where the octaves are! All that happens is that the average value within each octave is plotted, and then it's three times further to the centre of the next octave than it would be if you were plotting 1/3 octaves, so the overall slope is the same, whether you use 1/1, 1/3, 1/6, 1/12 or 1/24th octave analysis. I can do all of those in software - but only the first two with the Acoustilyzer. The result is simply displayed with more and more resolution, but - the slope remains the same.

Also, I forgot to mention, when I use the microphone to analyze pink noise through my studio monitors or through a club PA system (that I set up, of course), it slopes as I would expect pink noise to. I'm fine with it being one way or the other, but give me both and I'm confused...

It's very simple - if you want to see the slope you are used to, use a FFT equal-step-size analysis. But if you want to see something that's more representative of what you hear, then use 1/3 octave analysis.

It's very simple - if you want to see the slope you are used to, use a FFT equal-step-size analysis. But if you want to see something that's more representative of what you hear, then use 1/3 octave analysis.

I don't really care which one it is, as long as it's consistant.  I still don't understand why the unit is displaying a slope with a microphone plugged in and showing flat with the line input.  Do you get these readings with the AL1?

I still don't understand why the unit is displaying a slope with a microphone plugged in and showing flat with the line input.  Do you get these readings with the AL1?

No - the AL1 is quite consistent. It behaves exactly as I described above, whatever input you use.

So than either my ML1 is malfunctioning, or my microphone does not have the response indicated in the certificate I received.

So than either my ML1 is malfunctioning, or my microphone does not have the response indicated in the certificate I received.

Without having both of them in front of me, I couldn't be absolutely sure about that - but it seems very unlikely that either piece of hardware is faulty. The mic's response would have to be massively in error, and the electronics in the ML1 is identical to the AL1 - it's the software that is completely different. So if there is a fault, it would most likely be a software one. But somehow, this just seems unlikely.

Are you saying that if you get a line-level signal fed into the line input, and when you attenuate it (do no more than this) and feed it into the mic input, that the spectral response is different, with no settings altered?