Peter, ... Right now, I am not confident I am using fuzzmeasure correctly. ... Each time I try to us it, I get a different looking result.
This is normal. The curve will never be exactly the same. You are not in an
anechoic chamber, your room is alive, it has reflections. For now, without any acoustic treatment, the curves show you how wildly the sound bounces across your room. Very long impulse responses combined with high values in the lower frequencies show the existence of standing waves. Don't know if you see this with your current monitors. It is a good idea to measure now, then with the Yamahas and then with the Yamahas and acoustic treatment. Just for learning.
Btw, in Fuzzmeasure you can shift-click and cmd-click on more than one measurements (left column) to combine the curves and see the differences and imagine average values. So it is good to name and save the measurements for later comparison. And to find measurement failures which you or the software may have produced.
A couple of hints:
Use a mic stand, place the mic in your listening position. Additionally you can measure in different locations to get a clue about other spots in your room. The recording position is interesting for example because your recording mic will only hear this spot. For measuring set the sweep to start a couple of seconds after you hit the button and get out of the way. Don't hide in a corner, your body would work as an absober. Best is to be out of the room while the sweep runs. And use 1/6 or even 1/3 octave smoothing in Fuzzmeasure. Finer resolutions are just confusing, you cannot handle small frequency bands yourself anyway.
Look at the frequency response of my mixing room:
This doesn't look good, although it is almost within the tolerance I find acceptable for my work. I temporarily removed some bass traps because I need them in the recording room but the curve doesn't look much better with them in place. I would need broadband absorbers to even out the mid range.
Now comes the interesting part, the added value of such curves for recording, arranging and mixing:
- I don't care a lot about too much 100 Hz from a guitar, the room delivers more than I actually have.
- I can boost the female voice frequencies of 300-400 Hz because my room response goes down from 150 Hz on.
- Around 700 Hz is a well known and rather unwanted range for an acoustic guitar, it gives this "honky" tone. I have a deep vally there, the room is too polite. I have to drop the EQ a little even if it sounds right.
- I hear always too much at 1 kHz than there actually is; another lie of the room.
- I am careful with boosting 3 kHz for female vocals because the room tries to hide them. I may boost to much and force this annoying 3 kHz "trumpet" some women produce when they go loud.
- I can trust the range above 4 kHz. This is important to me because I am 58 years old and did never hear much in the high frequencies.
See? A lot of information. But don't worry, by just looking at the curve, preferably at 1/3 octave resolution, we can take our technician hat off. It boils down to simple rules:
- Peter, do not drop the bass and the 1k range!
- Peter, watch for the ranges below and above 1 kHz!
- Be relaxed and have fun.
I have an EQ in the Logic output channel. It is made according to a 1/3 octave resolution of Fuzzmeasure and does not follow every detail. It's just a little helper, makes life easier because of less brain-work while validating sounds.
Next, reveration time - a very important analysis from Fuzzmeasure:
(The red cross means that this value is an illusion of Fuzzmeasure. I did not even measure there, started at 40 Hz because my monitors don't go that far down.)
Here you can see what your absorbers do, this is valuable information. 300 ms means that a room is very dry. 400 ms is a comfortable reverberation time, 500 ms is a bit long. From that, my room is not bad. I can work for a long time and don't get tired. But I don't have too much reverb, which would smear the whole sound.
Please note:
The reverberation time is important because it tells about the time domain. You cannot change this by an EQ and this is part of the problem. Many people try to equalize their room or turn the lows far too much down. But this doesn't change the timing of the sound. The room modes are a physical fact and if a sound wave remains for a second, turning the lows far down just ruins the overall sound. This is the world of bass traps, low frequency absorbers. They swallow the energy and reduce reverberation.
it's not clear whether I have to do certain baseline tests (e.g. loopback, impedance, etc) before I can get meaningful and consistent results.
Nothing necessary. Just get a good mic signal, don't overload your preamp or the curve will be wrong. No problem if it goes just up to -15 or -10 dB. For easier exploration you can shift the curve in Fuzzmeasure until a thick line goes where your average values are. This is not an absolute measurement, you just want to learn about the difference in frequencies and reverberation time.
---
Again a long post, I hope you don't mind. But it seems that we get a useful little "manual" for many people working on their recording- and mixing environment.