Stephen - I have a question which may be difficult to verbalize. When we mix a song over a set of speakers, we attempt to get the song sounding the best we can over those particular speakers. Then, if we hope to have that song translate over all speakers, one way of doing that is to mix the song over many sets of speakers and "tweaking here and tweaking there." So here's my question: shouldn't a mix that is flat, as evidenced by a flat EQ response on an EQ or frequency analyzer, ultimately be the best mix?
Speakers will change over the years as will speaker drivers, the materials they're made of, etc....which will give some more or less of a response in certain frequency ranges, etc..... It seems that once a song is sounding good over one set of speakers, then you play that mix over another set of speakers, the mix has to be tweaked again for those speakers but then the mix doesn't sound as good as it did on the first set of speakers. Then, if the mix is played on a third set of speakers, it has to be remixed again for those speakers but then parts of the mix are lost when replayed on the first two sets of speakers.
So where does it end? How do you know which set of speakers to mix for? Ultimately, wouldn't a mix with a flat frequency response be the best overall approach?
That's a great question, and I will tackle it in two parts.
First, let's address the question: is a "flat" mix the best? The short answer is a resounding NO!
I don't know of any professional who mixes by trying to create an even frequency response, and I think you would be very hard pressed to find any mixes that had a flat frequency response. If you did mix that way, by using an analyzer and trying to make all the frequencies equal in amplitude, I think the mix would sound horrible.
First off, our ears don't have a flat frequency response. They are more sensitive in some frequency ranges, especially in the critical mid-range where most of the speech/vocal frequencies reside. Our ears evolved to really zero in on human speech.
Second, the frequency response of our ears changes with the sound pressure level of the audio. The quieter the audio, the harder it is for us to hear low and high frequencies, which is the reason a lot of stereos used to have the "loudness" button that would boost bass and treble at lower volumes to try to compensate for the way we hear at different volumes.
You can do some searches and find graphs showing how much the frequency response of our ears vary by frequency and amplitude.
Getting back to the frequency response of a good mix, I almost never look at a spectrum analyzer while mixing any more, but back when I worked at the major studio, there was one built into the meter bridge of our big console. There was a switch that allowed you to turn the meters into a giant spectrum analyzer. I would look at it from time to time for fun, and what I observed back then that for typical rock/pop songs, the frequency response was very similar to a loudness curve... The shape kind of looked like two camel humps with one bump being in the low frequencies around 60 to 180 hz or so, and the other in the upper mid-range/low high-end frequencies, typically around 5Khz - 12Khz, with a nice roll off of the ultra high end and ultra low end, and a nice smooth valley in the mid range.
Of course, it's not that simple. You couldn't just program in that curve and get a perfect mix every time by trying to match it. Every song is different, with different types and amounts of instrumentation and/or vocals, so there is definitely not a one size fits all curve. Plus, instruments are usually coming in and out throughout a song, so it would also depend on which part of the song you are analyzing. Even if you average out the response for the entire song, there are still too many variables from one song to the next to come up with a curve that works, or even with a curve for each style/genre of music.
Some people have tried, and there are some automated mastering systems out there, as well as some automated mixing systems, but I haven't heard any of them that do a very good job at all yet. It's much more complex than simply looking at a frequency response curve.
That's also not even considering elements that make a mix unique. Even if there was a "perfect" frequency response for a mix, how boring would that be if all songs were mixed using the same frequency response curve!! Songs are about emotion and creativity, not conformity! Sometimes you want an element to really jump out in the mix, or be very aggressive or even harsh sounding, or you may want it to be very mellow and subdued. A great mix really helps to emphasize the mood and emotional message of the song, and there definitely isn't a one size fits all setting that will accomplish that!
Now that we got that out of the way, let's move on to the very important topic of mixing with multiple pairs of monitors.
Almost every studio I have ever worked at had at least two sets of monitors that you could easily switch between. In my own studio, I have 3 sets of monitors I can quickly switch between, as well as a sub-woofer that can be switched on or off with any of the 3 sets of monitors. The majority of my mixing work is done on only 2 sets of monitors, with some occasional checks on headphones, and usually a final check on my office computer speakers as well.
Most studios have one large set of monitors that are used for checking the really low end of a mix, as well as for cranking up for the clients. Then, they usually have one smaller set of monitors that you listen to at lower levels to focus on getting the all important mid-range right.
In my studio I have one relatively large pair of monitors that are my main monitors that I know very well, and give me a really good representation of the full frequency spectrum. I usually do most of my tweaking work on those, when I'm trying to dial in the sound of the tracks with EQ and other processing. Then I switch to my little Auratone speakers, which have no low end and no high end, and really focus on the level automation of the mix to dial in the volume balance of all the tracks. I also make sure that I can still hear everything on these very mid-range limited speakers. If I lose the bass when listening on the little speakers, then I know I need to work more on the bass sound so there is enough mid-range to still be heard on small speakers while not making it too loud on full range speakers. I also have a mid-sized set of speakers that I listen to once in a while, but not as much as they were originally part of a surround sound system and I can figure out everything I need from the other two sets of speakers. Since the kids now all seem to listen with earbuds or headphones these days, I always listen at least once through with headphones as well. You'll hear a lot more detail in headphones, so sometimes I'll catch some weird noise or something that I might have missed when listening on speakers, but mostly it's just an overall reference check to make sure the mix works on headphones as well.
The key is learning your monitors and being able to trust your monitoring environment!
The same monitors will sound different in different rooms, which is why there is no one set of perfect monitors. You need some acoustic treatment to get your room acoustics under control if you want to get to where you can really trust what you are hearing. Acoustic treatment is a topic for another day, though.
After you have your room treated and monitors position to be as accurate as possible, then it's just a matter of lots of listening and lots of practice! The more you listen and mix with your monitors, and then check your mixes on other monitors/systems, the better you will understand how things are supposed to sound on your monitors (in your room). In addition to working on your own mixes, take the time to listen to great mixes from other artists to get a feel for how they sound on your monitors. The more you can listen, the more you will learn how your monitors & room are supposed to sound, and the better your own mixes will become.
Also, no matter how good you get, you always need a reality check, especially if you've been working for awhile and your ears are getting fatigued. This is another time to listen to other mixes that you admire, even if just for a few seconds, so you can get a reality check on how your ears are doing. You may find that you keep adding more and more high end to everything as your ears get tired, so it's a good idea to check against a known mix once in a while to see if you've gone too far. Plus, you should try to take a quiet break at least once every hour or two to let your ears refresh a little bit. Check your emails or do a little exercise, or anything quiet you can do to give your ears a break for 10 or 15 minutes.
But, getting back to different speakers and different systems... When you are just starting out, you will definitely need to check on as many systems as possible, and do fine tuning of your mix until it sounds good on ALL the systems. You can't just make it sound great on one system and think you are done. You need to make your mix work and sound good on ALL systems. Sure, it might not sound great on every system, but it should at least sound good on every system. The more experienced you get and the more you learn how your main work system sounds, the less checking you will need to do on other systems. You will eventually just know how a great mix is supposed to sound or your system, although you will most likely want to double-check everything on another system that you are very familiar with (such as your office computer, car system, or wherever your listen to the most music when you are not working), as well as headphones.
Of course, this is all a lot easier if you have an accurate system (this includes the room and monitors) to work with, so that it actually sounds good, without being hyped, on your main work system. If your system has significant issues, it can be very difficult to get a good mix no matter how well you know your system. For example, if your system has a big dip in its low end response, you will constantly be fighting the urge to add more low end to your mixes, which would then make for a bass heavy mix on more accurate systems. So, getting your room and main monitors as accurate as possible will greatly help you create great sounding mixes. The goal is to have at least one accurate, full range, system that you can trust, and then one or two other sets of monitors you can easily switch between to help you focus on other elements of the mix, or give you another reality check.
No matter how many sets of monitors you may choose to use, the ultimate goal is to have an accurate system that you can trust and that you know well, so that you can greatly reduce the amount of time you need to spend checking your mix on other systems. You want to get to the point where you know when it sounds good on your studio system and can trust that it will then translate into a good sounding mix on any system.