I see topics such as "what's the best monitor in such price range" "active or passive" "sub or no sub" recurring all the time, so I thought I'd summarise a few of my thoughts on such subjects for those of you who might benefit from them. And as you can see from the size of the text below, there ain't an easy answer!
- 1. The myths of Frequency Response
It is very difficult (if not impossible) to determine the audio quality of a speaker based on specs like 100w and 34Hz-22.000Hz... in fact, it is even difficult to determine the power handling and freq. response of said speaker based on this information alone. Let me elaborate: a speaker that is quoted to have a freq. response starting at 34Hz means in no way it will sound deeper (or better) than one quoted to start it's response at 45Hz because crucial information like dB roll-off, energy dispersion and A LOT of other crucial info is often omitted. 34Hz -6dB at... means I will hear nothing of it, but 45Hz ±0dB will mean that part of the sound is intact. Most commonly you will see a ±3dB (which is considered good) next to your tasty 22Hz-37KHz freq. response, which begs the next question: where in the wide spectrum do i get +3dB and where do I get -3dB? Again, this detail is often left out (remember that 3dB means a doubling or halving of the energy in a frequency, and depending on where this occurs it could be a significant detail). Other data like SPL, dispersion and the angle of coverage in degrees (both vertical and horizontal) can be significantly more revealing than a relatively abstract 100w, which in itself has almost zero relevance to audio (and even perceived power).
The easiest and most accurate way of reading specs is via graphs and diagrams and ideally every manufacturer should include them in their info sheets, but truth is, most people can't read graphs and what's more, they don't care. All they want to know is: Do they sound good, and can I blow my windows out with them... which brings me to point 2
2. Active or Passive?
Ok, so you now know how to read specs and graphs properly and can see that choosing a decent set of monitors has suddenly become a much narrower task. You should also notice that low cost active monitors have almost disappeared completely from your list of candidates. Why is that? Because to manufacture good quality amps properly balanced to the cones/horns they are driving, with good quality crossovers, cabinets etc. is an expensive business. "So should I buy passive ones with a decent amp?" do I hear you ask? In my view: yes, if you're a bit strapped for cash you can go much further with a decent set of passive monitors or even hi-fi speakers coupled with a solid quality amp than any low-end active monitors, even if doing so complicates matters slightly: for one you want to choose an amp that's not gonna blow your speakers as soon as your volume knob leaves 0, so put a brake on your boy-racer background and choose wisely (but now you can read specs so that's not really an issue, right?).
This in not to say active speakers are no good. On the contrary. There are many benefits for using active monitors, the most obvious ones being that amps and cones will be properly balanced thus minimising the risk of distortion through overdrive or blowing the drivers. Also most good quality active monitors will be magnetically shielded and take balanced feeds, greatly reducing or completely eliminating interferences and ground loop noises... but like I said, be prepared to fork out a hefty sum for a good pair.
At the end of the day, ALL loudspeakers in both categories (hi-fi and studio) are designed with the intention of having the most extended and flattest response that their production budget will allow, so don't let your bias (and marketing propaganda) influence your decision without conducting your own research! Furthermore, amplifiers are a lot less subject to factors influencing freq. and almost all good quality amps have a near perfectly flat response, so in my opinion it's worth devoting critical budget to speakers instead, where the quality can vary to much greater degrees.
"Great. got it. Let's roll!"... easy tiger, I'm not done yet...
3. An overlooked process: Tuning
The best quality gear can only go so far if not set up properly and vice-versa, your dad's modest little 70's hi-fi can go a long way (ok maybe not, but you know what I'm saying) if you spend a little time tuning it properly, and doing so can be a daunting task if left to uncle Useyourears.
You've read the specs, made your move, unpacked your jewels bursting every cell in the bubble wrap, fire them up but... shock horror!... your ears are telling you there's something wrong, there's this cold-hearted stab that smacks you in the face when your congas come in, but you can't quite put your finger on it. What you are probably experiencing are a mixture of constructive and destructive phase relationships caused by reflections from your walls which cause certain frequencies to stand out more and others to cancel out, mixed with phase alignment issues due to the speaker's relation to each other! They're not as obvious as you cranking up the Q on a resonant filter, but very noticeable nonetheless. Try this: set your output to mono, unplug one of the speakers, get close to the other and stare it right in the cones! Does it sound sweet? Good! it means there's nothing wrong with your choice of speakers thank fuck for that!... So, a little more work is needed and for that you're gonna need your reading glasses and your tool bag.
The advantage of listening to your audio in a small home studio as opposed to a live gig, or even in your living room, is that you only really need good sound in one spot: the chair you're sitting in. This reduces a lot of problems caused by cancellations but note I haven't used the word 'eliminates'.
If you really want to get a good picture (literally) of what your frequencies are doing, the best approach is to use an omnidirectional mic (also known as measurement mic) and hook it up to a decent quality Analyser (there's a few good plugins as well), preferably one that can take 2 inputs: one from the source (your DAW) and one from the mic. This will also show your delay... but I'll get to that in a minute... in the meantime, place your mic where your head would normally be, fire up some pink noise* and crank the volume up (White noise is no good, I'll explain why in another article maybe).
What you should notice immediately by looking at your analyser are some nasty spikes (downward) more prominent at lower frequencies and repeating at their relative harmonics. These represent your room's acoustics and their relative resonant frequencies... and cause a effect commonly known as 'comb filtering' Often you will get a whole array of unrelated spikes resulting from all kinds of frequencies bouncing all over the place, especially at higher volumes, quickly giving you the impression that your speakers sound like crap! (Mind you're listening to pink noise at this point, not exactly a Mozart piano-concerto-like experience anyway
but back to our pretty spectrum graph picture! An all too common mistake at this point would be to even out those spikes using EQ to compensate. Very bad!... a) you can't really boost frequencies which have cancelled out, b) you won't achieve a better sound... so I presume these are not options. Don't worry, there's alternatives. Try repositioning your speakers so they're not directly facing flat surfaces, add a few obstacles along their way and fill the room with as many rugs, curtains and soft toys as you can and you will quickly get rid of a lot of the harshness of those spikes. This will also dramatically help with room reverberation.
Ok. We've talked about frequencies. What else is there? Well, there's phase relationship problems (constructive and destructive) which results from the left and right speakers intersecting each other incorrectly. This is a much more complex topic, but to keep things simple, much can be improved by placing the speakers perfectly parallel to each other and at the exact same distance from you, and while this can be a significant aspect on large systems, it's a relatively minor issue on small ones given that due to the short distances between speakers (and lower volumes) only higher frequencies are worst affected... and they tend to create less of a problem (and are less noticeable), so not worth fussing over them at length. But still, worth mentioning.
If after all this the results are still not satisfactory, as a last measure before calling the builders to treat your room, try this while looking at your analyser: TURN THE VOLUME DOWN! See? Magic. Your spikes are disappearing faster than myspace accounts... Yes yes, I hear ya, "you can't produce and mix at low volumes, you GOTTA hear everything!" Truth is: you CAN hear everything at low volumes (an SPL of around 80db is considered to be where the human ear is most responsive, which is about as loud as conversation speech) and the advantage is two-fold. Train your ear to mix at low volumes and not only will you hear everything, but also get a much clearer picture of the dynamics in your tunes and you will not suffer from ear fatigue after 15 minutes. A HUGE factor when you are trying to retain objectivity over your sounds (and fidelity of your speakers, to stay in topic) but we're entering into Psychoacoustic territory here, so I'll leave that for another time... but fear not, I'm not leaving you hanging just like that. I've saved the best for last.
4. To sub or not to sub:
If your productions contain low frequencies, logically it would make sense to be able to hear those! No? Yes!
Once again there seem to be a lot of debates about the benefits vs disadvantages of using subs in your monitoring and few people quoted places like Abbey Road not using subs in their monitoring. This is plain fantasy!
One of the reasons many top producers choose not to use subs in their daily mixing is because low frequencies cause ear fatigue (and headache) faster then open/hi frequencies and chances are that on a long day session, you will mostly work with sounds and instruments that don't focus primarily on the low end content. But you can be sure that when they work on kick and bass (or playback the mix as a whole) the sub will be ON!
And here it's fair to point out that all the rules we talked about previously about setting up an accurate, flat response over the whole spectrum becomes that tad bit more difficult, but the more necessary should you choose to implement a sub in your rig. And why shouldn't you. Right?
Bear in mind that low frequencies are always the most problematic (and most noticeable) when tuning a rig, so be prepared for a bumpy ride when tuning your sub following what we've previously talked about, and this is the point where most anti-sub campaigners give up the fight. That's primarily because low frequencies are a lot harder to make directional (can be done, ask me nicely I'll tell you how) and will be bouncing all over the place and barging through obstacles like a raging bull trapped in a phone box. So what now?!
Here's what: Delay (as in time delay, not an audio effect). That's where your 2 way input analyser will come in handy: Turn off the speakers leaving the sub and play your favourite pink noise tune again and read the delay (milliseconds) between your direct out and the mic reading. Then turn off your sub and do the same with the speakers. Compensate the delay by applying the correct time difference to your speakers (not the sub). Great. Now the sounds from the sub and speakers sit perfectly on top of each other and reach you at the same time. (Mind you, you could have achieved the exact same result by physically moving the sub to line up with speakers, but you want to know how the pros do it, right?!) Continue by adjusting the volume of the sub so it sits along the flat spectrum nicely (swallow your pride, put a brake on your temptation and TURN IT DOWN! I seriously can't stretch this far enough) and lastly measure the crossover frequencies and apply low-pass to the sub and high-pass to the speakers until you finally have your NICE, HARD EARNED, FLAT RESPONSE.
I hope this helps to dispel some myths and clear up some confusion about the importance of choosing the right monitors.
Ultimately it's the music that really counts and as a dear friend once said: "You cannae polish shite"... he is scottish
* You can get pink noise by applying a pinking filter (3dB per octave roll-off across the whole frequency range) to white noise
GAFM ***
