PLEASE NOTE: This article has been archived. It first appeared on ProRec.com in August 1999, contributed by then Editor-in-Chief Rip Rowan. We will not be making any updates to the article. Please visit the home page for our latest content. Thank you!
The time has come, Gentle Reader, for another rant.
I spend a little time out there on the popular Internet newsgroups, chat rooms, and discussion boards, surfing and listening. I like to lay low and read, and just absorb what people are thinking and saying.
Sometimes, the arrogant audio wannabes rear their ugly heads, and start spouting pure balderdash. The wannabes fall into two camps, who I refer to as the C12s and the 57s.
The C12s are the esoteric gear sluts. These are the guys that claim that everything digital sucks. Except maybe 24 bit recordings made at 192 KHz. Everything solid state sucks – all amplification must be provided by vacuum tubes. Pro Tools really sucks. The only good mics to use for vocals are vintage U67s and C12s, and actually, only certain runs of serial numbers within these products. All mixes should sum to mono. You can’t get a good recording unless the room you’re in has at least a 16 foot ceiling with at least one room dimension in excess of 32 feet. That only “perfectly flat” monitors in “perfectly flat” rooms can provide a good mix reference.
You know these guys.
Then there are the 57s. These are the reactionaries. They claim that all you really need is a 4-track cassette deck and a 57 to make a killer recording. That you can turn any old apartment into a recording studio with just a little foam padding on the walls. That you can’t tell the difference between MP3 and 24/96. That you don’t need compressors. These are the guys who, at least once a week, reinvent the idea of sticking a 57 up against a cranked Marshall. And of course, you can get a great mix on a jambox.
You know these guys.
The funny fact is that both groups are right, from their own frames of reference.
Trivial Pursuits: No Absolutes
The C12s, who have a frighteningly narrow definition of “good sound,” spend a tremendous amount of time analyzing the precious few pieces of gear and techniques that can capture and reproduce audio that meets their exacting criteria. Everything that falls outside the boundaries is summarily disregarded, and those who espouse views outside the boundaries are to be publicly shamed.
On the other hand, the 57s seem to accept any sound as good sound. Sometimes they simply have broad, accepting tastes – but, more often, they simply can’t discriminate between “good” and “bad” sounds. Usually they are very insecure when interacting with people who they suspect are a lot more experienced than they are, and they puff up with defensiveness and pride as they tell tall tales about their “killer mix” they made with a 4-track and a jambox at their friend’s house. The C12s become the “enemies” and a war ensues.
I’m really keen on debunking. The fact is that I fall into both camps simultaneously. I love esoterica – I grew up during what I consider to be the “golden age” of hi-fi recording – the 70s and 80s, when truly full-range recording and hi-power full spectrum reproduction were finally possible. I have spent quite a few hours listening to virgin vinyl records and well-mastered CDs played back on equipment by McIntosh, Threshold, Stasis, B&W, Klipsch, Tannoy, etc.. I remember when achieving the goal of almost-distortion-free recording and reproduction was still news.
Maybe you do too.
Along the way I got bored with what I ultimately found to be a trivial pursuit. I fell in love with older recordings, and heard miserable attempts to capture rock and roll sounds on “distortion-free” equipment. I learned that sometimes, lo-fi sounds better than hi-fi.
Case in point. I don’t usually pick on recordings, but I will make an exception. If I am picking on your favorite record, sue me. The CD is called “California Project” and it is a Telarc recording of the Beach Boys soundalikes Papa Doo Run Run. This is to my ears the perfect example of hi-fi gone amuck. An arguably flawless performance, made without compression, using only pristine signal paths and the finest recording gear available, this CD is as sterile as a prepubescent eunuch with radiation poisoning and three X chromosomes. It utterly fails to rock. A classic example of how not to record rock and roll music.
Once I started appreciating the finer points of distortion, I started questioning flat frequency response as a yardstick or desired quality in a recording. Then I started questioning everything else. I would pull out “Please Please Me” by the Beatles – a terribly lo-fi recording, that simply rocks all day long, and marvel at how sonically interesting it was, even though it was so lo-fi.
Where, I asked, should I draw the boundaries of taste? What, then, was “good” sound? I pulled out my classical recordings – in particular a 40-year old recording of Van Cliburn performing Rachmaninoff’s 2nd Piano Concerto. Lacking the deep bass response of my newer recordings, with a sheen of tape hiss covering up the high frequencies, this recording was still much more sonically interesting than my recent recordings which were, by any objective standards, better reproductions.
I have come to despise myths, prejudices and standards of “good” sound and recording techniques. From both camps – the C12s and the 57s. To the C12 camp, I say: if you can’t enjoy music on a jambox or a cheap car stereo, you have a problem of hyperactive discrimination. Loosen up, losers. And to the 57’s: if you have broad tastes that’s one thing, but if your recordings are muddy and distorted because you can’t make them sound differently, then you suck.
I decided to parse through old emails, discussion group comments, reader feedback, and other sources of error for fun and amusement. The result was this article. I hope it offends some people. They deserve it.
I got this new CD today and it was so compressed and limited my output meters just went to 10 and stayed there. How could they ruin music like that? a.k.a. Compression is Bad
Note this dork did not say anything about how the CD sounded! The fact is that some great recordings have almost no dynamics. It’s MUSIC, man, not Speaker Aerobics. If you’re doing in-your-face rock, one way to make that music get in your face is to compress the living snot out of everything. For music like that, the little “clipping” light on your compressor / limiter means “keep going, you’re almost there”.
Louder is better a.k.a. Compression is Good
This is the flip side of “Compression is bad” and is equally irritating. The advent of digital brickwall limiting has permitted some real abuses of technology – namely the “how loud can we make this” phenomenon.
I love compression – don’t get me wrong. I use it all the time. That’s not the problem. The problem is that in both the “compression is bad” and the “louder is better” camps the goal has become some kind of creepy exercise in dynamics (or lack thereof) and the real question is ignored.
What is the real question? It’s the ONLY question: “How does it sound?”
This studio’s monitors only went down to 35 Hz. And this guy was trying to convince me he could mix in there. You can’t do a mix on a system like that.
Huh? Good thing nobody told George Martin that. Turns out that a lot of people get really worked up about their monitoring rooms and systems. I will agree that the flatter your room and system response is, the more likely you are to get the mix right.
“Really, Sherlock, you figured that out all by yourself?”
However, room response is never flat. Ever. Ultimately, the only way to get a good mix is not to achieve the perfect room – in fact for most people, mixing in a perfectly flat anechoic room would be very difficult, because it would be so unusual. The only way to get a good mix is to know the room you’re in.
That’s why lots of people can get good mixes on NS10s. That’s why experienced engineers will keep their 20-year old, tired monitors instead of upgrading to the new, flatter, lower-distortion models. Not because the older speakers are the best technology, but because they’re the most familiar. Ultimately, mixing is about subjective frames of reference, not about objective standards of measurement.
For that reason I like to spend time in my control room – reading, working on nonmusical stuff, whatever – with a CD playing on the monitors. Monitors I’ve used since 1990. Subliminally I am developing and honing my frame of reference. If my room has a problem with bass buildup at 70 Hz, for example, then my brain will, over time, start anticipating and compensating. The room may not be flat – maybe not even close – but my brain will compensate, to a degree.
In an ideal world of free bandwidth it might be reasonable to push for frequency headroom of 2X (96 KHz). I mean, why not? It’s the same reason that just about any car can go 2 or 3 times faster than you’ll ever really need to go.
But bandwidth is NOT free. Most people working with digital audio are VERY bandwidth constrained. It’s always a tradeoff. Going from 44.1 KHz to 96 KHz means that you’re going to get less than ½ the track throughput – whether you’re tape or disk based. Unlike the move from 16 bit to 24 bit – which I thoroughly endorse – and which only increases bandwidth requirements on the disk (mixing and effects were already handled at 32 bits on all pro DAWs), the move to 96 KHz sampling more than doubles the load on every part of the signal chain. And where the move to 24 bit depth provided us with a whopping and noticeable 256X improvement in theoretical dynamic range, the move to 96 KHz provides us with only a 2X improvement in reproducing frequencies that only my dog can hear, and only if I have speakers that don’t exist.
Now, I can understand that by moving the Nyquist frequency out to 48 KHz, we allow for more gentle filtering, we are allowed looser, more forgiving design tolerances, we improve jitter, and so forth. The question is: how are YOU going to spend your money? Because you are going to immediately lose more than ½ your bandwidth. Is it worth it to lose over ½ your bandwidth for a miniscule – if perceptible – improvement in sound?
If you have a multimillion dollar operation, with a big budget to be able to afford serious 96 KHz bandwidth, then it’s like I said: if it’s virtually free, why not amaze your dog?
But the products that amaze me, baffle me, and just totally crack me up are the “semi-pro” 96 KHz products. You know the ones. They cost about $500 and offer 24/96 sampling.
What’s so funny? Here’s what’s funny. The sub-$1000 price point means that these products are targeted to people who are recording vocals with SM 58s onto little multimedia computers designed for web surfing and monitoring over home stereos. Nothing wrong with any of that, except the marketroids at the pro audio companies are going to try to convince you that if that’s your rig, your next investment needs to be 96 KHz converters.
Is anyone awake out there? What is that going to buy us? Are we going to get a more detailed reproduction of the noise in our Behringer mixers? Are we going to capture the subtle nuances of our air conditioning unit with this newfound frequency bandwidth?
The fact is that if you can’t afford the finest-quality 96 KHz converters, with more tracks and processing bandwidth than you’ll ever use, then you already have a lot more sonic bottlenecks in your recording process than your sampling rate, and a lot better places to spend your dough.
Are you even aware it’s coming? Because it is. The guys in the boardroom have done it again, and this time it’s the 192 KHz Audio DVD. The very concept makes me giggle.
What’s amazing is that before 96 KHz has even really hit the streets it’s already obsolete, because we now have a 192 KHz playback format. This is amazing to me.
What is inevitable is the well-documented – and quite real – placebo effect. It goes something like this.
Joe Consumer probably will never be able to actually explain what 192 KHz sampling rate means, but you can be for damn sure that when the guys are standing around the stereo drinking a tall cold one, Joe will wax eloquent on the improvement the new DVD player is over CD and proudly proclaim the superiority of 192 KHz, “‘specially in the imagin’ and the dynamic noise ratio. S’got somethin’ to do with them rectifires.”
Feeling suddenly inferior, all the guys will want one, and they’ll have to bullshit the wives into it. After having sunk a good chunk of change into the player and the discs, what fool would stand up in front of his friends and honestly admit, “Nope, it was a total waste of time and money. Can’t hear the difference.”
Once the consumer format catches on, then the musical community will quickly follow. “Do You Do 192?” will become the catchphrase when artists call around to local demo studios. Get 192, or perish. Fear, uncertainty, and doubt have always played a major role in audio trends, and hypersonic sampling rates is the latest craze.
I can see it now, and let me tell you, I’m cringing. We humans can really be a bunch of fucking lemmings.
6 gauge speaker cable
Why is the audio community forever willing to spend $600 for speaker cable to hook up a pair of $600 monitors? Why do we forever try to wring 0.0000001% distortion out of a signal chain that feeds speakers with inherent THD levels of 1-5%?
I actually have 6 gauge ultra-low impedance speaker cable on my monitors. Looks like garden hose. Local power company got jealous when they saw it. Made my wife’s unmarried girlfriend horny just looking at its masculine girth.
Found it in a neighbor’s trash pile. For free.
That’s the appropriate price.
Anything with tubes is better
OK, first off, I love tubes and tube gear. Always have, probably always will. We can argue the whys all day long, and I really don’t care why. Maybe I just like glass and glowing coils and the smell of warm transformer windings.
However, and this is a big one for me, the audio world is awash in dorky tube gear. These days manufacturers will stick an 12AX7 in anything – like maybe a power supply – and call it “tube gear.” We have products bearing invented words like “Solidtube” and “Tubessence” and “Tuberiffic” (ok, maybe not “Tuberiffic”, but it would be funny). Usually, clever little windows with mirrors have been employed to show off the Highly Exalted Tube and make the unit glow a sexy orange.
Meanwhile nobody gives the real tube gear a fair shake.
If you want to experience real tube equipment, I encourage you to spend the money for the stuff that uses classic designs with modern manufacturing. That gear is excellent and will stand the test of time.
The “other stuff” – that low-quality tube gear – can be useful, however. A few preamps and compressors come quickly to mind that use cheap components in an otherwise fairly normal designs. I won’t mention names, but you can buy these in the popular mail order catalogs. These preamps and compressors are cool for “browning up” sound – pump stuff through them, turn ’em up too hot, and toast your music to a dark brown.
Useful if that’s the sound you want. But NOT representative of the sound of high-quality tube gear.
Tubes are required
OK, now you lost me. Sure, I love tube gear. But it’s a tool, not a crutch. You’ll encounter people out there that will tell you that you can’t get “good sound” without tube gear.
Balderdash! Tubes and transistors are different tools. So you learn to use them differently. Tubes “do their thing” at moderate levels of distortion. A good piece of tube gear might have only 80 dB of dynamic range, but it might have an apparent headroom of well over 30 dB, due to beneficial harmonic distortion. So you learn to run this stuff in the red. Solid state gear does not create the same harmonic distortion – and usually should not be driven into distortion.
I know a guy who was complaining that his Mackie 1604 mixer was a dirty-sounding piece of shit. Turns out he was unfamiliar with the gain and summing structure of the mixer and was running it all way too hot. Well, yeah, run it like that and there’s no headroom.
You have to master the tool before you can truly judge its usefulness. A scalpel is a marvelous tool, but that doesn’t automatically make me a surgeon.
Analog is better a.k.a. Digital is shit
This is really a repeat of the “tubes are required” myth. How many times have I heard this one? I love analog tape running hot. I love the distortion and the soft-saturation. It’s awesome. No problem.
But if you can’t get a good sounding digital recording it’s probably because you haven’t mastered the tool.
Maybe you’re used to running RMS +4 on analog tape. Don’t do that with digital. A lot of people had to learn that the hard way. Maybe you’re used to the beautiful harmonic distortion of saturated tape. With digital you don’t get that. So you learn to use other gear – tube mics and preamps, amp and tape simulators, soft-limiting compressors, whatever.
I’m not going to tell you the reverse; i.e. digital is better. It isn’t. Just different.
You have to use mic X on source Y
This always takes the form of “you use a 57 on snare drum”, “you use a U87 on vocals”, “use a D112 on kick”, “use a 421 on toms”, “use an SM 81 on acoustic guitar.” These are fine starting points, but they are not prescriptions.
There are no mic prescriptions. Yes, some vocalists like recording with U87s, others with RE20s, others with Coles ribbon mics, and still others with 57s and 58s. A D112 is a good kick drum mic, but for a punchy sound for jazz or punk, try a 57, you’ll like it, and generally I prefer the less-bassy AT Pro25. I almost always reach for an AKG D1000E or AT 4050 over an SM 57 on lead guitar for rock music, it always seems to cut through the mix with more character and authority. And I recently fell head over heels in love with the Crown CM700 on everything from guitars to drum overheads and from viola to vocals.
OK, if you’re lost, trying desperately to get a snare drum sound, and you haven’t tried a 57, then you’re in need of help. Then the mic prescriptions come in handy. But don’t ever let anyone tell you that a given source requires a given mic.
It just ain’t so.