There was all sorts of buzz earlier this week about TV on the Radio's apparently abysmal performance on Saturday Night Live. If you like the band, as I happen to, you have to wonder what happened: their recordings are immaculate, and they have a solid live reputation.
Today, music blog Idolator performs an interesting experiment, embedding the band's performance of "Dancing Choose" on SNL directly above the same song performed on last night's edition of The Colbert Report. The difference is immediately noticeable.
So what the heck happened on SNL? Idolator jokes that they're making the sound bad on purpose for publicity, but in fact it sounds like the SNL sound guy used a default pop mix, making the vocals and next-lead instrument--in this case, they chose the sax in the horn section--loud and dry (with no effects), and putting most other instruments in the background. This type of mix usually accentuates the snare drum to maintain the high-energy rock feel (hear how loud it is?), and the bass (here, the low-end synth) to help pin everything down. Guitars? Backing vocals? Horn section? Sorry, we ran out of time.
This kind of mix is the default because it works with a strong singer--especially if they're using a backing track or lip-syncing portions of their performance--and it ensures that listeners don't have the most common complaint with live sound: "I couldn't hear the words."
But this mix doesn't work as well with art or indie rock bands, where the singers generally don't have highly trained dynamic voices and where the textures provided by the instruments are as important as the vocals. (I think of this as the old REM mix, used on their first few records, where you could barely hear Michael Stipe above the guitar.) Whoever did sound for Colbert seems to have been better prepared for this type of band, adding a little reverb to the vocals and giving the guitars much higher prominence. They also did a great job with the horn section and keeping everything balanced as more instruments got added to the end.
Of course, there are a lot of factors other than the skill and ears of the people doing sound--set-up time, number of microphones and channels, and communication from the band are all important. And, as a wise old producer once told me, you can't polish a bad performance. (He actually used a scatological term, but you get the idea.) But if you ever wonder why the sound sucked at a live show, this is an interesting case study.