I listened to two CDs for the first time last night, and while I'm very happy with both of them, the disparity in sound reminded me how much production can make or break a record.
The first one was Planet of Ice by Seattle band Minus the Bear. I've read good things about this band for years, but the descriptions of this 2007 album--it's apparently more progressive and "math"-y than their previous releases--finally got me to make the purchase.
Musically it's a fantastic achievement, one of the most progressive and interesting modern rock albums I've heard since Mirrored by, which also came out in 2007. Odd time signatures, check. Guitar synths, check. Glitchy electronic breaks, check. Self-consciously repetitive pentatonic guitar riffs, check. Long songs with multiple parts and at least one false ending (the last song, "Lotus"), check. It reminded me of 70s progressive guitar/keyboard bands like Yes and Styx, but without the high-pitched vocals and lyrical pretentiousness. My wife heard some Police influence, and I have to mention Sunny Day Real Estate and Rush as well. It would be a great driving record. I'd love to hear the band live.
But as engaging as the music was, about 2/3ds of the way through the album I found myself tuning out, to the point where I had to force myself to pay attention. Some of my reaction was probably related to composition and album sequencing and instrument choice--it's hard to stay engaged through 10 songs that feature the same types of sounds--but the production played a part. The sound seemed to be concentrated in the middle frequencies, with a lot going on in three distinct EQ bands and very little going on outside those frequencies--there was almost no super-low-end and very little audible bass guitar or bass synth, for example. The overall tracks sounded compressed, with no peaks or valleys in volume. The stereo effects were not particularly noteworthy. In other words, the music was interesting and intricate, but the production was...well...a bit dull.
Then I put on Portishead's Third, the group's first record in 10 years, and remembered what incredible production can sound like. I have good memories of their first album, 1994's Dummy, which seemed to be the universal San Francisco soundtrack that year, but their second self-titled release left me cold for some reason--I think it was trying too hard to sound aggressive and ended up sounding monotonous. But Third is not only compositionally solid, with a pleasing variety of instruments and moods and tempos, but it's the best-sounding CD I've heard in years--the last time production impressed me this much was with Radiohead's Kid A in 2001 (their more recent records sound just as good, but I've come to expect it from them so am not surprised).
On Third, every audible frequency is represented. The stereo separation is incredibly wide--like listening to analog tape in a studio setting. (I wonder if they recorded to tape before moving into a digital workstation.) Many sounds are placed very precisely in the stereo field, while others--particularly Beth Gibbons' voice--are spread across the field and hard to place, as they should be. Some sounds are very dry and therefore seem to come forward, while others have enough reverb to sound like they're further back--including some of the drums, which is unusual but worked perfectly. There were lots of specific effects that weren't particularly complicated or hard to accomplish, but very effective--one repeated beat in "Machine Gun," for example, is dry on one side of the stereo field and duplicated with a slapback echo on the other.
The end result: Third was unignorable. Instead of being forced to rouse myself, I was constantly drawn into the sound. Instead of lying back on the couch, I sat up straight a couple times to try and figure out what was going on, and finally stayed sitting so I could keep my head in the perfect stereo sweet spot. I couldn't have carried on a conversation over it. It demanded attention.
For what it's worth, I agree with this Salon article arguing that Portishead's new album puts the final nail in the coffin of trip-hop, the genre the band helped invent back in the 1990s. But unlike the Salon writer, I think that's a great thing--the genre was confined to a half dozen acts, overhyped and played out almost from the moment it appeared, and retreading that old ground 10 years later would have been predictable and disappointing.