A Favorite Test Album: Radiohead Amnesiac
Radiohead’s Amnesiac album is a favorite test album here. It tests many things about a high fidelity audio system that most other albums do not.
Many people say test tracks must be music where the musicians use exclusively acoustic instruments. That one can only compare a reproduction of the sound of a musical instrument to an instrument whose sound one is familiar with in the real world. The ‘absolute sound’.
First, people who use this technique aren’t any better, in my experience, at understanding what they are hearing when listening to a system than anybody else. And maybe a little below average at that.
Second, real acoustic instruments have a wide variety of sounds e.g. all guitars do not sound alike. So, which one are they comparing the sound on the stereo to?
Third, most acoustic instruments are extremely easy to reproduce. All percussion, flutes, most instruments in a 3- or 4-piece jazz composition, etc.
Fourth, classical music, the gold standard of acoustic music, is usually recorded so badly that it is largely just a wall of sound. Great systems help; but still, poor, so poor, oh so poor quality.
OK. Back to Radiohead. Back to heavily processed very complex very well-recorded music.
Using consensus-based evaluation, after listening to a track on many different systems, at least some of which do not suck, one learns where sounds are supposed to come from: their location in the sound stage, how far away they are, how large they are, where the are moving to, what the decay is supposed to be, etc.
One learns that some systems can disambiguate a pair of notes and some cannot.
One learns that some systems can communicate that the vocals are full of angst and others the vocalists always sound bored [voice is the only overlap with the ‘absolute sound’ people, though they seem to not pay much attention at all to this ‘most popular acoustical instrument EVER’ that we all are so familiar with – especially those with talkative spouses :-)]
One learns that many of the sounds on these tracks are so weird that they are extremely hard for inferior systems to reproduce at all well – it often will sound like something is broken (though sometimes that is the way it is ‘supposed’ to sound 🙂 ) [whereas reproducing a violin or piano, possibly the most complex acoustic instruments, can be done by the most modest of stereo systems well enough to fool some people all of the time].
Anyway, Radiohead in general is great for testing hifi systems, and especially the later albums where the quality continues to improve. Amnesiac is just the first Radiohead album that I happened to fall in love with.
Anyway, that is our opinion and these techniques have worked well for us. For example, they are extremely useful when we want to know immediately the capabilities of an unknown system.
I agree that listening to complex music and often not acoustic in nature can be much more revealing of a systems capabilities. It would be helpful and even enlightening if you detailed the tracks you like and what specifically you listen for? What can you pass on to make us all better listeners and better judge a system?
Hi Brian,
That is an interesting question. To come up with an algorithm, a step-by-step approach to scoring a playback system [and choosing test tracks to use].
I am going to post my original take on using Amnesiac’s first track to judge systems [first posted on the old Sprintricity magazine. We miss that place :-(]
Hope to run into you [several times :-)] at RMAF in a few days! 🙂
Take care,
-MIke
Let’s just say that while I think that your approach has some validity, I have a somewhat different approach that involves attending live concerts, usually the Houston Symphony or the Shepherd School of Music (Rice University) but occassionally other venues, such as the Symphony of the Pacific, the Sydney Opera House and a few European halls . The Shepherd School is interesting in that it has a number of venues which are optimized for particular performances, large orchestral, chamber music, opera and a very large custom built pipe organ. Those performances provide a basis aginst which I can evaluate the performance of my system. Yes, my system does suffer in comparison! At this point, rather than argue validity of approaches, I want to question your comment about availability of well recorded music. Assuming that you have access to vinyl recordings from the first decade and one half of the stereo era (late 1950’s through mid 1970’s) which were recorded with tubed electronics, I think that you do have a more than adequate source. Note that here I am thinking the better RCA Shady Dogs, selected Mercury recordings (SR90000), and most assuredly many Decca SXL 2000 and 5000 classical issues as well as selected EMI Columbia SAX issues to name a few. Yesterday I heard sarah Chang performing Barber’s Violin Concerto and was able to come home and listen to one of recorded performances. I am assuming that she was probably playing the same instrument.
Hi Folks! Looking forward to seeing some of you at Rocky Mtn ..
I like Amnesiac as a test album. I like later (i.e, later than Kid A) Radiohead in general, and I often use House of Cards (from In Rainbows) as a test track, to the point that everyone around me is probably sick of hearing it.
But … I disagree with Mike’s remark that ” …, classical music, the gold standard of acoustic music, is usually recorded so badly that it is largely just a wall of sound.” In fact I think that much classical music (including related genres such as classical opera and choral music) is extremely well recorded compared to everything else. Much of it is recorded at least somewhat ‘live’, in a hall , with the hall miked instead of the individual instruments. Compression is used gently, if at all. Full orchestra recordings are such a wonderful system test: complex + emotionally communicative.
Chamber music and small jazz ensemble are somewhat simpler to reproduce. Big classical and big rock and roll (to generalize) are much more difficult. Rock & roll and related (e.g., funk) are often so compressed for radio play that even great playback can’t sufficiently overcome the limitations of the original recording in terms of ‘big, open and Relaxed’ like a live performance would be.
I don’t know, Neli…. emotion?
Now, if they close mike the lead violin, or cello, then yeah, emotion can be heard. And the music is full of emotion, but the sound? Somewhere in the recording chain something is awry and steps all over the information stream.
Maybe it is just me, but it take a heckuva lot of mental hand-waving and imagination for me to recreate the orchestra and emotional content and playing skills of what are no doubt experts musicians out of the sounds heard during playback of most classical music.
I know this is kind of pointing at the elephant with no clothes on that people have by unspoken consent have chosen to ignore, but… maybe we shouldn’t ignore it anymore?
Close miking/ multi-miking of current classical recordings does create real problems; however, I am unsure why you have limited your criticism to the classical genre as many of the same abuses are present in music in general. Again, if you read my post you will note that I was very specific about what classical music I would use as a reference. I perhaps should have also said that there are a wealth of really good reissues of these early recordings. I am particularly fond of the ORG 45 rpm London reissues. In fact, many of their reissues whether classical or other genres are excellent as are the RCA Living Stereo reissues from Analogue Productions.
Hi Fred,
You are right, as stated in both your posts, that I was thinking mostly of classical on CD.
In many ways I brought up this apparently more controversial than expected topic as a straw man. To see if you all could shoot the idea down, to shoot holes in it.
The idea that even single-miking something as complex as an orchestra has profound limitations compared to a similarly complex piece of modern music more or less created on the soundboard.
And that this leads to use of modern music as audiophile test tracks if one wants to hear the full capabilities of an audio system.
I see this as primarily a limitation of microphones (something IsoMike is trying to address) but perhaps there are other limitations as well.
You certainly can teach me a whole heckuva lot of stuff about the best classical recordings – and that it would probably take the rest of my life to learn what you know 🙂 Sure would be fun … !
But at the high, abstract level of choosing test tracks, I was just venting my despair that I mostly use classical tracks for completeness sake, and not as a true scythe.
YMMV 🙂
Take care,
-Mike
It is always somewhat disarming when someone responds with a polite and courteous response. Again there is much truth in what you have to say. Even during the Golden Age of Analogue Recording, the equipment, particularly the mikes had limitations, although some of the old Neumann and Telefunken tubed mikes being used are legendary. What I perceive as the real difference is that the best engineers of the day understood the limitations of the equipment which they used and developed strategies to mitigate those weaknesses, such as use of one mike for voices, another for instruments. The same held true of equalization of tubed tape recorders. My point is that level of care and understanding is seldom present today and is what makes the difference. I suspect that the members of Radiohead care very deeply about the music they produce and that is why it sounds the way it does.