Test Tracks for Evaluating Loudspeakers
This topic is sure to cause a lot of debate among readers. Everyone tends to have their own favorite tracks that they like to use to evaluate loudspeakers, and I’ve often found that tracks someone else is using to test a speaker gives me no insight whatsoever into how it performs. That’s not to say that they don’t for the listener, but I find they’re often listening for one note or small little passage in a piece so that it sounds “right” to them. As I’ve discussed in an earlier post on flat versus non-flat responses, sometimes “right” is what your ear is used to or what stands out the most. Other times, it’s just that you have to be familiar with the track to understand what you are listening for. I use some tracks as well because of a certain section that I’ve found speakers to have trouble with. But in those cases it’s a track that I’ve listened to hundreds of times on several dozen loudspeakers.
But even if there are a few notes that can reveal a speaker’s downfalls, one track isn’t the end all for any speaker I test. For example, a speaker might underperform on one piece of music but excel at another. Because of this, we like to evaluate our loudspeakers with multiple pieces and types of music before settling on a final design.
OK, so now that we’ve opened up the gates for a flame war, the real point of this post was to talk about the music I like to use to evaluate speakers and specifically what I’m listening to during those tracks. While you don’t need to use the same tracks I do, this article can hopefully this can give you a little insight how you might think about evaluating your speakers in the future.
The first CD I always start (and end with) while voicing a speaker is Revelator by the Tedeschi Trucks band. This CD has good, but not great recording quality. The music itself is amazing in my opinion, which also makes it enjoyable to demo. This CD works well for me for two reasons. First, Susan Tedeschi’s voice is very finicky on this CD. If you don’t have the crossover dialed in, her voice sounds either too shouty or too dull. Once you get it dialed in just right, her voice just pops. The second reason is that this band has a lot going on. There were 13 band members performing on some of these tracks. To me, this gives speakers a good test of low level resolution (often related to distortion performance). Lesser speakers fail to properly resolve and separate all of the instruments in some of the complex passages. They get muddied together and nothing really stands out. A good pair of speakers really adds separation and depth to the tracks and lets you pinpoint each instrument.
This is going to sound weird to most audiophiles – there may even be some that will stop listening to anything I write after this – but the second track I normally move on to is something in the heavy metal genre. I like to use Soilwork’s The Living Infinite album for this. Why, you ask, do I do this when most people would never consider using this type of music to audition and how do I even know what it’s supposed to sound like? Well again, there are a couple of reasons. First, it’s extremely easy for me to detect tonal imbalances in the frequency response when listening to metal. Many times, a boost in a certain range can sound good on certain music as it tends to elevate one instrument or piece of the mix. However, metal tends to highlight this because there is so much energy all across the spectrum. Things like a boosted midrange might sound fine on more easy listening type of music, but become shrill and piercing on metal. A bass boost around 100 Hz might make some music sound more full, but when you crank metal, you get a woofing sound on the lower end of guitars. I use this as a baseline check to make sure there are no significant issues, of course, I can always go back and measure, but this is much faster. Second, because metal can often be a full bandwidth assault, it really tests the overall composure of your speaker and can highlight power compression issues. Obviously big speakers tend to do better than smaller speakers for this, but you can still get a comparative measure of similar speakers using this.
Next, I usually move on to Nickel Creek’s track Eveline on the album Why Should This Fire Die? This track provides a good mix of string instruments and it also has a very dynamic section in the middle. This middle section gives a good indication of not only the dynamics of your speaker but also gives you clues about how much intermodulation distortion and Doppler distortion you might be hearing. This track is audibly cleaner at higher volumes on a prototype 3-way we have been working on than either of our 2-way kits. The prototype is still using the LDW7 as the woofer for reference, but it becomes apparent quickly that the midrange is being affected less.
Next, I like to break out the Keb’ Mo’ album Slow Down. Keb’ Mo’ typically has very good recording quality and a lot of tracks on this album have really nice, punchy bass lines. This also covers the male vocal portion of evaluation. Male vocals like Keb’ Mo’, who can sometimes hit lower register notes help identify upper bass boost issues. Too much energy in this region can sound chesty and resonant. Too little can sound thin and recessed.
I typically test the really deep bass output as well to really see how far a speaker extends. I often use Lindsey Stirling’s Crystalize for this. It’s a mix of dubstep and violins. If you love one but never liked the other, give this a try. It’s a really interesting track. There are many other tracks out there that you can use to test how deep your speakers play, but most normal audiophile tracks don’t play below about 35 Hz. Synthesized music usually has deeper bass and at this point I’m not checking for the tonality, only extension.
I typically listen for weeks before finalizing a design to make sure I have time to let any subtle issues sink in. Certain problems can sometimes need a lot of listening time to identify them. When I sometimes buttoned up a crossover only to weeks later decided there was something I wanted to tweak a little. Only through hours of listening to a ton of different music will lead to that conclusion.
However, I have to wrap this article up so I’ll leave you with one more piece of music that I use. I bought a disc from Walmart many years ago. It was a 3 pack of BBC recordings entitled BBC Big Band Sessions. I picked it up on a whim since it was only $9, but it turned out to be very well recorded with good dynamics. These CDs have everything from trumpet, saxophone, piano, clarinet, etc. on them, so it works well for evaluations. I played trumpet in a small band for 8 years, so I tend to think I have a good idea of what most of these instruments should sound like. This album lets me make sure that I’m getting that.
So that’s what I use and why. As I said, there's no reason you need to use these tracks or even agree with my selections. In the end, the most important thing is that you love the music you are listening to and that your speakers make you happy when they reproduce it. I’d love to hear your thoughts on what you use and why, or if you just have suggestions for good music, I’m always looking for more. Feel free to leave a comment on what you use.