Improving Speaker Reviews

To Doug Schneider,

I read your preliminary review of the Paradigm Founder 100F loudspeakers and was encouraged to provide you with some comments after you said you welcomed reader feedback.

First off, let me say that I have read many of your articles over the years, along with many other articles by the various other writers whose articles appear on your network. I have a tremendous amount of respect for the network of websites you have created. I feel these are some of the better audio-review websites out there. I also think that the loudspeaker frequency-response measurement database on your website provides a tremendous resource to the audio community. (I actually live almost next door to the National Research Council facility in Gloucester, Ontario, and bike through there a few days a week. I often wonder which building houses the anechoic chamber and think how cool it would be to be able to see how the measurement process works in there.)

Now, I do have some constructive feedback. I believe there are various ways in which you (and your columnists) could improve your reviews. Most of this would involve making the reviews a little more objective, and a little less subjective.

For example, in your column on the 100F, you talk about how the bass is not only full and deep, but also “tight and fast.” I am always skeptical when I read descriptions like this. There are a few reasons why I say this.

First, the science shows that bass is dominated by the room. So, when you listen to multiple pairs of speakers in the same room and describe their bass differently, are you describing the speakers or the room?

Second, I don’t think there is even such a thing as fast or flabby bass. I think fast bass is simply less bass and flabby bass is simply more bass. And again, this is dominated by the room. A speaker has extension, output, and distortion; however, the response itself must be EQ’d based on the in-room response.

The science also demonstrates that sighted listening can have a huge influence on our perception of the sound. For example, Floyd Toole has said that when he started working at Harman, many of the engineers there did not believe in blind testing. Instead, they felt their expertise allowed them to sort out the biases involved in sighted testing. However, after they developed one of their first blind listening tests, many of the engineers found that the speakers they “knew” (or rather, thought) they would prefer did not turn out to be the ones they actually did prefer in the blind test.

I think you could improve your speaker reviews by always listening blind against a reference. You already have one of the best references available with the Revel Ultima Salon2. If you were to set up your review speakers side by side with your Salon2s, EQ both sets of speakers below the room transition frequency so that both speakers have as smooth a response as possible in the frequencies that are dominated by the room, and test them blind, you would have a much more objective view of the various strengths and weaknesses of the speakers.

I did something similar in my listening room between pairs of GoldenEar Technology Triton Two and Revel Performa3 F208 loudspeakers. I did the testing in mono, after reading that Floyd Toole said it was easier to spot the differences this way. I set up one of each pair at the front of my room. I have miniDSP’s nanoAVR, which I used to EQ each speaker model below the room-transition frequency. And then I hooked up one speaker to the “A” set [of outputs] on my receiver, and the other to the “B” set. I also volume-matched each speaker. So, I was able to push two buttons to switch back and forth between the two speakers. I didn’t do the test blind (although I would like to), but in my sighted test, I found myself preferring the Revel F208 by a wide margin.

Anyways, that’s my feedback. Like I said, I think you do great work.