Ringing False: Digital Audio's Ubiquitous Filter Filter Listening Impressions

Sidebar 2: Filter Listening Impressions

Although Keith Howard had sent me four copies of the listening-test DVD-A, the only writers in my team who had suitable players were Kalman Rubinson, Wes Phillips, and me. I gave discs to Wes and Kal, but Wes felt that the differences between the filters were too small, and his reactions to those differences too inconsistent, for him to be able to offer any meaningful comments. So the following discussion is of necessity restricted to Kal's and my reactions.

Neither Kal nor I knew what the filters were when we performed our listening comparisons. All we knew was that each set of eight samples of the four pieces of music began with the original, untreated track, followed by seven filtered versions, the filters used presented in the same order for each music excerpt. In the discussion that follows, the comments were condensed from our opinions of each filter, formed from listening independently to all four music excerpts. Kal and I didn't discuss our reactions at any time, nor did I look at Kal's comments until I had completed my own listening tests.

First, both Kal and I found differences very hard to identify, if at all. Kal reported that, after his first 40-minute listening session, his "brain was numb." Mine too. In fact, it took three sessions before I could get a handle on at least one of the filters. This was Filter 4, which I felt projected sibilants and rosin noise slightly forward in the soundstage compared with the unfiltered original, and tonally sounded a little "wiry." It gratified me to learn that Kal agreed, feeling it "thinner and a little harder than 1, 2, or 3."

We disagreed about Filter 2, which Kal thought sounded lighter and more forward; I also thought it more forward, but warmer-balanced overall. Kal found Filter 3 midway between the original and Filter 2. I found Filter 3 had less lower-midrange bloom around the thumps when the guitarist hits the body of the instrument in "Desert Flower," but otherwise it sounded identical to Filter 2.

We both liked Filter 5. I felt it very similar to the original in tonal quality, if a little flattened in perspective; Kal felt he liked it better than he did the original in the midrange, though he, too, described the presentation as "closer" to the listener. I found Filter 6 identical, or almost so, to Filter 3; Kal felt he could discern more detail with Filter 6.

Keith had asked us specifically to compare Filter 7 with Filter 4. Kal found 7 "fuller or duller" than 4; I found it slightly less "wiry" and closer to the original. Finally, Kal described Filter 8 as being a little brighter than Filter 5; I thought them identical.

In direct comparisons with the original music excerpts, I thought all seven filters reduced the sizes of the spaces around instruments and voices. However, this was not something I felt I could identify without direct reference to the originals.— John Atkinson