IDR, Dither and Noise Shaping
IDR relies on the L2’s internal 48-bit resolution to represent and process the audio signal, even if the input signal is 16-bit. This new, ultraprecise digital signal is then requantized to your desired, selected, final bit depth, be it 16-, 18-, 20-, 22- or 24-bit. Simple truncation of the 48-bit internal signal would result in rounding errors of the digital word and would produce distortion of low-level signals—a loss of resolution of low-level information such as reverb tails, ambience and directional cues that our ears require to reconstruct a mental image of a soundstage.

Requantizing by truncation is improved by adding a precisely controlled amount of noise called Dither to the signal. This can convert low-level, nonlinear distortion (noise) caused by the truncation of the 48-bit word into analoglike steady hiss. Since an increased noise floor is not desired, the noise is shaped to fit in the part of the audio spectrum least noticeable to the human ear. The IDR method would capture as much as 27 bits of perceived resolution even when outputting a 24-bit signal and as much as 28 bits when at 88.2 or 96 kHz (four bits better when at double sample rates). This technology is based on mathematical work and research done by company consultant, the late Michael Gerzon.

There are three Dither modes selectable on the L2. No Dither or Off is the same as the Bypass button when you want to just clone a copy but at a different bit depth, say, 24-bit down to 16. Type 1 Dither adds a 5dB increase of broadband noise to eliminate all low-level distortion and signal-dependent modulation artifacts. Type 1 is the choice for high-quality mastering applications. Type 2 is a narrow-band dither that is 5 dB quieter than Type 1, but with some low-level distortion. It’s a trade-off between adding less noise and keeping a little of the distortion.

Noise-Shaping options on the L2 are Off, M or Moderate, N or Normal, and U or Ultra. With Noise Shaping off, there is equal distortion, or noise, at all frequencies. The Moderate option reduces perceived hiss by 6 dB, while Normal reduces by 12 dB and is best for the creation of production masters. Ultra will reduce noise by as much as 18 dB, but the caveat is that no further digital processing or editing should happen to this signal because of the additional high-frequency energy used.

And the Rest
Waves has included a very good manual with the L2, because a lot of the concepts and operations are new and, if incorrectly used, can yield irreversibly poor results (such is the case with all powerful digital processors). Single-page “quick-start” instructions walk the user through the proper setup and operation. The unit is extremely easy to use—with no LCD screens or submenus to navigate—and once set up properly, operation is basically a one-knob adjustment and all front-panel settings are stored in nonvolatile memory for your next session.

The L2 Ultramaximizer makes a superb final processor for all digital audio recordings. Selling for $2,395, the unit is a worthwhile investment for any studio seeking to maximize both digital level and resolution while eliminating any danger of clip.

Waves, 306 W. Depot Ave., Suite 100, Knoxville, TN 37917; 865/689-5395; fax 865/688-4260; www.waves.com

Barry Rudolph is an L.A. based recording engineer. Visit his Web site at www.barryrudolph.com.

Back to Page 1

Reprinted with permission from Mix Magazine, May, 2000
© 2000, Intertec Publishing, A Primedia Company All Rights Reserved