Since Crescendo is an AudioUnit / VST plugin, we can extract maximum utility by using it in a DAW. The screenshot below shows my initial setup for calibration and for playback of iTunes tracks and website audio.
At the bottom left you see my mixer, above that, and to the right, I show the various plugins in use. At the very top I have a looped 10 second calibration sinewave recording set up to play.
Let’s look first at the mixer:
On the left is the audio track, labeled Audio 1, and which accepts live input from my In 1-2 audio channels. This track sends its output to Stereo Out, which defaults to Output 1-2 for me. And that output channel is just to the right of the input channel.
For iTunes playback, I have Soundflower installed for Input 1-2. For my output, I’m using a MOTU Ultralight Mk3.
Master controls the output level of all output channels, and is redundant in this case. So Master should be set for unity gain.
I have the input monitoring selected in the Audio 1 channel so that I can hear and process live input on Input 1-2.
I already calibrated my system, so both Master and the output channel levels are set to unity gain. These sliders should never be touched, in order to preserve the calibration. All volume adjustments should be made ahead of the output channel, either in the Audio 1 channel shown here, or at the source of the music in iTunes.
My output channel has really two separate and unrelated groups of plugins. The top group is my level monitoring and ear-candy, consisting of a Loudness meter, a Reverb, and CLAS.
The bottom group of plugins is my personal Crescendo, consisting of a dynamic range reduction compressor, my HyperRecruitment EQ, and Crescendo itself (aka vTuning).
Here’s just a quick glance at the ear-candy:
The Loudness meter is on the left, my reverb in the middle, and CLAS on the right. The only interesting thing here might be the settings I’m using on CLAS.
Remember from the Fletcher-Munson iso-loudness contours that any boost given at low bass or high treble have magnified effect, compared to boosts across the middle of our audible range.
So I’m only using 3 dB of boost at the softest levels on both bass and treble. And believe me, that is quite a bit… These boosts operate in SPL space, which translates to at least twice as much in Loudness space at those highs and lows. But I do like bass…
The really important thing to remember here is that all sound shaping should be placed ahead of the Crescendo processing, never after it. If you have hearing impairment, you don’t hear EQ and compression the same way that everyone else does. So if you want to hear what they hear, place it ahead of Crescendo.
Now we come to the actual Crescendo processing:
The top plugin is an RMS compressor with threshold -50 dB, ratio 1.5, and makeup gain 8 dB. This compresses the dynamic range of the incoming sound to about 2/3 of its original range, centered around my 0 dBVU level. It is spectrally neutral, not coloring the sound, and not trying to correct hearing at this point. Just reducing the overall dynamic range.
I do this because my hearing corrections are pretty strong, and it becomes much easier to reach those extremely faint levels in the reverb tails if I bring them closer to the average level. Crescendo really could dig out that faint stuff, but getting just the right vTuning level to reach them, and keep all the lower frequencies properly corrected – not too much, not too little – becomes touchy at these extreme correction levels.
If you have mild hearing loss, in the 30-40 dB range at 4 kHz, then you won’t need to do this.
Beneath the compressor is my HyperRecruitment and Decruitment EQ correction, just ahead of Crescendo. I only need 2 dB dip at 1.8 kHz, and 6 dB shelving boost above 4 kHz.
I also incorporated a 10 kHz hi-cut filter here, because I’m listening right now to some rock that has too much air in the recording. I probably can’t hear much above 10 kHz anyway. But that air gets boosted by Crescendo, if I let it, and the result pushes the output till it hugs the clipping level. So why bounce noise that will never be heard, and might lead to clipping overload? With this hi-cut filter in place, my output levels peak around -4 dBFS.
If you don’t have any HyperRecruitment nor Decruitment then you won’t need this EQ. But if you do, it is important to place it after any dynamic range compression, and just ahead of Crescendo, as I show here.
Finally, we see the Crescendo plugin itself. Here you see that I’m using vTuning = 58 dB which means that my 4 kHz threshold elevation is about 58 dB.
With that much vTuning I’m also giving Crescendo an additional 6 dB of Headroom, to avoid clipping output levels. I made up for that attenuation by increasing my amplifier gain during the audio calibration process.
And speaking of calibration, you can see that the looped -18 dBFS 1 kHz sinewave track produced a measured output in my headphones of 81.6 dB.
[ … and remember to disable all plugins during calibration, except for vTuning, which must remain enabled because of my Headroom setting. But even Crescendo should have its HdphX and Process disengaged… ]
I’m also using Sennheiser HD650 headphones, as indicated to Crescendo, and I have HdphX engaged to avoid ear fatigue from excessive stereo separation. If you use Crescendo, you owe it to yourself to get a pair of high-quality cans. It’s really worth it for the listening…
With this setup I have an absolutely wonderful time. It is totally addictive!
- DM
