During the filming of our new video series, we had the opportunity to record the output of our wacky creations using Earthworks TC30 microphones. These microphones have extremely accurate frequency and transient response which is why we chose them to represent the sound of our Will It Tune creations. I can now present fairly accurate wave forms for the “speakers” with and without DPS. Read on and get ready for some Sound Science!
The top wave form looks like the output of a fairly standard speaker. The “thicker” areas represent more RMS power (average power over time) and quick transients (the little spikes) peek out of the larger mass of the wave form. As expected, lower frequencies will show up as more RMS power because they require more energy to reproduce. Long vocal notes will also show up as more RMS power (and a thicker wave form) because the vocals are mixed loud in the song. And since RMS is power over time, they can be as powerful as the bass line in some cases.
Now take a look at the bottom wave form with DPS ON. See the difference in RMS power? The DPS technology is “active” which means it is constantly reacting to the incoming program material and making adjustments in real time to provide the maximum amount of power possible within the limitations of the playback device. These particular “speakers” (if you can even call them that…watch the episode!) had quite a few limitations. But by creating a custom profile for them, I was able to divert power away from frequencies where it was being wasted into frequency ranges that needed help. The DPS “toolbox” allows me to use this new power available from the small 20 Watt amp we were using to make the speakers sound pretty darn amazing. This is all done by hand/ear and, in case you were wondering, the DPS “toolbox” is NOT a glorified multi-band compressor. There is some special sauce at the core of DPS.
If you look closely at the DPS ON wave form you will see large peak transients spiking out of a fat and consistent body of RMS power. This wave form actually looks more like the wave form of the actual recording! However, the transients are actually LOUDER than those of the original recording. This has the effect of adding life, color, excitement and drama to the program material. It is these non-measurable intangibles that we strive for when creating profiles for DPS. So at the end of the day it looks like the DPS technology has not only distributed the power evenly across the audio spectrum specifically for these speakers, it actually INCREASES the dynamic range.
When all frequencies are tuned for a given DPS application, the device now has more headroom available. This headroom should result in an increase in volume but sometimes it can’t! The bottleneck becomes the actual digital audio engine itself. Most systems (Windows and iOS for example) use PCM audio as the backbone of their audio engines. This means “digital zero” is the loudest possible signal that can be reproduced before very nasty clipping will occur. Modern music is already mastered right up to this “digital zero” level so if we are INCREASING the dynamic range for a louder output, where can all of this new sound go? It’s like putting Dr. Bruce Banner into a Honda Civic and making him angry. The Civic may become slightly distorted! We’ve worked around this issue by building headroom into the Windows version (since it becomes a part of the whole computer audio system) and creating HIFI profiles for iOS. This isn’t a problem for integrated DPS applications since we are part of the design of the system.
Hopefully you found this technical insight into DPS interesting. Respond to this post if you have any questions or wish to elaborate.