It's not dynamic range compression at all. I'd HATE that.
My short explanation is, your hearing becomes progressively less sensitive to bass as the sound intensity drops. If you listen to music at the exact level it was mixed at, no problem, everything is in balance. If you drop your play level by 20 or 30 dB, what you hear will sound thin. A fixed "loudness contour" adjustment takes a step closer to repair, but since music varies in level constantly, a fixed curve would only be the right correction at one level, in other words, wrong a lot of the time. And a traditional loudness control's action is coupled to a volume control position, and knows nothing of the actual sound level at your ears.
Dynamic EQ knows the exact sound level at your ear, so its applying the exact compensation for a play level offset. Since that exact level changes, and the exact correction is frequency dependent, it's not just an EQ curve, but dynamic correction. It's an exact compensation for what your hearing is doing at low levels, with the goal being to present the original mix balance at all play levels. If you turn up your play level to something like the original mix level, Dynamic EQ does nothing, it doesn't need to do anything. Dynamic EQ is level-aware, frequency-aware, and hearing response aware. Because of all of that, if you weren't aware that it was on, you wouldn't be able to tell except for the rather normal sounding spectrum at low levels. There's no side-effect, and no compression artifact. You'll hear the difference when you turn it off, though.
Remember, no compensation is actually wrong in this case, it's just that there's be no good way to do loudness comp until DSP came along, so it was always wrong, usually worse than nothing. Dyamic EQ is not "compensate using loudness contours" because loudness contours cannot ever be applied as a whole fixed curve. They are intended to show hearing response given a fixed level stimulus. Music isn't fixed in level or frequency, so the curves never really did apply.
Dynamic EQ is also not dynamic compression. which is level-blind above threshold, frequency-blind, hearing response blind, and under subjective control of whoever set it up. Dynamic compression is almost always audible, and could never be considered any sort of compensation, though very useful in mixing and processing for specific purposes, like broadcast.
That's at least how I would describe it. Here's their description:
http://www.audyssey.com/audio-technology/dynamic-eq
Some people freak out about changing anything in their audio dynamically. The problem is, your hearing has already done the damage that way, Dynamic EQ is simply the correct compensation. Turning it off actually imposes the rather significant error curve of your low level hearing response.
