Nothing really special there. There is even no amp there, the AK chip used is a combination of a DAC with a standard around 30mW headphone amp that delivers around 1Vrms. Any iDevice has a similar one, many Android phones also do. The only difference is - ES100 uses 2 of them, and can drive many cans balanced. This gives it enough headroom to drive HD6xx. DFR can do the same job single ended, means it will not require a ballanced cable. HD6xx are very efficient, they don‘t really need power, but because of their 300Ohm they need a bit more voltage than a usual mobile device. 2Vrms are giving them enough headroom, and both ES100 and DFR can deliver those - one in a balanced mode, another single ended.
On an android phone ES100 cannot be better than DFR by definition when paired over BT. It‘s not because of BT, it‘s because when using BT, there is no way to bypass 48kHz upsampling and not that good audio subsystem of android. Even if in the BT developer settings you change the BT sampling rate to 44.1 from 48 kHz, an android phone would first upconvert everything to 48kHz, process it, and then downconvert to 44.1kHz.
This upconversion alone reduces the quality a bit, smearing the 3d imaging.
This issue is not present when an iDevice is used as a source.
On the other hand, ES100 doesn‘t reach the resolution of DFR, but still sounds very pleasing, as I said. My hope was to get a best semi-mobile SQ with it, I couldn‘t get that, that‘s why I returned my ES100. For a different purpose (like yours) it would be clearly better.
As I said, I was not looking for absolute mobility, but wanted to get a best semi-mobile sound out of the combo that can fit into my pocket when I‘m indoors in my house.
So, for now I found this to sound best to my ears after many a/b tests with a switch box (transport is ipod touch, it‘s permanently attached to this stack):