Sony MDR-7506: the same 16 dB between dongle and desktop DAC, and it changes nothing
The 7506 needs 0.40 V and 2.51 mW for a 110 dB peak. The lowest-output dongle in the catalogue finishes 9.5 dB above target, the most powerful desktop DAC 25.6 dB. Both answers are «yes», and one of them costs $76.
- Gear in this setup
- FiiO KA13Check price at Thomann
- Sony MDR-7506Check price at Thomann
Yes to both, and the gap between them is wasted money. KA13: 119.5 dB. DX3 Pro+: 135.6 dB. Target: 110 dB.
Sony rates the MDR-7506 at 63 Ω and 106 dB/mW, which converts to 118.0 dB/V — the highest voltage sensitivity in this catalogue. A 110 dB peak therefore needs V = 10^((110 − 118.0)/20) = 0.40 V, drawing 2.51 mW and 6.31 mA. The KA13 delivers 1.188 V into 63 Ω after its own output impedance takes its share: 9.5 dB of headroom. The DX3 Pro+ delivers 7.588 V: 25.6 dB of headroom, i.e. 16.1 dB you will never use, because using it would mean listening at 135 dB.
This is the same comparison that ruins the ATH-R70x, with the opposite result. The two sources are 15.9 dB apart no matter what is plugged in; whether that gap matters depends entirely on where the headphone's requirement sits. The R70x needs 2.73 V and the cheap dongle fails by 7 dB. The 7506 needs 0.40 V and both sources clear it before the volume control leaves its first third. «Which DAC is better» is not a question about DACs.
The pearl: at these levels the volume control becomes the weak link. 0.40 V of the KA13's 1.22 V ceiling is peak level; ordinary listening sits 20-25 dB below that, i.e. in the bottom eighth of an analogue potentiometer's travel — exactly where channel-balance error on cheap pots is worst. With a sensitive headphone the audible problem migrates from the amplifier to the attenuator, which is why a high-output DAC can measurably worsen a 7506 while measuring better on every other test.
Damping, checkable here. FiiO publishes <1.7 Ω for the KA13 and Topping <0.1 Ω for the DX3 Pro+; against 63 Ω that is a ratio of 37 and 630 against the 8 the rule asks for. Both are clear. This is the exception — most manufacturers publish nothing at all, and then no one can say what the output impedance is doing to the bass.