Originally Posted by goodvibes
In this case, using the volume control in the DAC is better than using an outboard dig control upstream because the inboard one is extremely well engineered, bit perfect and avoids an additional a2d-d2a conversion processes of a later one upstream. Using a later analog control may be preferred but it's no slam dunk that there's benefit. In this case, it would be difficult to incorporate and in your scenario it would see large signal at it's input so low level tracking could be an issue.
The noise isn't coming from the DAC. The larger signal level of full on is masking the noise of the opamps when the PS is dirty or there is a ground loop etc. The noise is not a product of the DAC V control.
You can read all my previous posts. I have never said the noise come from the DAC built-in volume control circuit because it is not well engineered.
In my case:
(i) the amount of current noise generated by my "dirty" IPAD 4 5V supply is always the same -> N
(ii) the signal strength coming out from the OPA1602 I/V conversion with volume level set at 255 -> S
When using the DX90 DAC built-in digital volume control, at any volume settling lower than the max volume level 255, the signal coming out from the OPA1602 (after I/V conversion) is lower than full strength: -> s
Using an external amp with analog or digital volume control at later stage, allow me to run "S" instead of "s" from the point after the OPA1602, which give better signal to noise ratio in the later 2 stages of OPA1611 and BU634 amplifications.
Can you see now how S/N is greater than s/N?
All you said can be true if the 9018K2M can drive the headphone directly like the all in one chip in IPAD 4. Unfortunately it is not the case.
Edited by borrego - 5/7/14 at 5:35am