Ess hump solution?


Hi Khadas team,

After read long post from ASR about ESS hump, it seems somebody already has the solution for ESS hump, will Khadas apply this solution for the new version of tone board? I think one op amp is not that expensive. Thank you.



Hi Khadas team,

After read some topic inside datasheet, may I ask the value of thd compensation register, more specifically register 13, 22-23, 24-25.
According to ds, thd compensation is disabled and all compensation bits are set to 0, I would like to ask how khadas board set those registers?

If I would like to change the value of them, how can I do it? Thank you.


It seems khadas board doesn’t use thd compensation registers at all. Will you do some tweaks in the new firmware release?


In regards to this post, is their a issue with the sound of the tone board? What is the Ess hump


Please read the link in the original post for details. In summary the ‘ESS hump’ is a hump that typically shows in the plots of intermodulation distortion vs. input level for a wide range of DACs using ESS chips, including the Tone Board. It’s probably inaudible, but not definitely so. There’s interest in what causes it, and how it can be avoided, as a small number of manufacturers seem to have avoided it.


Right I understand, saw the link, I will be avoiding the tone board until this issue is resolved. Sucks it’s someth that is easily identifiable.


Easily identifiable maybe, but not so easily fixed. The few who know the secret are keeping it close to their chest, and charging MUCH more for their DACs. It seems the fix will need a redesigned PCB, not just new firmware. For some perspective, even with the ‘hump’ the Tone Board measures better than anything I’m aware of at the price.


In that case hopefully Khadas or a 3rd party can release another option


I see somebody from ASR stated this issue can be solved and ESS engineers know how to solve it, can Khadas consult ESS and solve it in new revision???


Thank you very much for your feedback. We have communicated with ESS, and are discussing a solution.


Thank you, hope khadas can solve this issue without too much work.


is this ESS hump a real world problem ? for how many customers ?
I think I can live with it as I’m quite sure noone will ever notice anything wrong while listening to the whole audio system


It’s a problem, before do real ABX testing, nobody can tell the result. But from the measurement, it’s there.

Somebody from ASR shows some clue and a potential solution to this issue, hope khadas can solve it fast, and give customers a guide about how to solve/avoid it.


I wonder if it’s cased by the different filters the dac is capable of doing, was reading somewhere that it can do selectable filters to change the sound


No, filters won’t affect this.


Oh ok glad I ruled it out lol, apparently new dac is coming soon so hopefully it will be better