• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Measurements: "ESS Hump" revisited (Khadas Tone Board V1.3)

Veri

Master Contributor
Joined
Feb 6, 2018
Messages
9,596
Likes
12,036
Yes... but it is a symptom fix (and gives only slightly better results), not addressing the root cause... which probably cannot be fixed. It does not change the output signal of the DAC chip, only the OpAmp stage is changed and this never has any major impact (unless you run into clipping etc).

For a number of reasons his exact method cannot be applied here but I might try what's actually possible on this board. I'm not overly optimistic but we will see if that helps (significantly).

A smoother IMD+N orTHD+N vs Level plot does not mean there is significant less distortion of the presented kind. It may just mean the distortion is more constant in absolute levels, similiar to general circuit noise. And that more constant (=humpless) distortion base-line might be higher than the noise level. The guy didn't even check the spectra, let alone the residual. Looking at the IMD+N or THD+N vs Level plot is not sufficient.
BTW, the easiest way to drown the hump visually is to make a noisier output circuit. Then we'll have a lower SINAD but no apparent hump anymore (distortion hidden in the noise).
Is this all a fatal flaw in ESS9038Q2M ..?
 
OP
KSTR

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,690
Likes
6,013
Location
Berlin, Germany
Is this all a fatal flaw in ESS9038Q2M ..?
I would think it's neither a flaw nor fatal. The error is there but still very low in level, at around the analog noise floor. A bit worse than SOTA, but still basically OK in absolute terms.
I can't speak for the chip in general, only for this specific implementation of it (which is really low cost). The chip is very complex to use and has a myriad of setup options, plus ESS's documentation is severly lacking. Many ways to get smth. wrong here....
 

Veri

Master Contributor
Joined
Feb 6, 2018
Messages
9,596
Likes
12,036
I can't speak for the chip in general, only for this specific implementation of it (which is really low cost). The chip is very complex to use and has a myriad of setup options, plus ESS's documentation is severly lacking. Many ways to get smth. wrong here....
That's fair. Very interesting thread, in any case :)
 
OP
KSTR

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,690
Likes
6,013
Location
Berlin, Germany
Does that phrase not predate the IMD discovery by a good margin?
Or, were you just speculatively joking?
Gut feeling and educated guessing suggests me that if there is an audible difference between a "typical" ESS vs. say, an AKM, then this is it.
I will do loopback music recordings and further analysis and maybe even post ABX'able material.
 

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
3,219
Likes
1,172
Gut feeling and educated guessing suggests me that if there is an audible difference between a "typical" ESS vs. say, an AKM, then this is it.
I will do loopback music recordings and further analysis and maybe even post ABX'able material.
I have the original KTB. I used to thought it was sharp sounding before firmware update. Then after firmware update, it was OK, when voltage matched compared to E30.

Do you think a firmware update could change anything you measured?
 

Herbert

Addicted to Fun and Learning
Joined
Nov 26, 2018
Messages
527
Likes
435

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
3,219
Likes
1,172
As the hump became famous, don´t you think they would mention the improvement? They do not: https://www.khadas.com/post/tone1-upgrade-to-official-v2-firmware
I am not even sure if the hump was the reason for the harshness I heard. Also, I am not sure if they would/could openly tell us all the small fixes they made in the new firmware.

I am still curious what firmware version KSTR‘s KTB has. If old one, I am curious updating to new one would change any of his measurements or not.
 

tuga

Major Contributor
Joined
Feb 5, 2020
Messages
3,984
Likes
4,281
Location
Oxford, England
Does it happen only with particular ESS chips? The DAC3, which shows no hump, uses the ES9028PRO but resamples everything to 211kHz:

index.php
 
OP
KSTR

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,690
Likes
6,013
Location
Berlin, Germany
To put things a bit into perspective I compared the Khadas Tone Board to the RME ADI-2 Pro, at the same settings (240Hz@192k, -25dBFS).

As before in both cases the horizontal cursor line represents the analog noise base-line (RMS noise in a 22Hz...22kHz bandwidth). Any harmonic that is approaching or even exceeding that level will likely be audible in the noise as it is not buried anymore (assumed we already start to hear the background noise).

KTB:
FD -25dBFS.gif


RME:
1642792227350.png
 
OP
KSTR

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,690
Likes
6,013
Location
Berlin, Germany
Does it happen only with particular ESS chips? The DAC3, which shows no hump, uses the ES9028PRO but resamples everything to 211kHz:

index.php

My guess on other ESS DAC chips and implementaions is as good or bad as anyone's, so much for that. For now, generalization should not be made at any rate:
I would think it's neither a flaw nor fatal. The error is there but still very low in level, at around the analog noise floor. A bit worse than SOTA, but still basically OK in absolute terms.
I can't speak for the chip in general, only for this specific implementation of it (which is really low cost). The chip is very complex to use and has a myriad of setup options, plus ESS's documentation is severly lacking. Many ways to get smth. wrong here....
 

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
3,219
Likes
1,172
To put things a bit into perspective I compared the Khadas Tone Board to the RME ADI-2 Pro, at the same settings (240Hz@192k, -25dBFS).

As before in both cases the horizontal cursor line represents the analog noise base-line (RMS noise in a 22Hz...22kHz bandwidth). Any harmonic that is approaching or even exceeding that level will likely be audible in the noise as it is not buried anymore (assumed we already start to hear the background noise).

KTB:
View attachment 180822

RME:
View attachment 180823
Hmm....

A while back, I started a whole thread to compare ktb vs Topping e30, just because I thought ktb sounded pretty harsh sometimes.

When voltage matched, at least with my setup at that time and with my test song, I couldn't really hear a difference with blind tests.

So now you are saying that potentially the harshness I heard could be real and it might not due to voltage and filter mismatch?

Interesting....
 
Last edited:

gvl

Major Contributor
Joined
Mar 16, 2018
Messages
3,427
Likes
3,982
Location
SoCal
Topping D10B is based on the same chip as KTB, has lower noise floor than the KTB and no hump to speak of. So it can be done. AFAIK Topping likes to keep the ASRC engaged.
 
OP
KSTR

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,690
Likes
6,013
Location
Berlin, Germany
Topping D10B is based on the same chip as KTB, has lower noise floor than the KTB and no hump to speak of. So it can be done. AFAIK Topping likes to keep the ASRC engaged.
Just ordered a D10 balanced to find out myself. I'm expecting to see the same characteristic residual patterns though lower in level (low enough to hide in the noise floor) but maybe we'll see a surprise and it's really totally different.
 

noi1996

Member
Joined
Jan 22, 2022
Messages
6
Likes
4
Great thread!!!

I'm interested in how much of an effect hardware mods like the ones Soncoz designers did actually have on the ESS hump.
Changing the value of the RC network in the feedback of the op-amps in the IV stage should probably have an effect.
The latest Topping DACs seem to use 10 ohms and 470 pF.
 
Last edited:

gvl

Major Contributor
Joined
Mar 16, 2018
Messages
3,427
Likes
3,982
Location
SoCal
The moment I tap the oscillator (Quartz chrystals react to force on them) or cool it with cooling spray (they react to temperature also), the skirt "explodes" literally, indicating a rather strong momentary master clock frequency change.

Sounds like audiophile isolation pads isn’t such a dumb idea after all.
 

BoredErica

Addicted to Fun and Learning
Joined
Jan 15, 2019
Messages
629
Likes
899
Location
USA
No, the hump is only microscopically different here. I'm pretty sure the spectra and the residual patterns will be exactly the same compared to what I saw here.
Microscopically different as in looking at the IMD graphs?
1642839819680.png
1642839839189.png
The toneboard has about x4 the hump of the Motu, no?
 

Mnyb

Major Contributor
Forum Donor
Joined
Aug 14, 2019
Messages
2,645
Likes
3,626
Location
Sweden, Västerås
Forgive me if i’m wrong but is not Ben ( a member of this forum ) the designer of SONCOZ also the designer of the KTB ? Maybe he has some insights , this is technical beyond my skill set , so good luck with this curious investigation :)

However is not that far worse performance than this still is transparent when listening to actual music ?

I’m sure that when the root cause is identified some contrived test tone can be designed and it could be possible to hear this .

However pursuing perfection is never wrong , especially if it’s about implementation that only costs knowledge and not more expensive components.

ESS themselfs should have better documentation , that would benefit all products designed with thier chip set ?
There should be an ESS user group where people can help each other out .

And again thanks for the effort , this is the stuff that makes this forum great !
 
OP
KSTR

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,690
Likes
6,013
Location
Berlin, Germany
Microscopically different as in looking at the IMD graphs?
The toneboard has about x4 the hump of the Motu, no?
4x is 12dB and that is not much in my book. The point is that the pattern will be the same (likely, at least) which is the point I'm going to make ... because to my knowledge almost nobody has looked at the distortion residuals before. That's why I ordered a D10 B which has even nice IMD plot.
 
Top Bottom