• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Bass control differences between DAC? What to check with measurements?

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
1,185
Likes
386
#1
I bought 3 DAC in the last few months due to ASR. :D

My most recent DAC is Gustard x16. My impression is that X16 have better bass control than Topping E30. For example, feeling the bass shakes easily and cleanly with x16. E30 less so.

It is possible to measure bass control?

I have used a low cost ADC to capture RCA output from both DAC, dB matched as closed as possible within 0.2db, playing a few songs.

I have audacity app. Can audacity show bass control differences? What do I check for in the graphs?

Thanks!
 
Last edited:

polmuaddib

Active Member
Joined
Mar 1, 2020
Messages
147
Likes
146
#2
First, try DeltaWave software. It is easy to use. If it shows 99.99% or something like that the two files are the same, then look no further - the bass is the same, it's all in your head. But if it shows significant difference... but it won't.
Of course, then you can question if the low cost ADC captured properly...
 
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
1,185
Likes
386
Thread Starter #3
First, try DeltaWave software. It is easy to use. If it shows 99.99% or something like that the two files are the same, then look no further - the bass is the same, it's all in your head. But if it shows significant difference... but it won't.
Of course, then you can question if the low cost ADC captured properly...
Yeah, the issue likely will be the capture device. It has serious crosstalk issue and I can't even capture two channels at a time. But I still should look into Delta wave first...
 

pma

Major Contributor
Forum Donor
Joined
Feb 23, 2019
Messages
1,472
Likes
2,379
Location
Prague
#4
My most recent DAC is Gustard x16. My impression is that X16 have better bass control than Topping E30. For example, feeling the bass shakes easily and cleanly with x16. E30 less so.
What is the output impedance of those two DACs?

Second, though Gustard X16 has correct "fast" filter
1613667718475.png


all filters of Topping E30 are wrong
1613667756083.png


Believe me or not, this is important.
 
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
1,185
Likes
386
Thread Starter #5
What is the output impedance of those two DACs?

Second, though Gustard X16 has correct "fast" filter
View attachment 113351

all filters of Topping E30 are wrong
View attachment 113352

Believe me or not, this is important.
Per shenzhen audio, Gustard x16:
RCA Output level: 2Vrms VOLUME 00dB)
Output impedance 100 ohms
XLR Output level: 4Vrms(VOLUME 00dB)
Output impedance 300 ohms

I could not find such info from Shenzhen nor apos site for e30...
 
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
1,185
Likes
386
Thread Starter #6
First, try DeltaWave software. It is easy to use. If it shows 99.99% or something like that the two files are the same, then look no further - the bass is the same, it's all in your head. But if it shows significant difference... but it won't.
Of course, then you can question if the low cost ADC captured properly...
Do you know what this means? Thanks!

DeltaWave v1.0.61, 2021-02-18T12:19:54.6838371-08:00
Reference: topping_e30_optical_in_e30_minus15.5db_billie_elis_42seconds_right_channel.wav[L] 1852200 samples 44100Hz 16bits, stereo, MD5=00
Comparison: gustard_x16_optical_in_gustard_minus15db_bille_eilish_42seconds_right_channel.wav[L] 1852200 samples 44100Hz 16bits, stereo, MD5=00
Settings:
Gain:True, Remove DC:True
Non-linear Gain EQ:False Non-linear Phase EQ: False
EQ FFT Size:65536, EQ Frequency Cut: 0Hz - 0Hz, EQ Threshold: -160dB
Correct Drift:True, Precision:30, Subsample Align:True
Non-Linear drift Correction:False
Upsample:False, Window:Hann
Spectrum Window:Hann, Spectrum Size:32768
Spectrogram Window:Hann, Spectrogram Size:4096, Spectrogram Steps:2048
Dither:False
Trim Silence:False
Enable Simple Waveform Measurement: False

Discarding Reference: Start=0s, End=0s
Discarding Comparison: Start=0s, End=0s

Initial peak values Reference: -10.999dB Comparison: -11.231dB
Initial RMS values Reference: -25.245dB Comparison: -25.476dB

Null Depth=10.959dB
X-Correlation offset: 798 samples
Drift computation quality, #1: Excellent (1.54μs)


Trimmed 0 samples ( 0.00ms) front, 0 samples ( 0.00ms end)


Final peak values Reference: -10.999dB Comparison: -10.999dB
Final RMS values Reference: -25.243dB Comparison: -25.243dB

Gain= -0.2307dB (0.9738x) DC=0 Phase offset=18.09097ms (797.812 samples)
Difference (rms) = -63.63dB [-66.48dBA]
Correlated Null Depth=63.1dB [57.07dBA]
Clock drift: 0.08 ppm


Files are NOT a bit-perfect match (match=1.82%) at 16 bits
Files match @ 49.9802% when reduced to 10.73 bits


---- Phase difference (full bandwidth): 12.0009273149523°
0-10kHz: 9.14°
0-20kHz: 11.68°
0-24kHz: 12.00°
Timing error (rms jitter): 1.9μs

RMS of the difference of spectra: -109.327530018368dB
DF Metric (step=400ms, overlap=0%):
Median=-42.9dB
Max=-28dB Min=-49.1dB

1% > -49.04dB
10% > -47.25dB
25% > -45.88dB
50% > -42.9dB
75% > -40.61dB
90% > -35.29dB
99% > -10.27dB

PK Metric (step=400ms, overlap=50%):
RMS=-91.5dBFS
Median=-93.5
Max=-84.7

99%: -85.2
75%: -90.71
50%: -93.52
25%: -95.49
1%: -99.37

gn=1.02691066954221, dc=0, dr=7.85006309800256E-08, of=797.8117604836

DONE!

Signature: 0c79b1fb0b0a3302e092eca085067077
 
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
1,185
Likes
386
Thread Starter #7
What can I get from this? Bass ok, but not the high freqs?
Edit: This is not an accurate one due to using bad ADC and wrong file resolutions. Newer one below are more accurate using better ADC.


delta_of_spectra.PNG
 
Last edited:

polmuaddib

Active Member
Joined
Mar 1, 2020
Messages
147
Likes
146
#9
It means that the recordings are different. I believe that the author of this great software is a member and he might give more insight.

Is it possible that you percieve the lack of HF as a better bass?
 

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
8,103
Likes
16,673
Location
The Neverlands
#10
That is not a filter. That is audible high-frequency roll-off (-12 dB at 20 kHz).
The hearing limit of @Pdxwayne has been established to be 16kHz. The Gustard filter will still be 0dB there.
-3dB would be 19kHz.
To even the playing field @Pdxwayne should set filter 2 (L fast) which is similar to E30 sharp filter.
44.1kHz music rarely has >20kHz signals and never at 0dB so any mirrors will be at 24kHz which are fully removed by filter 2.

What can I get from this? Bass ok, but not the high freqs?
I assume this has been recorded with UCA202 at 44.1 or 48kHz ?
The plot shows up to 14khz the output is exactly the same in amplitude and phase so there is no worthwhile difference to speak of when it comes to the signal representing the sound of the recording.

I suggest a level matched blind test ... again. With a level playing field (both filters and amplitude) with say 20 'attempts'.

The output resistance of a DAC will be low in both cases and is not of any consequence when it is feeding a 'normal' amplifier.
 
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
1,185
Likes
386
Thread Starter #11
It means that the recordings are different. I believe that the author of this great software is a member and he might give more insight.

Is it possible that you percieve the lack of HF as a better bass?
Yes, I have been busy in https://www.audiosciencereview.com/...ave-null-comparison-software.6633/post-678356. Author very responsive.

I made many mistakes in my measurement setting and I think I finally get it together correctly.

The comparisons is about bagpipes high notes and not really bass related comparison. Indeed, lack of HF may be making X16 sounded to have better bass....
 
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
1,185
Likes
386
Thread Starter #12
The hearing limit of @Pdxwayne has been established to be 16kHz. The Gustard filter will still be 0dB there.
-3dB would be 19kHz.
To even the playing field @Pdxwayne should set filter 2 (L fast) which is similar to E30 sharp filter.
44.1kHz music rarely has >20kHz signals and never at 0dB so any mirrors will be at 24kHz which are fully removed by filter 2.



I assume this has been recorded with UCA202 at 44.1 or 48kHz ?
The plot shows up to 14khz the output is exactly the same in amplitude and phase so there is no worthwhile difference to speak of when it comes to the signal representing the sound of the recording.

I suggest a level matched blind test ... again. With a level playing field (both filters and amplitude) with say 20 'attempts'.

The output resistance of a DAC will be low in both cases and is not of any consequence when it is feeding a 'normal' amplifier.
I got lots of help in the detlawave thread. My last input is at
https://www.audiosciencereview.com/...ave-null-comparison-software.6633/post-678356

All captured with Scarlett Solo Gen3, right channel only. E30 with filter 1. X16 with filter 2.

Here is the difference between DAC when comparing 15 seconds of bagpipe high notes....this is just one channel, so you may need to consider both channels cumulative db difference. Audible?

Thanks!

x16_vs_e30_solo_right_channel.PNG
 

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
8,103
Likes
16,673
Location
The Neverlands
#13
Is this done using the UCA202 ?
There is no difference up to 10kHz.
The pagpipes and bass have no harmonic content to speak of above 18kHz and your ears do not hear that anyway. How could that change bass reproduction if the actual signal is exactly the same up to at least 10kHz. The differences above 10kHz could also be chalked up to the amplitude being small and noise levels of the UCA202 in the ADC being higher.
One is looking at difference signals at very low levels above 10kHz with a crappy ADC.
For all intents and purposes the signals of the DACs are the same.
The perceived differences are most likely to originate from something else beside a difference in waveform.

The only part that needs more investigation is the 'knowing what is playing' part.
For this reason I would recommend to AB (level matched, filter response matched) and blind. Meaning someone else plugs in either DAC and you don't know which DAC is playing. The tester may even use the same DAC many times in a row as long as the procedure (unplugging and re-plugging) is the same and there are no audible 'hints'.
Do 20 attempts.
 
Last edited:
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
1,185
Likes
386
Thread Starter #14
Is this done using the UCA202 ?
There is no difference up to 10kHz.
The pagpipes and bass have no harmonic content to speak of above 18kHz and your ears do not hear that anyway. How could that change bass reproduction if the actual signal is exactly the same up to at least 10kHz. The differences above 10kHz could also be chalked up to the amplitude being small and noise levels of the UCA202 in the ADC being higher.
One is looking at difference signals at very low levels above 10kHz with a crappy ADC.
As mentioned, I am using Focusrite Scarlett Solo for measurement now (single channel only). UCA202 not good!
 

pma

Major Contributor
Forum Donor
Joined
Feb 23, 2019
Messages
1,472
Likes
2,379
Location
Prague
#15
One is looking at difference signals at very low levels above 10kHz with a crappy ADC.
Yeah, this method gives no chance. I do not want to sound harsh, but talking from quite some experience.

I have some reservations to "make 20 attempts" or so. This way you will find only extreme sound differences. In case of tiny differences, you will lose concentration soon. This is difficult to speak about in a long distance on-line conversation, because only simplest topics are covered. And also the scale is too wide, from self-fabulation to well documented experiments.
 

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
8,103
Likes
16,673
Location
The Neverlands
#16
I am using Focusrite Scarlett Solo for measurement now
Missed that part. Try to record on the highest bitrate it supports and works well with.

The remarks regarding the levels of signals above 14 kHz remain though.

Look at the spectrum plots you made of the bagpipe section:
1613719650782.png


There is virtually no signal there and you are just comparing noise levels of DAC's and ADC above those frequencies. That's what you are looking at and comparing.
 

solderdude

Major Contributor
Joined
Jul 21, 2018
Messages
8,103
Likes
16,673
Location
The Neverlands
#18
Yeah, this method gives no chance. I do not want to sound harsh, but talking from quite some experience.

I have some reservations to "make 20 attempts" or so. This way you will find only extreme sound differences. In case of tiny differences, you will lose concentration soon. This is difficult to speak about in a long distance on-line conversation, because only simplest topics are covered. And also the scale is too wide, from self-fabulation to well documented experiments.
You don't need to do these 20 attempts the same day. I have done some tests over weeks (where the wife swapped or didn't swap connections when I was not home).
Statistical enough attempts are essential when you want to eliminate bias.

One more heretic comment - subjective perception of bass has nothing in common with frequency response in bass region, which is almost always flat in DACs. It is rather about mids and highs content and distortion.
correct. There were no differences upto 10kHz though. There won't be much harmonics changing the tonality of a bass above 10kHz. That is more in the 150Hz to 2kHz range.
20Hz to 8kHz is very easy for any DAC.
 

pma

Major Contributor
Forum Donor
Joined
Feb 23, 2019
Messages
1,472
Likes
2,379
Location
Prague
#19
Measurements to be done at the power amplifier output. This will catch possible influence of different DAC's ultrasonic and VHF content and its possible intermodulations in audio band. Nothing's easy, dude , except for overlooking ;).
 
OP
Pdxwayne

Pdxwayne

Major Contributor
Joined
Sep 15, 2020
Messages
1,185
Likes
386
Thread Starter #20
Missed that part. Try to record on the highest bitrate it supports and works well with.

The remarks regarding the levels of signals above 14 kHz remain though.

Look at the spectrum plots you made of the bagpipe section:
View attachment 113492

There is virtually no signal there and you are just comparing noise levels of DAC's and ADC above those frequencies. That's what you are looking at and comparing.
Shouldn't we focus more around 10 to 15 Khz differences?

Here is another look without using log view. Let's say at 10 to 11Khz, E30 is louder by 0.5db per channel. So, I assume both channels around 1db louder? Audible at all if 1db louder?

Regarding audio tests, my wife and daughter already sick of my last one near end of year between E30 and KTB....so little chance they will willingly help....
: (

e30_bagpipe_freq_analysis.PNG
 
Top Bottom