Schiit Modi Multibit 2 vs. Topping E50. Multibit (+NOS) vs. Delta Sigma

[Sokel]

I first levelled both files as good as possible, and additionally normalized the volume to 0.1 dB with rx10. I think that's all I should have done.

The difference is in time you started recording,you can see it in the waveform.

 

[Bow_Wazoo]

Yes, well, I didn't cut the recording to within a millisecond. That's not a problem, but it doesn't really matter...

 

[Sokel]

Yes, well, I didn't cut the recording to within a millisecond. That's not a problem, but it doesn't really matter...

No,it's not as we get an idea.
As much as it seems similar the zoomed critical area differences are as much as -3db for W,that's clearly audible.
That -3db changes the whole balance in favor of the lower freqs,that's why the friend thinks "fuller" probably.

 

[charleski]

Just run a measurement sweep through both of them. It's quite reasonable to expect people would spot a 3dB response variation in the frequencies where we're most sensitive.

 

[Bow_Wazoo]

No,it's not as we get an idea.
As much as it seems similar the zoomed critical area differences are as much as -3db for W,that's clearly audible.
That -3db changes the whole balance in favor of the lower freqs,that's why the friend thinks "fuller" probably.

As I said, I can't get the files to a more accurate overall volume. According to MusicScope and RX10 analysis, the TPL of both files is equal to 0.1dB.

 

[Sokel]

As I said, I can't get the files to a more accurate overall volume. According to MusicScope and RX10 analysis, the TPL of both files is equal to 0.1dB.

Maybe you didn't understand what I said.
W and Y files are reasonably matched,no problem with that.
It's not that W file is -3db lower in the whole spectrum.
It's -3 db lower only in the zoomed area,there's more differences as well as you can see at the shape.
Is if is someone reduced the level between ~500-2000Hz in the W file in half.

That's a broader view before and after this freqs:

As you can see before ~500Hz is almost the same but after that...

It's not your mistake,the match was good,it's just that they sound different.

 

[Bow_Wazoo]

yes, i understood that.
But we have no influence on that.
The measurements of Schiit do not bring any conclusions

https://www.schiit.com/public/upload/PDF/APx%20Test%20Suite_%20Modi%20Multibit%202.pdf

 

[Sokel]

yes, i understood that.
But we have no influence on that.
The measurements of Schiit do not bring any conclusions

https://www.schiit.com/public/upload/PDF/APx%20Test%20Suite_%20Modi%20Multibit%202.pdf

No,we don't,is just how it's made.
As @charleski said a frequency sweep recording can be more useful to see exactly what's going on.

 

[Jimbob54]

Maybe you didn't understand what I said.
W and Y files are reasonably matched,no problem with that.
It's not that W file is -3db lower in the whole spectrum.
It's -3 db lower only in the zoomed area,there's more differences as well as you can see at the shape.
Is if is someone reduced the level between ~500-2000Hz in the W file in half.

That's a broader view before and after this freqs:


View attachment 278067

As you can see before ~500Hz is almost the same but after that...

It's not your mistake,the match was good,it's just that they sound different.

The important question for me then is not can one spot the difference, nor which is preferred, but which is more accurate?

 

[Sokel]

The important question for me then is not can one spot the difference, nor which is preferred, but which is more accurate?

That is easy to see if we have the original file,maybe @Bow_Wazoo can share it so we can compare (roughly).

 

[Bow_Wazoo]

No,we don't,is just how it's made.
As @charleski said a frequency sweep recording can be more useful to see exactly what's going on.

Good idea.
I'll make it

 

[Bow_Wazoo]

I have added the original files

That is easy to see if we have the original file,maybe @Bow_Wazoo can share it so we can compare (roughly).

Multibit vs. Delta Sigma - Google Drive

drive.google.com

 

[Sokel]

I have added the original files

Multibit vs. Delta Sigma - Google Drive

drive.google.com

Is that recorded or the original?
We want the original.

 

[Bow_Wazoo]

The first two are the originals.
W and Y the recorded.

 

[Jimbob54]

That is easy to see if we have the original file,maybe @Bow_Wazoo can share it so we can compare (roughly).

I know where my money is ;-) $5 says the E50 is the most accurate

 

[Sokel]

Let's see.
(we must consider that ADC comes to play too in the recorded ones but it's the same for both,so... )

Original (blue) vs Multibit (white)

...and zoomed:

Original(blue) vs Delta Sigma(white) :

..and zoomed:

Clearly different but judge for yourselves. @Jimbob54

Edit:Scales for better comparison

 

[Jimbob54]

Let's see.
(we must consider that ADC comes to play too in the recorded ones but it's the same for both,so... )

Original (blue) vs Multibit (white)

View attachment 278071

...and zoomed:

View attachment 278072


Original(blue) vs Delta Sigma(white) :

View attachment 278073

..and zoomed:


View attachment 278076

Clearly different but judge for yourselves. @Jimbob54

Edit:Scales for better comparison

The D/S looks particularly unhappy when zoomed 500- 2khz. Would have thought given previous comparison of the 2 DAC outputs it would be the MB looking more like that but what do I know!

 

[Sokel]

The D/S looks particularly unhappy when zoomed 500- 2khz. Would have thought given previous comparison of the 2 DAC outputs it would be the MB looking more like that but what do I know!

Yes,it's like it's boosted in that area (that's not distortion,only amplitude)
Wish we had more DACs to compare.

 

[Snoopy]

Can't say I care for schiit. They don't support DSD.. and input is limited to 24/192 pcm.

 

[pkane]

It doesn't look much in the above screen but zoomed enough in the area we perceive sound better and... :

View attachment 278061

There are very large timing errors between the two captures (not just a simple delay or clock drift, but large jitter, actual large timing errors between samples). With DeltaWave you can still align these, although it'll warn you that the error is too large. The level difference is 0.1dB.

Once you perform the alignment, the result "looks" much better than the raw, unmatched waveforms: