Serious Question: How can DAC's have a SOUND SIGNATURE if they measure as transparent? Are that many confused?

[raif71]

I can't think of why that would be a particular issue, and if not clipping the amp.....

You could use amp Y still, you just won't be able to use it fully.

Yes, we're in agreement that amp Y would not be used fully and hopefully use it before it got clipped.

 

[Chrispy]

Yes, we're in agreement that amp Y would not be used fully and hopefully use it before it got clipped.

But you think IMD is an issue?

 

[raif71]

But you think IMD is an issue?

Never crossed my mind about IMD but since we're measuring based on a single tone frequency as a starting point, we can never be sure the dynamics of music with its varying frequencies can affect how both amps will sound playing this music.

 

[Killingbeans]

I suppose if the DAC is able to drive both amps to reasonable hearing level say 80db, that should be ok. However 80db on a single tone may be different than driving a whole range of music frequencies. That may introduce differences, yes ?

Not in the DACs. Their outputs (hopefully) goes to high impedance inputs on a preamp or an amp. Not frequency dependent.

Never crossed my mind about IMD but since we're measuring based on a single tone frequency as a starting point, we can never be sure the dynamics of music with its varying frequencies can affect how both amps will sound playing this music.

Now you're talking about load dependencies in different power amplifiers. Doesn't matter at all when comparing DACs.

 

[Bow_Wazoo]

I'm currently in a total mind-shift.
Triggered by the Schiit products.
First the comparison of the A90D to the LYR+.
Followed by the E50 to Modi Multibit 2.
Fixed by the Modi Multibit 2 on NOS,
there was great curiosity about how even more exotic architectures in the DAC universe "sound".
That's how I accidentally ended up with the LAB12 DAC1 Reference.

dac1 reference - lab12

Non Oversampling, tube output DAC, sounding as closer to the vinyl sound as possible.

www.lab12.gr

The measurements are creepy,
but the sound is amazing.

1khz sine

10khz sine

IR

1khz stepped sine

The tube based Filter

FR

Distortion

I'm an electrician technician, and I love clean measurements.
That's why I've ended up at Topping again and again in recent years.
The Topping products also sound really good, but somehow also boring...

Could it be what Mike Moffat is saying? Multibit generally sounds more musical than Delta Sigma?

https://www.youtube.com/live/UeDNftTgaBA?feature=share

When comparing the E50 vs. Modi Multibit 2, I actually had the impression. Only in my head? Perhaps. Even if. If it works for me (becomes a reality)?! Why not.

And then this crazy construction of LAB12...
Sounds really cool. Measurements?!
A nightmare. Ultimately, meaningless.
Basically, I'm wondering about the need for a reconstruction filter, because as soon as I enable oversampling in the Neutron Player, most of the dirt is gone.
But don't use.OS. I'm feeding the DAC with 44.1khz (raw dope).

10khz sine + 4x OS

IR + 4xOS

Maybe I just didn't like the world out there.
I want to go back to the matrix...

 

[Killingbeans]

Only in my head? Perhaps. Even if. If it works for me (becomes a reality)?! Why not.

Absolutely. But personally I don't like the idea of spending €2900 on a DAC just to feed that part of my imagination.

If it's not in your head, I'd be more interested in finding out how the effect can be replicated much cheaper.

Can probably be done with a DIY line level buffer for a few euros/dollars.

 

[Bow_Wazoo]

Well, basically I live in the opinion that if you practice your hobby with reason, is down with me.
Nevertheless, 2900 € would not be worth the fun for me either. Actually everything was pure coincidence. Around the corner from me someone sold this dac for a very good price.
I couldn't resist. To be honest my plan was to listen to it, do some measurements, and sell it again for a profit.
But at least in my chain, this fat greek sounds very exciting, and authentic.

 

[antcollinet]

I'm currently in a total mind-shift.
Triggered by the Schiit products.
First the comparison of the A90D to the LYR+.
Followed by the E50 to Modi Multibit 2.
Fixed by the Modi Multibit 2 on NOS,
there was great curiosity about how even more exotic architectures in the DAC universe "sound".
That's how I accidentally ended up with the LAB12 DAC1 Reference.

dac1 reference - lab12

Non Oversampling, tube output DAC, sounding as closer to the vinyl sound as possible.

www.lab12.gr

The measurements are creepy,
but the sound is amazing.

1khz sine
View attachment 270903

10khz sine
View attachment 270907

IR
View attachment 270905

1khz stepped sine
View attachment 270904

The tube based Filter
View attachment 270901

FR
View attachment 270902

Distortion
View attachment 270900

I'm an electrician technician, and I love clean measurements.
That's why I've ended up at Topping again and again in recent years.
The Topping products also sound really good, but somehow also boring...

Could it be what Mike Moffat is saying? Multibit generally sounds more musical than Delta Sigma?

https://www.youtube.com/live/UeDNftTgaBA?feature=share

When comparing the E50 vs. Modi Multibit 2, I actually had the impression. Only in my head? Perhaps. Even if. If it works for me (becomes a reality)?! Why not.

View attachment 270909

And then this crazy construction of LAB12...
Sounds really cool. Measurements?!
A nightmare. Ultimately, meaningless.
Basically, I'm wondering about the need for a reconstruction filter, because as soon as I enable oversampling in the Neutron Player, most of the dirt is gone.

10khz sine + 4x OS
View attachment 270912

IR + 4xOS
View attachment 270910

Maybe I just didn't like the world out there.
I want to go back to the matrix...

View attachment 270908

Holy hell that is truly awful. It looks like it has almost no reconstruction filter, resulting in the "audiophile" perception of digital as being a stair step waveform.

perfect voicing that resembles the very best qualities of analog

Do you think any analogue source would present a 10KHz sinewave looking like that?

And oversampling "most of the dirt is gone"? Are we looking at the same waveforms? You can still see the "stair steps" albeit slightly smoothed by the small level of filtering there actually is - but now it looks like there is some high frequency fuzz added.

I can't quite work out if you are also asking if this is more "musical" or not. I can't say, because "musicality" for reproduction equipment is a meaningless concept. What it certainly is not though, is HIFI.

If this sort of distortion generator is what you like - stick an effects box in your stack somewhere.

 

[Bow_Wazoo]

Yes, after many others, that's my favorite chain up there

PS
The rather powerful DSP "effect box" (Neutron Player), that is the beginning of my chain,
doesn't produce the same effect.

 

[sonder]

Multibit generally sounds more musical than Delta Sigma

It's not the heaviest reading, but this article: Delta Sigma vs Non-oversampling (NOS) R2R DAC Designs from '15 is an interesting read.

[edited out distracting pointless bits]

 

[Killingbeans]

PS
The rather powerful DSP "effect box" (Neutron Player), that is the beginning of my chain,
doesn't produce the same effect.

Maybe because it doesn't have a checkbox that says 'add a truckload of distortion'?

It's not the heaviest reading, but this article: Delta Sigma vs Non-oversampling (NOS) R2R DAC Designs from '15 is an interesting read.

[edited out distracting pointless bits]

Not very interesting. A ton of conjecture, and nothing useful to be found, as far as I can tell.

 

[steve59]

I've got a pioneer PDR-19rw with a manufacture date of 03/99 on it. If I use the digital out from this unit and run it thru the built in da1 chipset in my c49 or even more current chipset in a Project pre box s2 digital with levels matched all 3 will sound identical?

 

[Bow_Wazoo]

For this I only need the preamp control.

Maybe because it doesn't have a checkbox that says 'add a truckload of distortion'?

For this I only need the preamp control.

 

[GXAlan]

Absolutely. But personally I don't like the idea of spending €2900 on a DAC just to feed that part of my imagination.

If it's not in your head, I'd be more interested in finding out how the effect can be replicated much cheaper.

Can probably be done with a DIY line level buffer for a few euros/dollars.

It works both ways. Everyone should have a transparent system with good measurements. But you cannot force yourself to like it if you somehow don’t like it. At that point, finding something you subjectively like is fine. You should always have a reference to compare to and enjoy since you can do so for cheap.

Likewise, if you like transparency, given the low barrier to entry to that club, don’t got searching for euphonic colorations…

 

[majingotan]

I'm currently in a total mind-shift.
Triggered by the Schiit products.
First the comparison of the A90D to the LYR+.
Followed by the E50 to Modi Multibit 2.
Fixed by the Modi Multibit 2 on NOS,
there was great curiosity about how even more exotic architectures in the DAC universe "sound".
That's how I accidentally ended up with the LAB12 DAC1 Reference.

dac1 reference - lab12

Non Oversampling, tube output DAC, sounding as closer to the vinyl sound as possible.

www.lab12.gr

The measurements are creepy,
but the sound is amazing.

1khz sine
View attachment 270903

10khz sine
View attachment 270907

IR
View attachment 270905

1khz stepped sine
View attachment 270904

The tube based Filter
View attachment 270901

FR
View attachment 270902

Distortion
View attachment 270900

I'm an electrician technician, and I love clean measurements.
That's why I've ended up at Topping again and again in recent years.
The Topping products also sound really good, but somehow also boring...

Could it be what Mike Moffat is saying? Multibit generally sounds more musical than Delta Sigma?

https://www.youtube.com/live/UeDNftTgaBA?feature=share

When comparing the E50 vs. Modi Multibit 2, I actually had the impression. Only in my head? Perhaps. Even if. If it works for me (becomes a reality)?! Why not.

View attachment 270909

And then this crazy construction of LAB12...
Sounds really cool. Measurements?!
A nightmare. Ultimately, meaningless.
Basically, I'm wondering about the need for a reconstruction filter, because as soon as I enable oversampling in the Neutron Player, most of the dirt is gone.
But don't use.OS. I'm feeding the DAC with 44.1khz (raw dope).

10khz sine + 4x OS
View attachment 270912

IR + 4xOS
View attachment 270910

Maybe I just didn't like the world out there.
I want to go back to the matrix...

View attachment 270908

That's some Schiit-zophrenic looking Sinusoidal waves there!

Unlike you, I didn't like the NOS sound and prefer Moffat's version of closed-form oversampling filter instead.

Triggered by the Schiit products.
First the comparison of the A90D to the LYR+.
Followed by the E50 to Modi Multibit 2.
Fixed by the Modi Multibit 2 on NOS,

I've had Schiit-y products before ASR and they've been incredibly reliable to me. My Schiit Saga OG preamp survived a coffee spill on the vents and my SET amp survived a 1 meter drop of DCA Aeon X Open headphone drop on the KT88 tubes while both are powered on without damaging the tube socket and the tube itself (the tube glass has surface scratches though but still works perfectly)

My desktop chain (Bifrost 2 OG + SET amp) might have horrendously audible distortion (~32dB to 40dB SINAD guesstimate or 1-2.5% THD+N, heck my Aeon X has lower distortion than this lol) but it doesn't sound that marginally different than my 115 dB SINAD (0.0018%) Chord Mojo 1 or the 106 dB SINAD (0.0005%) DAP

 

[DSJR]

I thougght this was over, but back in the past, there were far too many audio systems balanced around vinyl playback. For too many years, just slotting a digital source (CD or otherwise) upset the 'sound' as heard by people too well used to the 'tone' of their sound systems and blaming th enew digital source as the problem. The fact an Audio Note article was linked to as evidence put me off completely I have to say...

 

[sonder]

Observation which I hope doesn't create a lot of noise in answers: purely digital music seems to sound better through delta-sigma DACs, and conversely anything originally analogue less so.

I'm left wondering if the double pass of delta sigma modulation, first on input(s) via ADC, then again on output via DAC is a large factor here.

That is, only having one pass of delta sigma throughout the lifespan of a piece of music, may render "nicer" results.

Digital only music out through a DS DAC.

Originally analogue music (already been through an ADC at recording time) out through an R-2R DAC.

Multibit / custom FPGA for a middle ground.

Any input / somebody to rip it to threads or point me to more reading on the subject?

 

[BDWoody]

Observation which I hope doesn't create a lot of noise in answers: purely digital music seems to sound better through delta-sigma DACs, and conversely anything originally analogue less so.

It's generally going to boil down to those making those observations not taking normal human bias into account when doing the comparison.

There isn't going to be anything inherently more or less 'musical' about one type or another.

I've moved this into one of our catch all threads where this has been covered.

 

[Bow_Wazoo]

Yesterday I measured the FR again above 20kHz ,to look at the tube filter installed in the DAC1 Reference at the output (green graph).
If I activate the Neutron Player 4x OS and transfer the data to the DAC at 176.4kHz, I get filtering (red graph) as you would expect from a standard FIR filter. I'm just wondering where this filter comes from? It can't be from the DAC. Apparently none are installed. So only the Neutron Player would remain?!
I've asked both the Neutron Player programmer and the DAC designer, but I haven't received a clear answer to this day
Can someone enlighten me?

 

[solderdude]

From the OS algo. It has to have a steep filter.
If that is in the Neutron player then that's where it comes from.