• 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!

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

krabapple

Major Contributor
Forum Donor
Joined
Apr 15, 2016
Messages
3,168
Likes
3,714
Apologies if any of this ground has been covered .. I admit have not read the full thread.

Gee, who could have guessed? There's your fundamental mistake right there.

And then you doubled down, spewing out all sorts of old, discredited audiophile nonsense arguments. Thereby provoking the usual helpful responses that you proceed to ignore/dispute with more nonsense. Even your wife could hear it. And then, the argument from authority: you've been a sound engineer, so.

Same old same old.

If only you had done your homework.

tl;dr: Apologies not accepted.
 

Trell

Major Contributor
Joined
May 13, 2021
Messages
2,752
Likes
3,285
Gee, who could have guessed? There's your fundamental mistake right there.

And then you doubled down, spewing out all sorts of old, discredited audiophile nonsense arguments. Thereby provoking the usual helpful responses that you proceed to ignore/dispute with more nonsense. Even your wife could hear it. And then, the argument from authority: you've been a sound engineer, so.

Same old same old.

If only you had done your homework.

tl;dr: Apologies not accepted.

You've too high expectations on his reading comprehension skills.
 

krabapple

Major Contributor
Forum Donor
Joined
Apr 15, 2016
Messages
3,168
Likes
3,714
I think they tend to say rather the opposite: ""If you just use music instead of test tones, all discrepancies will appear!"
 

Killingbeans

Major Contributor
Joined
Oct 23, 2018
Messages
4,088
Likes
7,544
Location
Bjerringbro, Denmark.
Both.

"The discrepancies between claims from golden-eared audiophiles and objective data will vanish! AND the discrepancies between the infinite complexity of audio and conventional test procedures will appear!"
 

Thorsten Loesch

Senior Member
Joined
Dec 20, 2022
Messages
460
Likes
530
Location
Germany, now South East Asia (not China or SAR's)
Hi

Is this DAC overpriced?

The name just rolls of your tongue Can’t really find any measurements or serious reviews.

Thanks

I had a look.

It is not directly a product I have designed, BUT it appears to be pretty much copied verbatim from the iONE nano that I did design, plus an inverted output to give "balanced", all the way to using MAX97220 as analogue stage (U7 & U8) and the TI TS5A22362 output mute switches (U9 & U10).

1671527718970.png


This means there is more distortion and noise than necessary and a subjectively less resolving and open sound.

I used the MAX 97220 in the iONE nano as was originally meant to be well sub 200 USD and specifically because I was required to make sure it did not sound as good as the micro iDAC 2.

You can see Amir's dashboard here:

https://www.audiosciencereview.com/...w-and-measurements-of-ifi-nano-ione-dac.4754/

The clipping at 0dBFS BTW is a manufacturing flaw that affected some batches of iFi products using the TS5A22362 mute switch. It has a TL431 regulator to feed the switch with 2.75V DC. But the IC procurement bought and that was fitted had the wrong pinout and fed the switch with 3.3V causing the "clipping" Amir observed. Later revisions resolved this.

I have long since stopped using this switch in favour of discrete MOSFET switches that give me more control and allow rail-2-rail output with vanishingly low distortion, like you find in the ZEN DAC. Not that Amir particularly liked it any better, dashboard here:

https://www.audiosciencereview.com/...ds/ifi-zen-dac-and-headphone-amp-review.9885/

At least it measures as it supposed to.

I think it would have been better if the "engineer" (I use the term in the loosest sense of the word, it is Toby O. who I know personally) at iFi would have copy/pasta scaped the line out from the ZEN DAC and used it here, that would have in essence would made it sonically the same as the older micro iDAC 2.

Seeing the 350 USD price tag I would be inclined to not spend my own money on this unit.

Thor
 

JSmith

Master Contributor
Joined
Feb 8, 2021
Messages
5,153
Likes
13,213
Location
Algol Perseus
the iONE nano that I did design
Welcome to ASR Thor. :)


JSmith
 

Thorsten Loesch

Senior Member
Joined
Dec 20, 2022
Messages
460
Likes
530
Location
Germany, now South East Asia (not China or SAR's)
Welcome to ASR Thor. :)


JSmith

Hmmm. I don't really have commercial interests.

I split from iFi in '19 and have no audio related products selling, other than the work I did for iFi.

Much of my current work in electronics has shifted to industrial automation, autonomous drones etc.

I might of course have audio products again in future. But right now nothing at all.

Thor
 

TSX

Active Member
Joined
Jul 15, 2022
Messages
105
Likes
104
Location
DK
I had a look.

It is not directly a product I have designed, BUT it appears to be pretty much copied verbatim from the iONE nano that I did design, plus an inverted output to give "balanced", all the way to using MAX97220 as analogue stage (U7 & U8) and the TI TS5A22362 output mute switches (U9 & U10).

View attachment 251300

This means there is more distortion and noise than necessary and a subjectively less resolving and open sound.

I used the MAX 97220 in the iONE nano as was originally meant to be well sub 200 USD and specifically because I was required to make sure it did not sound as good as the micro iDAC 2.

You can see Amir's dashboard here:

https://www.audiosciencereview.com/...w-and-measurements-of-ifi-nano-ione-dac.4754/

The clipping at 0dBFS BTW is a manufacturing flaw that affected some batches of iFi products using the TS5A22362 mute switch. It has a TL431 regulator to feed the switch with 2.75V DC. But the IC procurement bought and that was fitted had the wrong pinout and fed the switch with 3.3V causing the "clipping" Amir observed. Later revisions resolved this.

I have long since stopped using this switch in favour of discrete MOSFET switches that give me more control and allow rail-2-rail output with vanishingly low distortion, like you find in the ZEN DAC. Not that Amir particularly liked it any better, dashboard here:

https://www.audiosciencereview.com/...ds/ifi-zen-dac-and-headphone-amp-review.9885/

At least it measures as it supposed to.

I think it would have been better if the "engineer" (I use the term in the loosest sense of the word, it is Toby O. who I know personally) at iFi would have copy/pasta scaped the line out from the ZEN DAC and used it here, that would have in essence would made it sonically the same as the older micro iDAC 2.

Seeing the 350 USD price tag I would be inclined to not spend my own money on this unit.

Thor
Thank you very much for your reply. That was very interesting. I’ll find another product than what iFi makes.

Thanks!
 

Thorsten Loesch

Senior Member
Joined
Dec 20, 2022
Messages
460
Likes
530
Location
Germany, now South East Asia (not China or SAR's)
I found this https://www.soundstagenetwork.com/i...g-converter-measurements&catid=434&Itemid=577

Am I reading the measurements right, by placing this DAC in the 100-110 range of DACs?

Thanks

Well, this is where it gets interesting.

If you believe that better objective measurements reliably equals better sound, all ifi products I designed will do poorly, next to a copy/paste job from the datasheet for a more recent DAC Chip sold cheap.

But does that mean those relative numbers will make it sound worse?

As it so happens, I know (not believe) different. For example appx. -90dB 2nd harmonics @ 0dBFS is reliably inaudible (harmonic masking etc.) never mind that very few speakers and headphones even manage 60dB SINAD (0.1% THD) at rated power input or say 112dB SPL.

So in my view lowering the Harmonic distortion of a DAC further serves no purpose if we desire a device to use to listen to music.

Similarly, having (say) 112...113dB(A) (meaning A weighted) SNR is low enough noise that it will be reliably inaudible when playing music at 112dB peak at the listening position in an absolutely quiet (background noise = 0dB absolute).

To note, 112dB peak equals about 100dB average SPL with modern recordings that are quite compressed and around 92dB average with less compressed older recordings and audiophile drivel. That is pretty loud.

And rooms with 0dB background noise do not exist, even using IEM's which block noise you will find the noise of blood moving in your body louder than the "0dB" absolute SPL.

So again, in my considered view, very low noise much beyond appx. 112dB cannot be a valid design goal.

Of course, someone else may argue that even if there no audible benefit to lower distortion or noise, it does no harm. And if we have two devices that sound identical, cost the same but measure differently the better measuring one is a better choice.

Here I even agree. If a device is better than it needs to be objectively to be audibly transparent and does so without giving a poorer subjective perceived result, by all means.

BUT the issue of audibility of difference, emotional response to reproduced music (after we do not listen to music because we want to be bored) etc. are far from clear-cut.

Of course that is the point where usually an ABX fanboi cuts in and points out that in an ABX test pretty much everything sounds the same.

To which I reply that that the ABX test works as designed to combine expectation bias and poor use of statistics to hide relatively subtle differences, at which point the discussion often goes off track, so I will leave it at that.

I would suggest one should ask if the device measures objectively well enough that it noise and distortion (and frequency response deviations, crosstalk etc. et al) are of sufficient quality to not cause audible problems withusic and after to simply based on listening, personal preference and even looks...

Thor
 

Reynaldo

Active Member
Joined
Mar 17, 2021
Messages
232
Likes
97
Location
Brazil, Blumenau SC
@Thorsten Loesch

I already had several DACs and today I have the ZEN One Signature with the iPower X.
I really like the quality of this DAC, especially the Burr-Brown chip.
If you participated in this project, congratulations.
Numbers are important, but what each one likes to hear is also important.
I found your interview here.

 

TSX

Active Member
Joined
Jul 15, 2022
Messages
105
Likes
104
Location
DK
Well, this is where it gets interesting.

If you believe that better objective measurements reliably equals better sound, all ifi products I designed will do poorly, next to a copy/paste job from the datasheet for a more recent DAC Chip sold cheap.

But does that mean those relative numbers will make it sound worse?

As it so happens, I know (not believe) different. For example appx. -90dB 2nd harmonics @ 0dBFS is reliably inaudible (harmonic masking etc.) never mind that very few speakers and headphones even manage 60dB SINAD (0.1% THD) at rated power input or say 112dB SPL.

So in my view lowering the Harmonic distortion of a DAC further serves no purpose if we desire a device to use to listen to music.

Similarly, having (say) 112...113dB(A) (meaning A weighted) SNR is low enough noise that it will be reliably inaudible when playing music at 112dB peak at the listening position in an absolutely quiet (background noise = 0dB absolute).

To note, 112dB peak equals about 100dB average SPL with modern recordings that are quite compressed and around 92dB average with less compressed older recordings and audiophile drivel. That is pretty loud.

And rooms with 0dB background noise do not exist, even using IEM's which block noise you will find the noise of blood moving in your body louder than the "0dB" absolute SPL.

So again, in my considered view, very low noise much beyond appx. 112dB cannot be a valid design goal.

Of course, someone else may argue that even if there no audible benefit to lower distortion or noise, it does no harm. And if we have two devices that sound identical, cost the same but measure differently the better measuring one is a better choice.

Here I even agree. If a device is better than it needs to be objectively to be audibly transparent and does so without giving a poorer subjective perceived result, by all means.

BUT the issue of audibility of difference, emotional response to reproduced music (after we do not listen to music because we want to be bored) etc. are far from clear-cut.

Of course that is the point where usually an ABX fanboi cuts in and points out that in an ABX test pretty much everything sounds the same.

To which I reply that that the ABX test works as designed to combine expectation bias and poor use of statistics to hide relatively subtle differences, at which point the discussion often goes off track, so I will leave it at that.

I would suggest one should ask if the device measures objectively well enough that it noise and distortion (and frequency response deviations, crosstalk etc. et al) are of sufficient quality to not cause audible problems withusic and after to simply based on listening, personal preference and even looks...

Thor
Again. Thank you very much for sharing your insights
 

Reynaldo

Active Member
Joined
Mar 17, 2021
Messages
232
Likes
97
Location
Brazil, Blumenau SC
Which chip is the DAC?
U15?

It is hard to find Burr Briwn chips, other than the 24 bit ones.
Which one is the 32 bit DAC chip?
It uses Burr-Brown DSD 1793.

I have here another DAC from Musical Fidelity the M1SDAC that uses the Burr-Brown DSD 1796, but I don't know why Musical Fidelity left only PCM up to 192Khz, it doesn't have DSD.
Sound quality I like a lot.
Probably these DACs don't have the same measurements as others like Gustard, SMSL and Topping but for me the sound is much nicer.
I've had the Gustard X16, SMSL and prefer those with the Burr-Brown chips.
I still have the Oppo 205 here which has a fantastic DAC, but for my music taste the M1SDAC and the ZEN One Signature sound better.
 

Thorsten Loesch

Senior Member
Joined
Dec 20, 2022
Messages
460
Likes
530
Location
Germany, now South East Asia (not China or SAR's)
@Thorsten Loesch

I already had several DACs and today I have the ZEN One Signature with the iPower X.
I really like the quality of this DAC, especially the Burr-Brown chip.
If you participated in this project, congratulations.
Numbers are important, but what each one likes to hear is also important.
I found your interview here.


As long as you like the result and are happy, that is what counts to me most.

Thor
 

Thorsten Loesch

Senior Member
Joined
Dec 20, 2022
Messages
460
Likes
530
Location
Germany, now South East Asia (not China or SAR's)
Which chip is the DAC?
U15?

It is hard to find Burr Briwn chips, other than the 24 bit ones.
Which one is the 32 bit DAC chip?

U15 is WM8805 SPDIF Receiver.

U6, partially hidden by the big capacitors.

It is a DSD1793. The DAC accepts 32 Bit input. The actual SNR/DNR on the analogue side is equal to 18 Bit, which I think is enough.

If we expect a 32 Bit DAC to fully resolve 32 Bit implies around 194dB Dynamic range, thus there are no 32 Bit DAC's in this definition.

Thor
 

Reynaldo

Active Member
Joined
Mar 17, 2021
Messages
232
Likes
97
Location
Brazil, Blumenau SC
I dunno.
My question was that the spec sheet says it is a 24bit DAC.
And there are 28 legs on the chip, so I assume we have power, ground, a clock…
In the images that I put the DAC was or not reproducing files in 32bit?

What iFi did I don't know, as explained before it could be that they didn't update the specs.

One thing I know, iFi is very professional and correct in disclosing the configurations of its products.
 

Reynaldo

Active Member
Joined
Mar 17, 2021
Messages
232
Likes
97
Location
Brazil, Blumenau SC
Another important thing is that the specification says it supports 32-bit/384kHz PCM and fully native DSD
And it doesn't convert to 32bit.

Today many DACs convert from 24bit to 32bit, and this is not what is written.

Discerning Digital​

The ZEN One Signature offers the following to transform home music enjoyment:

  • True Native hi-res DAC – supports 32-bit/384kHz PCM and fully native DSD
  • Powerful 16-core XMOS processor – performs full MQA decoding (to 384kHz)
  • Upgraded GMT femto-precision clock and customised digital filter eradicate jitter and other forms of digital distortion
  • Coaxial S/PDIF digital outputs support 24-bit/192kHz PCM
 

Holmz

Major Contributor
Joined
Oct 3, 2021
Messages
2,018
Likes
1,241
Location
Australia
In the images that I put the DAC was or not reproducing files in 32bit?
correct.

What iFi did I don't know, as explained before it could be that they didn't update the specs.

One thing I know, iFi is very professional and correct in disclosing the configurations of its products.

I am assuming that they down convert the 32 bit stream, to fit into a 24 bit DAC buffer.

I guess technically if it accepts 32 bit integer data, then it Is a DAC that accepts 32 bits… and they can call it a 32 bit DAC.
But that is not the same as a FDAC chip doing a real 32 bits of resolution.

And if we want to argue that 32 bits is better or worse than 24 bits, then I am probably not the right guy, as I can hardly tell the difference between a CD playing and the same song streamed from an iPad though the identical preamp, amp and speakers.


Another important thing is that the specification says it supports 32-bit/384kHz PCM and fully native DSD
And it doesn't convert to 32bit.

Today many DACs convert from 24bit to 32bit, and this is not what is written.

I think it is converting FROM 32 bit, to 24 bit, to fit into the 24 bits of the DAC chip.
I dunno though, it is still a question, but the spec sheet suggests that it is a 24 bit chip.

Hence I am not going to get a “boner de jur” over this.
It is probably good, but I already have a decent DAC, and… as mentioned… I really do not hear a huge difference.
 
Top Bottom