On DACs

[JoachimStrobel]

Oh, I think changing speakers is the most obvious audible difference for sure. Speakers are the most money I'd put in, followed perhaps by the amp. Although, I'm not sure if amps would make that much difference if the state of amplification one day approaches the current state of the art that DACs today have.

For instance, I thought I heard some sonic difference between my Topping PA3 amp and XTZ Edge A2-300 amp.., but I cannot be sure until some controlled blind-testing method is applied. Both definitely sounded perfectly fine through the speakers.

I have an 25 year old Yamaha amp in the bedroom. The volume control cracked and it got worse and worse. I then bought a Amp2 for my Raspi4 to drive the loudspeakers directly. This is a 60w Class D amp and I though it would be ok for my bedroom. Within 5 seconds after firing it up, I did not like it. I found the sound to be awful. Never thought that something like that is possible. I bought cleaning spray and fixed my old Yamaha amp. I placed the amp2 in the kitchen where it drives two low quality loudspeakers and does little harm.

 

[bobbooo]

Good way of putting it. Has Amir tested the onboard DACS on any widely-used phones? Would be interesting to see what the DAC specs are for a Galaxy 10 or the newest IPhone, etc.

Yes, in the Pixel 4A review, there is a comparison to the Samsung Galaxy S8+ (Qualcomm Snapdragon version), which is audibly transparent:

The S10, S9, S8, Note9 and Note8 series all have the same Qualcomm Aqstic WCD9341 DAC/amp, as do the many other phones available that use the same Snapdragon 855/845 chipset, which will perform similarly well. See my post here for more details. DACs in many other modern phones will be audibly transparent as well. As for iPhones, they haven't had analogue audio output for a while, but the $9 Apple 3.5mm DAC/amp dongle adapter to enable this is also audibly transparent.

 

[MrHifiTunes]

Hi all,

So, since an objectively good DAC is transparent and shouldn't add anything to the sound reproduction, among the top 10 DACs (by SINAD) reviewed on ASR, say between the $11,500 Mola Mola Tambaqui, the $750 Topping D90, the $500 Gustard X16, and, just in case, even the only $130 Topping E30, is it then correct to say these about them:

1. that there won't be any difference in the humanly audible sonic performance among them? (say psychological biases from knowing the prices, looks, etc, are removed through a properly designed and executed blind-test control).
2. that the relevant differences among them are only prices, aesthetics, material/build quality, features/connectivity, etc, everything else but the audible sound quality?
3. that if the audible sound quality and value are your only priorities (I know most of the times it's not the case but hypothetically), then you should always buy the Topping E30 than the 8,800% more expensive Mola Mola DAC?

Thanks!

Interesting questions. To put it in other words, If they measure the same, they should sound the same.
So we could here a difference in a DAC with THD+N of 120 and 119 DB ?
Why do companies bring out many DAC in different price ranges with same connectivity and usage? Or why pay 200-300 USD (or any nbr) more same measument results?

This is an interesting video aswell from a designer of ESS chip.

 

[BDWoody]

So we could here a difference in a DAC with THD+N of 120 and 119 DB ?

Our lovely and talented @RayDunzl came up with this a while ago to put some perspective to what those numbers actually mean.

https://audiosciencereview.com/forum/index.php?threads/the-shoutometer.2555/

These are very very small numbers we're talking about. That doesn't stop people from hearing things if they aren't using controls.

 

[TheWalkman]

Interesting questions. To put it in other words, If they measure the same, they should sound the same.
So we could here a difference in a DAC with THD+N of 120 and 119 DB ?
Why do companies bring out many DAC in different price ranges with same connectivity and usage? Or why pay 200-300 USD (or any nbr) more same measument results?

Why do people buy Maseratis instead of Hondas? Because people are willing to hand over bushel baskets of cash because the Maserati makes them feel all warm and fuzzy despite the fact that the Honda will get you to the office just as quickly.

 

[M00ndancer]

I have possibly the least electrical engineering knowledge of anyone on ASR, but I did have an experience that seems relevant: I have a friend whose job is engineering to minimize distortion in chips for cell phone amplifier tower, so all the specs Amir measures here are his daily bread and butter. I sent him some links to ASR reviews of expensive DACS and reviews from other places, of the "toe tapping" variety.

His response was "This is fascinating-- the chips in those DACS have been essentially the same for decades. We solved the problem of transparent conversion and amplification a long time ago, and every 1st year EE student knows it. ASR seems to be filled with people who are basically just having a ball analyzing and studying products in search of ingenious design solutions. The other sites are filled with hilarious nonsense from people using words they only kind of understand to misrepresent phenomena that they definitely do not understand."

I thought that was very telling, coming from an expert who has 0% investment in or experience with luxury hifi lifestyle culture, but total expertise in the science that underlies the equipment.

Best reference so far this year , I'm keeping this for future ref.

 

[MrHifiTunes]

Our lovely and talented @RayDunzl came up with this a while ago to put some perspective to what those numbers actually mean.

https://audiosciencereview.com/forum/index.php?threads/the-shoutometer.2555/

These are very very small numbers we're talking about. That doesn't stop people from hearing things if they aren't using controls.

I don't get the shoutometer thing. There is no system that can reproduce the sound of a barking dog 300 meters away is such a way that you can estimate the distance.( a hifisystem has a soundstage of a few meters)
So no system can reproduce more then 50 dB in sound differance?

 

[Killingbeans]

This is a 60w Class D amp and I though it would be ok for my bedroom.

I often see small Class D amp manufacturers being "creative" and setting the power rating at 10% THD. The normal 1% point is usually much, much lower. There's a risk that you were simply listening to a symphony of clipping?

I don't get the shoutometer thing. There is no system that can reproduce the sound of a barking dog 300 meters away is such a way that you can estimate the distance.( a hifisystem has a soundstage of a few meters)
So no system can reproduce more then 50 dB in sound differance?

It's not about estimating distance, but the intelligibility of the SPL.

When talking about hearing things at -120db, it should trigger the question: "While listening to music, can I really hear someone shouting at me from a distance of 650 miles/1000 kilometers?"

It doesn't even factor in the problem that at some point you'll need to be a vampire to avoid letting the noise of your own vascular system swamp the difference completely

 

[MrHifiTunes]

I often see small Class D amp manufacturers being "creative" and setting the power rating at 10% THD. The normal 1% point is usually much, much lower. There's a risk that you were simply listening to a symphony of clipping?



It's not about estimating distance, but the intelligibility of the SPL.

When talking about hearing things at -120db, it should trigger the question: "While listening to music, can I really hear someone shouting at me from a distance of 650 miles/1000 kilometers?"

It doesn't even factor in the problem that at some point you'll need to be a vampire to avoid letting the noise of your own vascular system swamp the difference completely

I get where you're coming from but that's even more confusing. So let's say I have the best ears and can hear someone shouting at me from 2km away.. that's about 66 Db. So all DACs should be considered Top notch. Why some get recommendations and others not?

 

[Killingbeans]

So let's say I have the best ears and can hear someone shouting at me from 2km away.. that's about 66 Db. So all DACs should be considered Top notch.

Exactly. Under normal listening conditions, including the ones that golden ear audiophiles experience in their setups, the bar for audibility of "dirty" DACs is probably set shockingly low. The dirtiness might become evident in some very specific situations if you know precisely what to listen for, but when you are simply enjoying music, even the most "revealing" system will be saturated by factors that dwarf the impact of the DAC's noise and distortion beyond any hope of audibility.

Why some get recommendations and others not?

Why Amir recommends some DACs and not others? I suspect it's mostly admiration for good engineering and/or good value.

I wouldn't put too much weight on the recommendations. Read the whole review and see what results made Amir reach his yea or nay, and then decide whether they hold the same value for you.

 

[MrHifiTunes]

Exactly. Under normal listening conditions, including the ones that golden ear audiophiles experience in their setups, the bar for audibility of "dirty" DACs is probably set shockingly low. The dirtiness might become evident in some very specific situations if you know precisely what to listen for, but when you are simply enjoying music, even the most "revealing" system will be saturated by factors that dwarf the impact of the DAC's noise and distortion beyond any hope of audibility.

Why Amir recommends some DACs and not others? I suspect it's mostly admiration for good engineering and/or good value.

I wouldn't put too much weight on the recommendations. Read the whole review and see what results made Amir reach his yea or nay, and then decide whether they hold the same value for you.

I think I still don't get it. My old CD player (1500 euro) have the same specifications as my Hifiberry DAC2HD (100 euro) and way above the mentioned 66Db. Still there is 15y of evolution between them and the hifiberry sounds better then the CD player, if used as DAC. Why is that? It is easily audio-able. Bass is tighter, image dept is deeper. (by no means the CD player sounds bad....but compared with other...) Where is that coming from?
Music is far more complex then a measurement signal (1 frequency and 1 amplitude). Many things happen, phase shifts and other electrical stuff which influence the output signal.

It reminds me of a old car review. A Chevrolet pick-up in comparison with a Ferrari. The Ferrari was way faster in the 0-60mph comparison ... But you can't conclude it was the better (read faster) car. The Chevrolet pick-up was faster on a test-track. Also faster from eg 40-80 mph.

 

[M00ndancer]

I think I still don't get it. My old CD player (1500 euro) have the same specifications as my Hifiberry DAC2HD (100 euro) and way above the mentioned 66Db. Still there is 15y of evolution between them and the hifiberry sounds better then the CD player, if used as DAC. Why is that? It is easily audio-able. Bass is tighter, image dept is deeper. (by no means the CD player sounds bad....but compared with other...) Where is that coming from?

Unless your CD player is broken or badly engineered there is no audible difference.

Quote from user Blake Klondike:
"I have possibly the least electrical engineering knowledge of anyone on ASR, but I did have an experience that seems relevant: I have a friend whose job is engineering to minimize distortion in chips for cell phone amplifier tower, so all the specs Amir measures here are his daily bread and butter. I sent him some links to ASR reviews of expensive DACS and reviews from other places, of the "toe tapping" variety.

His response was "This is fascinating-- the chips in those DACS have been essentially the same for decades. We solved the problem of transparent conversion and amplification a long time ago, and every 1st year EE student knows it. ASR seems to be filled with people who are basically just having a ball analyzing and studying products in search of ingenious design solutions. The other sites are filled with hilarious nonsense from people using words they only kind of understand to misrepresent phenomena that they definitely do not understand."

I thought that was very telling, coming from an expert who has 0% investment in or experience with luxury hifi lifestyle culture, but total expertise in the science that underlies the equipment."

It's quite simple, your brain is fooling you and succeeds, every time.
Mine and everyone else here as well. It's called perception bias.

There are lots of research and tests about the subject.

Grab you favorite beverage and start reading.

Bias in Auditory Perception

More interesting topics about the Lo-Fi human compared to the gear we have and use:

Audibility thresholds of amp and DAC measurements

Psychoacoustics Fundamentals

 

[Killingbeans]

I think I still don't get it. My old CD player (1500 euro) have the same specifications as my Hifiberry DAC2HD (100 euro) and way above the mentioned 66Db. Still there is 15y of evolution between them and the hifiberry sounds better then the CD player, if used as DAC. Why is that? It is easily audio-able. Bass is tighter, image dept is deeper. (by no means the CD player sounds bad....but compared with other...) Where is that coming from?

I seriously don't want to do this, so please don't bite me head off, but... there's a very real risk that it's comming from your imagination. Or rather, cognitive bias.

Music is far more complex then a measurement signal (1 frequency and 1 amplitude).

No matter how complex music is, it's still nothing but a bandwidth limited sum of single frequencies.

Many things happen, phase shifts and other electrical stuff which influence the output signal.

Phase shift is a displacement of a reference point in time. It means nothing unless it differs between different signals that need to be in sync. There's a lot of people out there who wants you to believe in all kinds of "electrical stuff" in order to keep the cash flowing. Don't listen to them. Find a real engineer and let him/her give you a crash course.

 

[rdenney]

CD players have analogue audio components in them, and those can have audible effects. I have half a dozen CD players, and while they all sound excellent, they do not all sound the same. I don't think that's the DAC, but rather error-correcting ability, jitter performance, and also the gain stages that follow the DAC.

And the assumption is that the objective is transparency, and clearly that is not always the case.

I have a Topping E30 on the way to use with a Cambridge Audio CXC I just snagged off ebay. I didn't spend more because there was no need, but I didn't spend less because I for what the Topping costs, I might as well have something in that can stand being strongly amplified and that has the connections I need for the intended purpose.

Sure, speakers are much worse. But this is a matter of making sensible optimization decisions. If I can make the DAC I'm using essentially perfect for very little cost (relative to the cost of the system), why not? That way, upgrading the transducers never runs into limitations of the electronics. What's the marginal cost of an E30 over something that would be essentially perfect? Not much, relative to the cost of any other component of the system, excluding the interconnects. For those saying the extra money should be put into speakers, my response is: What would that additional handful of dollars buy? And that's assuming the speakers aren't already in the system and therefore (mostly) a sunk cost. Maybe I'm wasting my time keeping combined system noise below -100 dB, but it's so cheap to do that even with stuff made 25 years ago that there's no reason not to, if I'm going to be buying the stuff anyway.

But mostly it's about good engineering, and rewarding it. Topping managed to design the E30 in such a way as to avoid marring the essentially perfect conversion performance of its DAC chip, and that cannot be said about all the DACs that Amir has tested.

My alternative was a Khadas Tone Board mounted inside the CD transport I bought the E30 to serve. Believe me, I thought about it.

Rick "who would buy a used higher-end DAC at the same cost if the performance was nearly as good and it looked good in the system" Denney

 

[honn]

CD players have analogue audio components in them, and those can have audible effects. I have half a dozen CD players, and while they all sound excellent, they do not all sound the same. I don't think that's the DAC, but rather error-correcting ability, jitter performance, and also the gain stages that follow the DAC.

And the assumption is that the objective is transparency, and clearly that is not always the case.

I have a Topping E30 on the way to use with a Cambridge Audio CXC I just snagged off ebay. I didn't spend more because there was no need, but I didn't spend less because I for what the Topping costs, I might as well have something in that can stand being strongly amplified and that has the connections I need for the intended purpose.

Sure, speakers are much worse. But this is a matter of making sensible optimization decisions. If I can make the DAC I'm using essentially perfect for very little cost (relative to the cost of the system), why not? That way, upgrading the transducers never runs into limitations of the electronics. What's the marginal cost of an E30 over something that would be essentially perfect? Not much, relative to the cost of any other component of the system, excluding the interconnects. For those saying the extra money should be put into speakers, my response is: What would that additional handful of dollars buy? And that's assuming the speakers aren't already in the system and therefore (mostly) a sunk cost. Maybe I'm wasting my time keeping combined system noise below -100 dB, but it's so cheap to do that even with stuff made 25 years ago that there's no reason not to, if I'm going to be buying the stuff anyway.

But mostly it's about good engineering, and rewarding it. Topping managed to design the E30 in such a way as to avoid marring the essentially perfect conversion performance of its DAC chip, and that cannot be said about all the DACs that Amir has tested.

My alternative was a Khadas Tone Board mounted inside the CD transport I bought the E30 to serve. Believe me, I thought about it.

Rick "who would buy a used higher-end DAC at the same cost if the performance was nearly as good and it looked good in the system" Denney

For instance, the difference in price between the E30 and Topping D90 is $619, which will get you the respectable Elac Debut Reference DBR-62, and still you have $19 left.

 

[rdenney]

For instance, the difference in price between the E30 and Topping D90 is $619, which will get you the respectable Elac Debut Reference DBR-62, and still you have $19 left.

I considered those speakers, and sort-of lust for a pair for one of my systems. But I really doubt they will produce fortissimo tympani strikes at 108 dB SPL out in my largish living room, which is required by one of my use cases.

Rick "responding to the statement that even spending what the E30 costs was excessive" Denney

 

[MrHifiTunes]

Unless your CD player is broken or badly engineered there is no audible difference.

Quote from user Blake Klondike:
"I have possibly the least electrical engineering knowledge of anyone on ASR, but I did have an experience that seems relevant: I have a friend whose job is engineering to minimize distortion in chips for cell phone amplifier tower, so all the specs Amir measures here are his daily bread and butter. I sent him some links to ASR reviews of expensive DACS and reviews from other places, of the "toe tapping" variety.

His response was "This is fascinating-- the chips in those DACS have been essentially the same for decades. We solved the problem of transparent conversion and amplification a long time ago, and every 1st year EE student knows it. ASR seems to be filled with people who are basically just having a ball analyzing and studying products in search of ingenious design solutions. The other sites are filled with hilarious nonsense from people using words they only kind of understand to misrepresent phenomena that they definitely do not understand."

I thought that was very telling, coming from an expert who has 0% investment in or experience with luxury hifi lifestyle culture, but total expertise in the science that underlies the equipment."

It's quite simple, your brain is fooling you and succeeds, every time.
Mine and everyone else here as well. It's called perception bias.

There are lots of research and tests about the subject.

Grab you favorite beverage and start reading.

Bias in Auditory Perception

More interesting topics about the Lo-Fi human compared to the gear we have and use:

Audibility thresholds of amp and DAC measurements

Psychoacoustics Fundamentals

Interesting lecture...Still I'm not totally convinced that this is the hole picture.
Still to this day they don't fully understand how the human ear works. Recently they discovered that the brain send signals to the hairs in the ear to make them move. So it's not only the incoming sounds-waves that make these hairs move.

I'm not an expert in wine tasting. I can't tell the difference between a 5 euro or a 20 euro bottle. I can tell which one I like. Some (read many) sommeliers can tell which kind of grapes are used and from which region the wine comes from. Those people can appreciate a 50 euro bottle for sure...I can't. It's beyond my ability...

I know some trained musicians, who can distinguish a stradivarius violin from any other brand. Some people can not even distinguish a violin from a cello and some are in between those 2 extremes. All those people are exposed to the same sound be it recorded or live.
As above both taste the same wine but different people have different evaluation/perception/appreciation/...

There are still many things unknown, undiscovered to this day. We measure the things we know and can interpret. But so many things we don't know...

 

[Killingbeans]

Still to this day they don't fully understand how the human ear works.

I agree. But the output signal from a DAC has nothing to do with the human ear. It's an alternating voltage and nothing else. Phase, frequency and amplitude... that's it. There's literally nothing more to it. What hits your auditory system and what it does whenever something hits it are two completely different things. A DAC only deals with the former.

Recently they discovered that the brain send signals to the hairs in the ear to make them move. So it's not only the incoming sounds-waves that make these hairs move.

Source?
EDIT: I guess that ESS talk? It points to increased resolution, not alternate sources of audio stimulus.

I don't quite understand it. The brain physically manifest the imaginary sounds?

Even if that's true, what does it have to to with a DAC?

I know some trained musicians, who can distinguish a stradivarius violin from any other brand.

Funny you should mention that :

 

[M00ndancer]

Interesting lecture...Still I'm not totally convinced that this is the hole picture.
Still to this day they don't fully understand how the human ear works. Recently they discovered that the brain send signals to the hairs in the ear to make them move. So it's not only the incoming sounds-waves that make these hairs move

There are still many things unknown, undiscovered to this day. We measure the things we know and can interpret. But so many things we don't know...

Recommended reading for new comers and inquisitive minds.

 

[rcstevensonaz]

I agree. But the output signal from a DAC has nothing to do with the human ear. It's an alternating voltage and nothing else. Phase, frequency and amplitude... that's it. There's literally nothing more to it. What hits your auditory system and what it does whenever something hits it are two completely different things. A DAC only deals with the former.

Source?

I don't quite understand it. The brain physically manifest the imaginary sounds?

Even if that's true, what does it have to to with a DAC?

Did you listen to the video clip from the CTO of ESS that was posted earlier? I don't have the background to judge the science; but he brings up several different elements of how a human hears and processes sound. Thus, factors that were not originally understood when designing DACs, were subtlety audible to listeners.