DO ALL DACS SOUND THE SAME??

[solderdude]

But you're talking about measurement the likes of dinamic range, Signal to noise ratio and distortion, done with measuring equipment.

Yes, we are generally talking about the signal quality. 2 voltages changing amplitude over time. There is no sound nor any sonic properties.... just 2 changing voltages.
Very easy to measure and is measurable far far beyond and below any human perception.

Sound is much more than dinamic range, Signal to noise ratio and distortion, and we have two ears, not measuring equipment.
Just my 0.02 $.

Sound is air being moved and indeed there are many errors in electric-acoustic conversion.
This can be measured and analyzed but has a substantial room for errors in those measurements.

Then there is the part of acoustics (which can also be measured but is complex so those measurements must be taken with a kilogram of salt).

Followed by the auditory system picking up those vibrations in the air.
And finally and most importantly where the music is perceived ... the brain and all of its quirks.

The transducer bit is already problematic, followed by acoustics which can be a real biatsch.
And to top it off ... perception which its huge variability and input from many other sensors, memory and all kinds of feelings.

Now .. where are the problematic aspects ?
In the electric domain where everything follows laws of physics or the brain with its many, many functions that are intertwined ?

 

[Nutul]

If you have always, and only, listened to your vinyl collection, and now listen to the very same albums (digitized versions) through a DAC, of course you'll find the result clearer, more defined/refined, even analytical.
That's not because the DAC is "cold". That's because the vinyl is too much "round and smooth" and limited to its poor FR.
The real thing is not what the vinyl portrays. Full stop.
If you're not prepared to hear what the intended recording sounded like, then, don't go digital; stay with vinyl.

 

[Roland68]

If you have always, and only, listened to your vinyl collection, and now listen to the very same albums (digitized versions) through a DAC, of course you'll find the result clearer, more defined/refined, even analytical.
That's not because the DAC is "cold". That's because the vinyl is too much "round and smooth" and limited to its poor FR.
The real thing is not what the vinyl portrays. Full stop.
If you're not prepared to hear what the intended recording sounded like, then, don't go digital; stay with vinyl.

What or whom is this post referring to?
I can't see any connection to any of the previous posts.

 

[Nutul]

What or whom is this post referring to?
I can't see any connection to any of the previous posts.

This one. Very first paragraph, latest words.

AKM vs ESS

Oh. Didn't know this. Must be my hearing.....or system. https://www.audiosciencereview.com/forum/index.php?threads/message-to-golden-eared-audiophiles-posting-at-asr-for-the-first-time.17598/

www.audiosciencereview.com

 

[bungle]

It is a bit funny when people talk about modern AKM vs ESS. Then they obviously need good speakers to test them. Ok, they pick say Genelec 8361A (https://www.audiosciencereview.com/...s/genelec-8361a-review-powered-monitor.28039/), a speaker with good reviews. Then they test in an anechoic chamber the AKM vs. ESS, just to get analog signal to Genelec. Then Genelec does ADC and finally its own DAC. What are we testing at the end? DAC (and ADC) is effectively a solved problem. We are only left to analyze if the implementation has gone wrong in some ways and badly (which is not a usual case). The chips do work, and have worked for a long time, especially on human ear audible range. Genelec definitely does not use latest and greatest chips (you cannot feed them with say DSD or use high bit rates, many models will max at 192 kHz / 24 bit, some even at lower, and they may re-sample to 48 kHz or 96 kHz internally when doing DSP -> in many threads you see here people complaining this too). They use chips that costs couple of euros at max, like 8351: AKM AK4621EF (see: https://www.audiosciencereview.com/...enelec-8351b-teardown-2nd-disassembled.22785/ - was the chip introduced about 10 years ago).

 

[TTT15]

I very much doubt that I could tell the difference between an ESS and a AKM chip but define work? For sure a DAC chip can very accurately reproduce a frequency with inaudible distortion and noise but by definition the job of a DAC is to convert digital to analogue so how do we know how good the chip is at interpolating the sample when mathematically proven Shannon-Whittaker requires an infinite sinc function which isn’t possible in practice. For sure we can infer from the effectiveness of the band width limiting but it isn’t something I’ve seen a figure for. A DAC chip can’t do the impossible which suggests it’s not a solved problem in the theoretical sense.

 

[antcollinet]

Sound is much more than dinamic range, Signal to noise ratio and distortion,

Sound might be more than those things.

But (and it is a big but) The way in which that sound can be changed by electronics is ONLY
Dynamic range (basically the same as SNR), Distortion, Frequency response. If a device only changes these at inaudible levels, and is compared to another device that only changes them at inaudible levels, then the two devices must sound the same.

We can trivially measure audio signals in the electrical domain to levels of accuracy far lower than our ears are able to detect when those electronic signals are converted to sound pressure waves.

 

[antcollinet]

Anyone else miss our old 'dumping-ground thread,' where all these sense-less arguments were eventually moved? This kinda stuff now seems to pollute many threads here at ASR....

Agreed @RickS any chance this thread can be merged with the appropriate "how can DACS sound different" one.

 

[voodooless]

I very much doubt that I could tell the difference between an ESS and a AKM chip but define work? For sure a DAC chip can very accurately reproduce a frequency with inaudible distortion and noise but by definition the job of a DAC is to convert digital to analogue so how do we know how good the chip is at interpolating the sample when mathematically proven Shannon-Whittaker requires an infinite sinc function which isn’t possible in practice. For sure we can infer from the effectiveness of the band width limiting but it isn’t something I’ve seen a figure for. A DAC chip can’t do the impossible which suggests it’s not a solved problem in the theoretical sense.

We’ve already been though this with you several times over now. You keep ignoring all the information you’ve been given time after time.

At this point, you’re just trolling.

 

[antcollinet]

A DAC chip can’t do the impossible which suggests it’s not a solved problem in the theoretical sense.

It's a solved problem in the audible sense for all people everywhere all of the time. The difference between the mathematical perfection of an infinite sinc function, and what modern well implemented delta sigma designs are achieving is far below audibility. It is what is measured here with every DAC review.

We’ve already been though this with you several times over now. You keep ignoring all the information you’ve been given time after time.

At this point, you’re just trolling.

Plus what he just said.

 

[TTT15]

Im merely noting the theoretical mathematics which I understand is proven and that given it’s nature we don’t appear to have a measurement figure that says how accurately a DAC implements the Shannon-Whittaker infinite mathematical conversion formula. Different DACs do take differerent approaches to this issue so there are potential for differences for which there doesn’t appear to be a figure. Of course there is then the question to what extent that is audible or preferred. I completely agree that most DACs dont sound different but the proven mathematics do suggest that there is at least the possibility for subtle differences in ideal circumstances at the margins. And of course that could only be proven with blind testing. But the maths is the maths.

 

[voodooless]

Again the nonsense we already debunked multiple times.. just stop, it’s getting really old.

 

[Mnyb]

While at it you most of our laws in science are either empirical or macroscopic approximations , the stuff ohm volta and ampere and maxwell et al did was before the discovery of the electron .. which way does current flow again with those negatively charged electrons ? And don't forget to invoke quantum mechanics , so what's really going on is something entirely different anyway ..and we cant describe something more complex than the hydrogen atom mathematically exactly due to the unsolved 3 body problem..

 

[TTT15]

I’ve not seen anywhere that Shannon-Whittaker is nonsense but open to be informed. Of course if it was a finite formula a chip could perfectly calculate it but it isn’t. I’m not even saying there is a difference in practice. I’ve tried dCS for instance and did nothing for me. I’m merely noting the proven maths as I understand it and advocating blind testing.

 

[Mnyb]

I’ve not seen anywhere that Shannon-Whittaker is nonsense

it's of course not nonsense , but all competent practical aplications ( non "perfect" ofcourse ) seems to do the job just fine and have done soo for decades .

More blind testing is a waste of time as it does not settle this as it was solved decades ago .

If any electrical audio component exhibits flat FR in the audible range almost no noise and almost no distorsion you have tested them all by just blind testing one of them , it does not even have to be a DAC

 

[Roland68]

I’ve not seen anywhere that Shannon-Whittaker is nonsense but open to be informed. Of course if it was a finite formula a chip could perfectly calculate it but it isn’t. I’m not even saying there is a difference in practice. I’ve tried dCS for instance and did nothing for me. I’m merely noting the proven maths as I understand it and advocating blind testing.

May I ask which dCS you tried?
And what did you compare it to?

 

[Beave]

TTT15's mind is really gonna be blown when he realizes that resistors, inductors, and capacitors also are not ideal mathematically perfect devices.

Yet that doesn't keep them from being implemented in working circuitry.

 

[antcollinet]

we don’t appear to have a measurement figure that says how accurately a DAC implements the Shannon-Whittaker infinite mathematical conversion formula.

Yes we do. THD. And frequency response.

Now will you stop?

 

[DSJR]

I think some people need to realise how BAD even a young person's hearing actually is in the animal world scheme of things. I was taught many years back that human brain processing is far more visually orientated and after all these years 'doing it,' I now firmly believe that the hoibby side of domestic audio reproduction has a lot to do with 'eye-fi' rather than 'HiFi Sound.' Why do you think all these expensive boxes exist which sell, yet often have little inside of any real value?

Dacs are pretty much a done deal these days I feel and really, with the best known and respected makes, could quite easily be purchased on facilities, looks, price and so on. The chip inside can be largely irrelevant these days, as 'wrinkles' in ESS implementations have seemingly been sorted out in most usage cases.

 

[VientoB]

I believe out brains can focus more on sound when they are not having to process information from other sources at the same time. So sitting in a dark room will allow us to focus on what we are hearing better.