breakthrough of the budget priced fully transparent DAC - when did it happen?

[Robin L]

How much of this is all in the mind anyway? Right now the chain in my audio system starts with a Topping E30. Subjectively, I'm hearing more than I recall from previous gear. How different is the sound of the Drop 6XX powered by the Topping L30 compared to the Stax Earspeakers powered by an OTL tube energizer/amp I used for the better part of 20 years? Which one told me the truth? I'm older, seems like my sense of activity in the treble range is diminished, or is it that the Stax that was my daily driver has exaggerated highs? My sense is that I'm hearing more individual musical lines with the Drop/Topping combo, though I'm not getting the illusion of "tactile proximity" I got from the Stax' projection of upper mids. How much is anticipating these sorts of improvements from reading a review? How much is it due to playing more music and becoming more aware of harmonic intervals and suchlike? How much is my mood, how much the time of day [listening to Van Morrison around 2:30 am right now, sounding better than I can previously recall], how much is the chemicals running through my body? "Fully transparent" to me means the illusion of the performers in the room with me. But they aren't, and the few times in the past that they were, the sound would be wonky in some other way. So maybe I can't get there from here, due to being familiar enough with the sound of a decent acoustic guitar to know I'm not getting it from any electronic simulacrum of the same. So maybe we're not fully transparent yet.

 

[TheBatsEar]

There is no DAC/system able to reproduce a remotely clean signal with 170dB dB dynamic

I think this is a good thing.
And i think your argument is absurd, because it's basically this: Unless your system can't duplicate the dynamics of an exploding atomic bomb, it's not transparent.

 

[sergeauckland]

F

Fully or good enough for most people's hearing is not the same for me.

Good enough for most people's hearing with normal music at "normal" listening level.
We have this since 20 years or so.

Objectively better in any way then the even the best golden Ears...
We are not even close in therms of reproducing real life dynamic
For example A balloon pop can be >160dB and
And for some even a sound at -10dB can be detected.
There is no DAC/system able to reproduce a remotely clean signal with 170dB dB dynamic
Even a hand clap peaks at >140dB SPL

And before someone comes with "A SPL level of x is unrealistic... "How mush dB(A)SPL is this at 20Hz?
The DACs output is linear with frequency the hearing is not.

1) How much music includes the popping of balloons?
2) Can these Golden Ears detect a sound of -10dB (Presumably SPL) in the presence of music programme?
3) How much music includes hand claps that haven't already been compressed at the recording level?

S.

 

[Lambda]

1) How much music includes the popping of balloons?
2) Can these Golden Ears detect a sound of -10dB (Presumably SPL) in the presence of music programme?
3) How much music includes hand claps that haven't already been compressed at the recording level?

This is not the point.

As i said normal music under normal listening conditions for average normal people. We are already fine since >20 Years
But this is not my definition of "Fully Transparent"

Fully transparent to me means Worst / best case for every test signal (or at least every thing you can realistically hear in real life) not just.
Not just compressed music but the faithful reproduction of every sound.

2) Can these Golden Ears detect a sound of -10dB (Presumably SPL) in the presence of music programme?

Also dose not matter... i can have live performance in a anechoic chamber starting with a pulsing 1-4khz sound at going from -10 to +10dB SPL and ending with a ballon pop.
If you want to call a system fully transparent it needs to be able to reproduce every life event indistinguishable from the original to every listener.

 

[freemansteve]

It's worth considering what is being recorded and listened to by most of us. And what limitations there are for recording kit and with the actual instruments and vocals when they are playing live.

Balloon pops don't do it for me, but everyone listens to different music or possibly just endless loops of sound effects.

My strat has a bandwidth of ~8kHz, is rather distorted and does not have much dynamic range or signal to noise. Nobody uses amps for their instruments that are not way noisier than anything in a HiFi system. A live drum kit or a full orchestra will cause distortion in your actual ears at full pelt.

Besides, any given DAC chip can be better or worse, depending on the system it is embedded in, i.e. FWIW, you can't just go by chip specs.

Mind you, without 'purists', who would we be able to fleece money from

 

[Lambda]

Besides, any given DAC chip can be better or worse, depending on the system it is embedded in, i.e. FWIW, you can't just go by chip specs.

Better then chips spec is hard...

Balloon pops don't do it for me, but everyone listens to different music or possibly just endless loops of sound effects.

My definition is of transparent is not base on music, Music is not the only form of audio(visual) Experience/art!
What we call music is highly Subjective.

My definition of Transparent is based on physiological limits of the human perception.
As said many times before for "normal" music in cd quality for the average listener most Smartphones(dacs) are "transparent"

But i don't think there are Fix definitions for what "transparent" or "fully transparent" means.
If your definition differs from mine that's totally fine.

 

[TurtlePaul]

This feels like moving the goalposts too far. Do you actually have a track of a -10 dB hum followed by a balloon pop that you actually listen to and can hear in your real world listening environment? I know that in my environment there is no possibility of hearing a -10 dBa noise. My computer in theory has fans running at low speed producing around +20 dB. In my room, they are effectively silent. The noise floor in my environment is rarely lower than 40 dB, understanding that some may have lower.

I feel like there is this seek for an inner truth with audio systems. There is no inner truth, it is a means to an end and that end is reproduction of recorded music and video material. That is it. There is already compromise in stereo systems because you are listening in a different room from the recorded source and with stereo reproduction not reproduction of the entire sound field. If you can't perceive the difference (via ABX) on real-world tracks, why bother?

 

[RickSanchez]

Fully transparent to me means Worst / best case for every test signal (or at least every thing you can realistically hear in real life) not just.
Not just compressed music but the faithful reproduction of every sound.

If you want to call a system fully transparent it needs to be able to reproduce every life event indistinguishable from the original to every listener.

A lot to potentially unpack there. Not going to try to address all of it, but at a minimum ...

• I think you need to go back and re-read point #3 from @sergeauckland . Your issue seems to be with microphone technology + the recording process, not with DACs.
• Your issue also seems to be with speaker design + technology, not DACs. Even the greatest omni-directional speaker in the world cannot exactly replicate the sound of balloon pop in a room.

 

[Lambda]

• I think you need to go back and re-read point #3 from @sergeauckland . Your issue seems to be with microphone technology + the recording process, not with DACs.
• Your issue also seems to be with speaker design + technology, not DACs. Even the greatest omni-directional speaker in the world cannot exactly replicate the sound of balloon pop in a room.

Sure.
Microphones, ADC, speaker, headphones area all also limiting and far form "perfect"
But the the DAC is also not quite there and this thread is specific about DACs.

It s hard to judged transparency for a synthesized sound since the reference sound never actually existed.
But using Synthesis we are not limited by Microphones an ADCs.
I also think we have headphones / in ear monitors that can reach over 130-140dB SPL peak.?
This is already hard to dive them at max level with current DAC/Amp technology and still have no audible noise and high SINAD at normal listening level.
If you want also to add some Headroom for EQ it gets even harder.

I by no means saying DACs are the limiting factor in current Hifi systems!
I think most DACs are good enough even for very high end Systems.

What matters more is how good the DAC is isolating from Noises ,interference, ground loops.
I would prefer Isolated or Quasi floating inputs and outputs with High balanced (and symmetrical) signal level over a view dB more SINAD.

From a Practical standpoint i can say get a Topping E50 and stop worrying about DACs.
But this is not Audio engineering forum

Do you actually have a track of a -10 dB hum followed by a balloon pop that you actually listen to and can hear in your real world listening environment?

Your looking at it from a practial point of view. I don't.
it is science not engineering. we can entertain hypothetical concepts.
i can create a track with this if you want to...
i would not want to listen to it! but as said before this is all not relevant for my definition.

There is no inner truth, it is a means to an end and that end is reproduction of recorded music and video material. That is it. There is already compromise in stereo systems because you are listening in a different room

Headphones?

Also Only because there is already a compromise don't mean wen have to stop all other effort to improve.
it will (maybe ) never be perfect.
but this dose not mean wen can't improve.

 

[Lambda]

I know that in my environment there is no possibility of hearing a -10 dBa noise.

I head a hearing test 1 or 2 years ago.
It was in a normal not very quite room. they played single frequency sine waves with random frequency and random left or right , slowly increased in volume in small steps.
I had to press left or right buttons as soon as i heard it.
I got as low as 6dB in the results for some frequency.

But i don't have golden ears. it was not silent room.
And i'm sure with training or changed testing i could score better.

Some times i was thinking i hear some thing but not quite sure so i waited till i was sure and then the volume already increased 1 or 2 steps.
If they would uses pulsing tones or if i could uses ABX between nothing and a known tone at low volume i'm sure i would do better.

I’m not saying -10dB with my ears but i would not surprised if some young golden ear can detect a tone at -10dBSPL with ideal conditions and training.

 

[phoenixdogfan]

Probably happened earlier than most audiophiles think. But I know the first DAC that was fully transparent, affordable and recognized as such was the Benchmark DAC1. It was closer to $1000 in 2004 than $100, but it was the one recognized universally in audiophile land as fully transparent (by both Stereophile and TAS, no less).

The DAC 1 replaced the DAC of the $3000 Audio Research CD 1 in my system. I used the Audio Research unit as the transport and sent the output of my CDs via AES/EBU into the DAC 1. I quickly realized its sound was equal or superior to the $20,000 vinyl rig I had in my listening room.

From that point on, it was game over for vinyl. When I realized I could rip all my CD's store them on a hard drive, connect the drive to a router and play everything back on my home WIFI using a Logitech Squeezebox feeding the DAC 1-- the turntables, isolation platforms, tonearms, moving coil phono cartridges, speed controllers, record cleaning machines, CD transports, and record collections (over 4000 CD's and vinyls) got sold.

It was the icing on the cake when I learned I could take the digital stream, feed it into a TACT 2000 and do room and speaker correction.

 

[TurtlePaul]

Also Only because there is already a compromise don't mean wen have to stop all other effort to improve.
it will (maybe ) never be perfect.
but this dose not mean wen can't improve.

This is the 'everything matters' fallacy. No, everything doesn't matter. Changes, even improvements, that are 20 dB below the noise floor and can't be identified in an ABX test don't matter.

 

[sergeauckland]

Probably happened earlier than most audiophiles think. But I know the first DAC that was fully transparent, affordable and recognized as such was the Benchmark DAC1. It was closer to $1000 in 2004 than $100, but it was the one recognized universally in audiophile land as fully transparent (by both Stereophile and TAS, no less).

It was recognised in ProAudio Land that the Sony PCM-F1 was transparent back in 1981. The Mitsubishi X-80 digital tape recorder before that, (1980) albeit they sampled at 50.4kHz so their recordings weren't directly compatible with CD.

S.

 

[Lambda]

This is the 'everything matters' fallacy. No, everything doesn't matter. Changes, even improvements, that are 20 dB below the noise floor and can't be identified in an ABX test don't matter.

I never said everything matters.
If something is below noise floor depends on noise floor and signal level both can be controlled. don’t know how you come up with 20dB...
But i sure can hear 1k sine wave 20dB below Brownian noise.

If something can’t be heard in an ABX because the rest of your system is not good enough you can argue there is no point to improve it.

This is not my point. Sure we have many other bottle necks long before DACs become a limiting factor.
But the Thread is about DACs specifically.
For any practical application they are good enough. For theoretical edge ceases (see my definition of fully transparent ) they are not.

 

[garbz]

It was recognised in ProAudio Land that the Sony PCM-F1 was transparent back in 1981

You're going to have to come up with a citation for that rather than a simple "it is so" comment.

I completely disagree with you by the way. The effects of aliasing are audible. The effects of complex anti-aliasing filters in early DACs also led to deviations in frequency response in the audio band trying to get this effect under control. While I could bite that your PCM-F1 with its higher sample rates were an issue, in the world of consumer audio the introduction of the CD was objectively *not* transparent and a lot of effort was put into resolve the audible issues introduced on multiple fronts.

Multiple fronts is the reason why Toshiba and Sony both started looking at high-res consumer formats, and the reason why engineers started looking at things such as oversampling to deal with aliasing in another way.

And then you get into the discussion of what is a DAC. The way it's used on this forum seems to imply external, and it was 1988/89 before we got to actually dealing with the audibility of jitter caused by the interface. Reminder: People didn't throw engineering terms in and say it had an effect, people heard an effect and then went to look for what caused audible problems. We know the thresholds for audible jitter are in the ns regime (far higher than even worst cases for equipment today) but with the introduction of the first DACs in the 80s jitter *was* that high.

Digital audio was definitely not *always* transparent.

 

[Robin L]

You're going to have to come up with a citation for that rather than a simple "it is so" comment.

I completely disagree with you by the way. The effects of aliasing are audible. The effects of complex anti-aliasing filters in early DACs also led to deviations in frequency response in the audio band trying to get this effect under control. While I could bite that your PCM-F1 with its higher sample rates were an issue, in the world of consumer audio the introduction of the CD was objectively *not* transparent and a lot of effort was put into resolve the audible issues introduced on multiple fronts.

Multiple fronts is the reason why Toshiba and Sony both started looking at high-res consumer formats, and the reason why engineers started looking at things such as oversampling to deal with aliasing in another way.

And then you get into the discussion of what is a DAC. The way it's used on this forum seems to imply external, and it was 1988/89 before we got to actually dealing with the audibility of jitter caused by the interface. Reminder: People didn't throw engineering terms in and say it had an effect, people heard an effect and then went to look for what caused audible problems. We know the thresholds for audible jitter are in the ns regime (far higher than even worst cases for equipment today) but with the introduction of the first DACs in the 80s jitter *was* that high.

Digital audio was definitely not *always* transparent.

I do believe the citation belongs to J Gordon Holt.

Sony PCM F1 Digital Audio Converter

In 1978, when I reviewed Sony's first audiophile-type PCM-1 converter, I earned the undying scorn of a large segment of audiophilia by reporting that, on the basis of a rather short testing period (which did however include some live recording), I was unable to hear anything the matter with its...

www.stereophile.com

As for absolutely transparent, I used an early Sony ADC/DAC [501] strapped to a Betamax way back around the dawn [1988] of digital, it wasn't 100% transparent. That PCM-F1 didn't have higher sampling rates, defaulted to 44.1k though it was also capable of 48K as well if I remember correctly. But compared to the various ways my Tascam 32 1/2 track reel to reel deck mangled the signal at 15 ips, it was mana from heaven. These things are relative. By 1998, most of what you're bitching about became non- issues, digital recording was more or less standard, jitter became a known variable that could be dealt with and nobody could afford reel to reel tape by then anyway. Those first LPs from 1949 were kinda wonky too, don't you know?

 

[garbz]

By 1998, most of what you're bitching about became non- issues

They largely became a known solvable engineering problems in 88/89. But there's a difference between being able to build a transparent DAC and everyone building only transparent DACs. A few turds no doubt came out in the decade that followed.

Not sure I like your use of the word "bitching". That implies I'm a whiner rather than the industry recognised a real problem and sought real solutions to address it. I completely disagree with the OP. DACs not only were not always transparent (internal or external), they were largely to blame for digital audio actually having a pretty crap reputation on first release. People these days saying "it's harsher" are dellusional. People in 1985 saying "it's harsher" were exclaiming about a serious well known, but at the time not well understood problem.

Those first LPs from 1949 were kinda wonky too, don't you know?

Yes, that was my point.

 

[Chrispy]

Bigger question to me is why did people think this was a recent event? Transparent dacs have been around for a long time, unless you buy into some of the stranger audiophilia.

 

[TheBatsEar]

Bigger question to me is why did people think this was a recent event? Transparent dacs have been around for a long time,

And OP asked for how long. I suspect he expected more than "a long time".

 

[Robin L]

They largely became a known solvable engineering problems in 88/89. But there's a difference between being able to build a transparent DAC and everyone building only transparent DACs. A few turds no doubt came out in the decade that followed.

Not sure I like your use of the word "bitching". That implies I'm a whiner rather than the industry recognised a real problem and sought real solutions to address it. I completely disagree with the OP. DACs not only were not always transparent (internal or external), they were largely to blame for digital audio actually having a pretty crap reputation on first release. People these days saying "it's harsher" are dellusional. People in 1985 saying "it's harsher" were exclaiming about a serious well known, but at the time not well understood problem.


Yes, that was my point.

Sorry if I pointed my complaint in your direction. I heard problems with digital replay in 1989, probably the first full year of regularly listening to CDs. My experience with an early ADC/DAC chain had audible problems, but not as great as the issues I had with the Tascam reel-to-reel machine. However, I had the worst of both worlds at the time, transferring LPs to the Tascam [so there would be proper pauses between tracks], then transferred to a digital copy on Betamax. Terrible sound, but that signal was reamed further by being transmitted via satellite. Solutions may have been at hand in 1989, but I really didn't encounter solutions to these problems until some years later. I suspect that Didital was harsher to a lot of people who had a rolled off front end via a turntable with a cartridge with the top rolled off. Also, the constant pre/post echo of LPs could also affect the sound. I don't know if the tc electronics M2000 [digital effects processor, 20-bit capable ADCs and DACs] I got in the mid 1990s was "fully transparent", but it was better than anything else I was using in 1996.

BTW, some of the early problems must have come from experienced engineers who were compensating for analog tape machines that had soft clipping and prematurely rolled off the upper octaves. Doing the same with a digital recorder will result in an excessively bright recording.