“…incendiary…”I was going to accuse Ken of derailing the thread with this particularly common incendiary device, but then realised that this thread was never really on rails to start with
usually when a thread swerves this way it has limited time to live but my concern over the whole black and white and nearly everyone is deluded by their mantra strawman is when it gets rolled out on a thread about DACs and is in fact talking about whole system performance (and has been pointed out by several that the system used may be flawed for the purposes of the test).
But I would love to see some examples of systems that are 'essentially perfect' but are uninteresting. Usually when you pull at the loose threads of comments like that you find engineering answers to the questions posed.
Then again I would note that I don't give much credence to the close in phase noise audibility argument. It's of course great fun as a DIY'er with the right tools to chase things down that far and that is a laudable goal, but I have never seen a coherent argument as to why it would be audible in a complete system.
Really, Bill? Seems a bit of an overstatement.
The measurement techniques have in fact improved lately and are sufficient to describe the natural sound so that "The ear/brain being the reactive/emotional mind" is in a state of content - but the current set of measurements used by manufacturers is not inclusive of them; I refer mainly to phase noise at 0.01Hz (or at least 0.1Hz) measurement.
We are constantly paying attention to an isolated area of interest (how is a particular DAC chip implemented on a PCB for example, or what areas of the current fabrication of a particular DAC are inferior) and forgetting that there are so many other elements of great importance, that could eventually help present the harmonics of the instruments in a way that we humans deem to be natural and pleasing. The spec sheet in isolation does not capture this, of course. But that does not mean that we should overlook the importance of good specification, on contrary.
What I stated here:
... was aimed at Mark4 who stated the following:
... which can not be right, in my opinion. Good solid engineering and the aim for the best specifications should always be priority No1.
What you said and what I tried to explain here are pretty much the same... I think...
The biggest issue here is that we get drawn into a discussion about a new design and something phenomenal coming our way (which I do encourage 100%; makes me wanna replace my May DAC in a heartbeat if it is really that good).... but are left with no specifications of any kind as a proof (maybe I missed it....???), which I find very frustrating indeed.
Do you use digital volume control or is the volume always at 100 %? How does your DAC do on this test https://www.diyaudio.com/community/...quality-between-some-dacs.386815/post-7040483 with the digital volume at maximum?
Thanks again, I'll try these out. To me this issue seems more of a recording problem so fixing it in DACs may not be first in manufacturers' backlog. It also reminds me of the infamous Telarc LP recording of Tsaikovsky's 1812 overture in which the cannon shot caused mistracking with several record players. I've seen a microscope image that revealed that the groove actually had an almost backward bend at the cannon shot causing the mistracking.
Not really if you look at the number of threads over the years on here where a statement similar to yours has come along and things have spiralled down into a flame war between a number of camps. I wasn't inferring that you meant it that way, just the way of the internet.
My understanding is that its what the client wants (artist and record company), which is to say loud. Also, they say it sometimes sounds better to clip than to compress hard and or to use a peak limiter. The various options are tried at mastering and a sound treatment is chosen. Therefore, the artist probably wants to end user to play it the same way it was heard in the mastering room, with intersample overs intact.
Quote from an interesting article:
"All empirical research is opportunistic – at least to some degree. We tend to focus on topics for which data and methods are readily available. There is a widely employed metaphor used to describe research that focuses on accessible topics to the exclusion of other important avenues of research, suggesting that you are searching for your car keys under the streetlight."
https://www.sciencedirect.com/science/article/pii/S0306919220301585
Easy to see the point when someone is talking about climate impacting agriculture, rather than audio design.
Seems to me Ken has raised an interesting point. Most of the discussion over the past couple of pages appears to be closer to focusing on how to get closer to making a system 'B.'
Please don't get me wrong, IMHO a lot of engineering work is needed to get a good result either way, system 'A' or system 'B.' The interesting thing to me about working towards system 'B' is that I see a some parallels to the article linked above and the story about looking under a streetlight. For audio designers the streetlight may by an APx555, a fancy scope, and or some other cool instrumentation. Also included would be tests for things like intersample overs.
Again please don't get me wrong, the above tools all have their necessary uses, but we might ask ourselves how the linked article helps illuminate how engineers may tend to think about audio design, at least in significant part. Is accurate sound really to be found under the streetlight where distortion and noise are so easy to visualize? Or, maybe improved dynamic imaging, reverb tail decay times, and other things not so handily measured remain to be found outside of where the light is best?
"All empirical research is opportunistic – at least to some degree. We tend to focus on topics for which data and methods are readily available. There is a widely employed metaphor used to describe research that focuses on accessible topics to the exclusion of other important avenues of research, suggesting that you are searching for your car keys under the streetlight."
https://www.sciencedirect.com/science/article/pii/S0306919220301585
Easy to see the point when someone is talking about climate impacting agriculture, rather than audio design.
Seems to me Ken has raised an interesting point. Most of the discussion over the past couple of pages appears to be closer to focusing on how to get closer to making a system 'B.'
Please don't get me wrong, IMHO a lot of engineering work is needed to get a good result either way, system 'A' or system 'B.' The interesting thing to me about working towards system 'B' is that I see a some parallels to the article linked above and the story about looking under a streetlight. For audio designers the streetlight may by an APx555, a fancy scope, and or some other cool instrumentation. Also included would be tests for things like intersample overs.
Again please don't get me wrong, the above tools all have their necessary uses, but we might ask ourselves how the linked article helps illuminate how engineers may tend to think about audio design, at least in significant part. Is accurate sound really to be found under the streetlight where distortion and noise are so easy to visualize? Or, maybe improved dynamic imaging, reverb tail decay times, and other things not so handily measured remain to be found outside of where the light is best?
Last edited:
Gave away the Benchmark DAC-3 to my daughter and kept the D90. That latter easily sounds better. Better bass, better imaging, wider soundstage, and it sounds less distorted than DAC-3 (whether what is heard its actually distortion, correlated noise, or something else).
Are you surprised though?
also, to be clear, are you suggesting that alone the phase noise is that important, or just that if phase noise is that low across the bandwidth, then by extension distortion and noise must also be low?
And yet you used to crow how the DAC-2 and 3 are used in studios all over the world. Funny how the best thing ever suddenly becomes a hand me down. I do struggle to believe digital audio is evolving that quickly.
Why does the blame for supposed audible differences have to be put on obscure low-level effects when half the DACs are driven into clipping and no effort has been put into controlling the test in the first place? That's searching for car keys in the dark without first checking if yours are among the dozen car keys that happen to lie under the streetlight.
- Status
- Not open for further replies.