The question is “simply” the effect of replacing a single device that measures well in a reproduction chain on human ability to reliably detect differences in an environment where they are very familiar with the sound characteristics of their system in that room or with those headphones. Period.
TBH, that isn’t “simply the question … period” (paraphrasing), it’s just one of several/numerous questions that’s been routinely rolled-out by audiophiles for many decades to try to explain the reason how/why they hear things that are actually inaudible. They’re nonsense questions because science has directly and indirectly already addressed them and moved on, typically decades before the audiophile community even thought of the question or in some cases even before there was an audiophile community. Mostly, they’re questions invented by audiophile marketers/reviewers in order to contradict or at least cast doubt on (and therefore justify dismissing) established science. AFAIK, your “question” was first raised by professional audiophile reviewers in the mid/late 1980’s, in order to cast doubt on and dismiss the growing trend of controlled DBTs/DBXs demonstrating no audible differences between even very significantly different components (with vastly different price points and even vastly different design topologies). Of course though, it was a question actually asked by science/engineers long before audiophile reviewers resurrected it to mislead consumers. Think about it; music/sound engineers have formally trained listening skills and spend 8 hours a day (or more), 5 or more days a week, working in a room (studio), doing nothing other than very carefully listening to/analysing “sound characteristics”. How many audiophiles listen to their audiophile systems 40-60 hours a week? And even if they do, much of that time is just casual listening, so who do you think can most “
reliably detect differences in an environment where they are familiar with the sound characteristics of their system in that room”? The reliable evidence, collected over 70+ years, does not support the idea/question you are repeating. There are many factors which do have an effect, such as the duration of echoic memory/fast switching, system accuracy, specific training, playback levels, etc., but familiarity of the listening environment isn’t one of them and numerous controlled tests demonstrate this. There have been many online DBTs where subjects use their own systems, many DBTs in studios where the resident engineer does no better than other trained subjects, even the occasional formal DBT run on a audiophiles’ own systems/environments and there’s no evidence to support any effect of familiarity on the ability to reliably detect differences.
I have owned at least 10 different DACs. None of them sounded the same to me in my systems, although I am quite sure they all measure near or over 100db in signal to noise ratio. I am not alone.
You’re definitely not alone. In fact, amongst those who actively listen for differences between DACs you’d be pretty much alone, or at least in a very small minority, if they had “
sounded the same”! I’ve certainly heard differences countless times and I can’t think of anyone off the top of my head that I’ve ever known personally who hasn’t. The question is why, which leads to your next point:
Three reasons or combination of reasons are responsible for this: [1] it is all in mine and others heads, [2] there is something going on with how different DACs interact with other components in the reproduction chain that results in audibly different performance, or [3] there is something going on between different design implementation of different DACs that results in audibly noticeable sound reproduction that is not captured on the bench, regardless of the systems they are deployed in.
No! Bizarrely you’ve not only missed a reason but the most obvious reason, there should be 4 in your list rather than 3. I’ll deal with all of them but start with the one you’ve missed: [4] There is something going on between different design implementation of DACs that results in noticeable sound differences that IS “captured on the bench”. For example a NOS DAC is a different design implementation and can result in noticeable sound differences which are measurable. But a far more common example is that most consumer DACs have at least somewhat different output voltages, which again is measurable and “audibly noticeable”. This last example is ultimately irrelevant of course because an amp can be used to compensate, unfortunately though, very few audiophiles compensate (volume match) accurately enough to stop this design difference being “audibly noticeable”.
[1] If #4 above has been eliminated, then this is by far the most likely reason. Being “in your head” (a perceptual error) is very common, has been demonstrated countless times, over more than a century and a half and certain aspects of audio reproduction actually relies on perceptual error, the stereo effect being the most obvious one.
[2] A DAC only interacts with two things: A. Whatever is supplying the digital data, it’s input, which has to comply with international digital protocol specifications and so will not have any audible effect unless something is broken, and B. An amp connected to the DAC’s output, which again will have no audible effect as the whole point of an amp is to amplify the voltage/signal and any competently designed amp for the task will have a far higher input impedance than any DAC’s output impedance. So again, the only possibility of an audible effect is if something is broken. There is no reliable evidence for anything else, a “C” option.
[3] Again, there’s no reliable evidence that such a thing even exists, let alone is audible. While science cannot prove a negative, there is a very substantial body of reliable evidence, going back two centuries or so, indicating such a thing does not exist. We know that sound is just air pressure variations and that these variations can be represented analogously with an electrical signal, there’s not even a hint that this knowledge is somehow wrong and that there’s something else.
So in the complete absence of any reliable evidence for your #2 and #3 reasons but a huge wealth of very solid evidence for reasons #1 and #4, we’re going to need some quite extraordinary evidence before we even consider #2 or #3 a remote possibility.
I am not a speaker guy but for example can a set of speakers change sonically due simply to the output impedance of the amplifier like the Campfire Audio IEM in my recent experience ?
Sure, even cables can make an audible difference, as given a long enough cable run, the power requirements of speakers and speakers with a low enough input impedance, then wire gauge can have an audible impact. Although in practice this would come under “user error”.
Does that potentially relate to other electronics or is there no relationship at all between output and input impedance of DACs, amps, pre-amps, speakers etc etc
It shouldn’t because the impedance between electronic components should follow the modern (50+ years) convention of audio components having a relatively low output impedance and the input having a relatively high impedance. There can be some rare exceptions however, for example some turntable outputs can have a relatively high output impedance.
G
