Filterless DACs question

[FunkyBassMan]

OK, thanks for all the answers . I have to say that these differences are likely not noticeable enough for me to worry about it, or spend enough time on it to make me feel worried.

My CD player has switchable digital filter characteristics, and frankly, when I tried them out I couldn't hear any difference. That's enough for me to just leave it and not get too much side-tracked by the equipment; I have too little time already to discover all the exciting new music I want to discover
(but I appreciate others may have different objectives)

Yeah, I totally hear ya on that! At the end of teh day, those subtle filter differences can easily get lost when ur just listening casually. In fact, that's kind of the point haha! Totally agree on your approach: trust your ears, don’t sweat the small stuff, and just enjoy the music. Life’s short and there's alwys more great music waiting to be discovered..

 

[gregorio]

It is this disconnect in timescales at which the pre-ringing happens, vs. the timescales of the claimed audible effects that I struggle with...

TBH, it’s not only the disconnect with timescales, there’s also a disconnect with the occurrence and magnitude, as well as the disconnect that even when it does occur it’s above the hearing range, as VNandor stated.

Also, you're dead wrong about ringing being totally inaudible.

Oh good, every single time you state I’m “dead wrong” it’s demonstrated you’re the one who’s dead wrong, so this will be easy!

w/ real-world audio content the ringing is typically far less pronounced because there's usually much less energy near that cutoff frequency.
But thats not to o dismiss ringing entirely.

That is exactly why it should be dismissed entirely! How is “much less energy” which is “near that cutoff frequency” of ~22kHz not entirely inaudible? Just being near 22kHz makes it inaudible, let alone the fact that it’s also extremely low level and of extremely short duration! You’ve then just simply invented a bunch of utter BS to try and get around this self-evident impossibility, such as:

[1) steep linear-phase filters can still subtly influence audio perception - not necessarily due to audible "ringing" in the conventional sense (like echoes) …
[2] The key to it is that your brain doesnt perceive these tiny microsecond events separately—it combines everthing into one overall impression.
[3] At the end of teh day, those subtle filter differences can easily get lost when ur just listening casually.

1. Sure in the “unconventional sense” of some inexplicable magic that’s contrary to the science. Although what you go on to describe; “slightly impacting the perceived transient clarity or spatial imaging” is in fact ringing in the conventional sense, so just nonsense!
2. The “key to it is that” you’re just making up complete BS! The only things the brain combines into one overall impression is biases and sounds that are audible.
3. They get lost because they don’t often exist to start with and even when they do exist they get lost with any kind of listening because the ringing is extremely short, extremely low level, in the ultrasonic range and therefore inaudible!

Do you have any reliable evidence to support your claims or are you just going to deflect and demonstrate it’s all made-up BS?

Has enough testing or any at all been done on the audibility of this specific ringing?

Quite a bit of research has been done that includes this specific effect. A couple of particularly compelling papers were published a few years ago:
High-frequency sound components of high-resolution audio are not detected in auditory sensory memory - Nittono 2020.
Auditory brainstem responses to high-resolution audio sounds: Effects of anti-alias filters - Owhan & Nittono 2021.

What is so particularly compelling about them is that they didn’t just perform a standard DBT but actually measured the entire auditory pathway from cochlear through to and including the auditory cortex and also, any subconscious behavioural responses. Basically, the claimed “temporal blur”, ringing artefacts and removed high freqs did not elicit any Auditory Brainstem Response, did not register in the Auditory Cortex’s Sensory Memory, elicited no subconscious behavioural response and unsurprisingly, the subjects could not differentiate. Obviously, if it’s not even registering in the auditory cortex then there is literally nothing to hear.

This appears to pretty conclusively support the logical, even self-evident conclusion that we can’t hear extremely short, extremely low level artefacts that are above the freq range of human hearing.

G

 

[FunkyBassMan]

TBH, it’s not only the disconnect with timescales, there’s also a disconnect with the occurrence and magnitude, as well as the disconnect that even when it does occur it’s above the hearing range, as VNandor stated.

Oh good, every single time you state I’m “dead wrong” it’s demonstrated you’re the one who’s dead wrong, so this will be easy!

That is exactly why it should be dismissed entirely! How is “much less energy” which is “near that cutoff frequency” of ~22kHz not entirely inaudible? Just being near 22kHz makes it inaudible, let alone the fact that it’s also extremely low level and of extremely short duration! You’ve then just simply invented a bunch of utter BS to try and get around this self-evident impossibility, such as:

1. Sure in the “unconventional sense” of some inexplicable magic that’s contrary to the science. Although what you go on to describe; “slightly impacting the perceived transient clarity or spatial imaging” is in fact ringing in the conventional sense, so just nonsense!
2. The “key to it is that” you’re just making up complete BS! The only things the brain combines into one overall impression is biases and sounds that are audible.
3. They get lost because they don’t often exist to start with and even when they do exist they get lost with any kind of listening because the ringing is extremely short, extremely low level, in the ultrasonic range and therefore inaudible!

Do you have any reliable evidence to support your claims or are you just going to deflect and demonstrate it’s all made-up BS?


Quite a bit of research has been done that includes this specific effect. A couple of particularly compelling papers were published a few years ago:
High-frequency sound components of high-resolution audio are not detected in auditory sensory memory - Nittono 2020.
Auditory brainstem responses to high-resolution audio sounds: Effects of anti-alias filters - Owhan & Nittono 2021.

What is so particularly compelling about them is that they didn’t just perform a standard DBT but actually measured the entire auditory pathway from cochlear through to and including the auditory cortex and also, any subconscious behavioural responses. Basically, the claimed “temporal blur”, ringing artefacts and removed high freqs did not elicit any Auditory Brainstem Response, did not register in the Auditory Cortex’s Sensory Memory, elicited no subconscious behavioural response and unsurprisingly, the subjects could not differentiate. Obviously, if it’s not even registering in the auditory cortex then there is literally nothing to hear.

This appears to pretty conclusively support the logical, even self-evident conclusion that we can’t hear extremely short, extremely low level artefacts that are above the freq range of human hearing.

G

Lol you're way off here. no one's claiming the ringing at ~22kHz itself is directly audible as a tone or echo-- obviously that's not happening. But your totally oversimplifying how human hearing actually works.

the ringing itself might be ultrasonic and super short, but you're ignoring well-documented auditory masking and temporal masking effects (AES studies, Moore 2012, Wright & Zhang 2009). These subtly impact the perception of transients and imaging—your brain absolutely integrates short-duration auditory info into an overall sound impression. This isn't "magic," it's basic auditory neuroscience, something you've repeatedly ignored.

Also, your nittono citations are about auditory responses to ultrasonic content specifically -not about transient perception affected by linear-phase filter ringing within the audible band. You're conflating unrelated topics. No one's talking about magically hearing ultrasonic tones directly—it's about how these rapid transients at the filter boundary subtly affect the brain's perception of timing and clarity within audible transients.

So, you're oversimplifying complex auditory science, misrepresenting research, and then accusing everyone else of "BS." lol - par for your course! Maybe step back and check your sources before dismissing others as ignorant.

 

[gregorio]

Lol you're way off here.

Let’s explore that shall we:

But you’re totally oversimplifying how human hearing actually works.

Sure, human hearing doesn’t actually work with a cochlear, an auditory nerve, an auditory cortex or a brainstem response. It’s all magical audiophile anatomy!

the ringing itself might be ultrasonic and super short, but you're ignoring well-documented auditory masking and temporal masking effects

Oh lordy, hours of fun and enjoyment trying to mask something with inaudible ringing. Didn’t you even bother to look up what auditory masking is, the fundamental principles? That’s funny, you cite a paper on auditory masking without bothering to even find out what it is!

Also, your nittono citations are about auditory responses to ultrasonic content specifically -not about transient perception affected by linear-phase filter ringing within the audible band.

So despite the test signal being transients and even the title of the paper actually stating “Effects of anti-alias filters”, it had nothing to do with transients or linear phase filters. Are you well? lol

G

 

[FunkyBassMan]

Let’s explore that shall we:

Sure, human hearing doesn’t actually work with a cochlear, an auditory nerve, an auditory cortex or a brainstem response. It’s all magical audiophile anatomy!

Oh lordy, hours of fun and enjoyment trying to mask something with inaudible ringing. Didn’t you even bother to look up what auditory masking is, the fundamental principles? That’s funny, you cite a paper on auditory masking without bothering to even find out what it is!

So despite the test signal being transients and even the title of the paper actually stating “Effects of anti-alias filters”, it had nothing to do with transients or linear phase filters. Are you well? lol

G

Haha, sure—hearing is totally just cochlea and nerves and brainstem, no magic needed, right? except we're not just machines measuring freqencies and anatomy; our brains do adaptive psychoacoustic stuff, like auditory masking and timing sensitivity. Yeah, the Nittono papers were mostly about ultrasonic content, fair point, but filter ringing isn’t always just ultrasonic noise. Even subtle transient stuff can actually matter to some people's perception—belive it or not.

Saying it's completely meaningless or just magical audiophile voodoo is ignoring science. But of course that's not surprising since calling things audiophile voodoo is your singular raison d'etre here lol

 

[gregorio]

no magic needed, right?

This is a science discussion forum, so what do you think? Duh!

except we're not just machines measuring freqencies and anatomy; our brains do adaptive psychoacoustic stuff, like auditory masking and timing sensitivity.

Right, so we do auditory masking of inaudible frequencies without an auditory cortex.

but filter ringing isn’t always just ultrasonic noise.

So a filter with a transition band in the ultrasonic range is not producing ringing “always just ultrasonic noise”? What happens, does it dance around the frequency spectrum to the beat of the funky music? lol

Even subtle transient stuff can actually matter to some people's perception—belive it or not.

Err subtle transient stuff cause by anti alias/image filters cannot affect people’s perception, that was clearly demonstrated by the cited papers! You can believe the BS you yourself invented if you like but of course you’re in completely the wrong forum for that sort of belief.

Saying it's completely meaningless or just magical audiophile voodoo is ignoring science.

Is the science I’m ignoring the science of hearing without a cochlear, auditory cortex or brain stem? Is that the cute science of audiophile magic, where a brainstem and apparently even a brain is optional? You’re really going for the “making a fool of yourself” big time. Impressive to witness!

G

 

[FunkyBassMan]

This is a science discussion forum, so what do you think? Duh!

Right, so we do auditory masking of inaudible frequencies without an auditory cortex.

So a filter with a transition band in the ultrasonic range is not producing ringing “always just ultrasonic noise”? What happens, does it dance around the frequency spectrum to the beat of the funky music? lol

Err subtle transient stuff cause by anti alias/image filters cannot affect people’s perception, that was clearly demonstrated by the cited papers! You can believe the BS you yourself invented if you like but of course you’re in completely the wrong forum for that sort of belief.

Is the science I’m ignoring the science of hearing without a cochlear, auditory cortex or brain stem? Is that the cute science of audiophile magic, where a brainstem and apparently even a brain is optional? You’re really going for the “making a fool of yourself” big time. Impressive to witness!

G

wow, your brilliance is truly awe-inspiring here. Yep, auditory masking absolutely requires that the masked frequency be audible itself, right? Because everyone knows the human auditory system checks frequencies individually on a spreadsheet before deciding what to mask—duh! Nice strawman though, top-notch work there..

And yeah, clearly filters only ever producee ultrasonic ringing that stays politely out of the audible band and never interacts with anything audible. I guess signals just stay neatly where you tell them to—because frequencies never overlap or interact, they just follow strict rules and respect theoretical boundaries. Glad physics and audio processing got your memo. I'm sure actual science is taking notes on everything you say so it knows where reality is wrong.

Also congrats on completely misunderstanding transient perception and temporal masking. Those cited papers totally proved no human can ever perceive subtle transient differences caused by filters, right? I must've missed the paper titled "Transient Perception is Fake News, lol" that settled this once and for all.

..and your cute little rant about brains and audiophile magic—adorable. Everyone who disagreess with you definitely believes in magic cochleas and optional brainstems, obviously. Great argument. Maybe consider actually understanding psychoacoustics next time before confidently embarrassing yourself further.

(apologies to others for the sarcasm - I'm just trying to get through to gregorio, and arrogant sarcasm seems to be the only kind of communication he uses. Of course, the science tells us this is exactly what one should expect.
Festinger, L. (1957). A Theory of Cognitive Dissonance. Stanford University Press.

"When someone is confronted with evidence that contradicts deeply held beliefs or opinions, they experience cognitive dissonance—a mental discomfort due to holding conflicting ideas simultaneously. People usually try to reduce this discomfort by rejecting, discrediting, or even attacking the person providing the conflicting information.

If the person has been confidently expressing a belief and is suddenly shown proof they're mistaken, cognitive dissonance sets in, causing feelings of embarrassment or shame. To ease this emotional tension, they might resort to angry attacks, sarcastic remarks, or attempts to undermine the credibility of the person presenting the evidence." )

 

[gregorio]

wow, your brilliance is truly awe-inspiring here. Yep, auditory masking absolutely requires that the masked frequency be audible itself, right?

Don’t you even know what “to mask” means? How do you hide or conceal a frequency that’s already completely hidden/concealed because it’s inaudible? What planet are you on?

And yeah, clearly filters only ever producee ultrasonic ringing that stays politely out of the audible band and never interacts with anything audible.

Huh? So fast linear phase anti-image filters will produce artefacts that can’t be measured in the audible band, do not register in the brain stem or auditory cortex, cannot be discerned in a DBT, produce no subconscious behavioural responses but you can still hear them and that’s not magic?

If the person has been confidently expressing a belief and is suddenly shown proof they're mistaken, cognitive dissonance sets in, causing feelings of embarrassment or shame.

You’re walking proof that’s not the case. Either that or you actually enjoy “feelings of embarrassment or shame”. You were shown the proof and since then your argument is about masking things that are already inaudible and filter artefacts on subtle transients that don’t register in the human brain but apparently you can hear them anyway. lol. And your reliable supporting evidence is ..…. (crickets). No cognitive dissonance there whatsoever. Have you never heard of a mirror?

While it’s entertaining watching you make a fool of yourself and just keep digging that hole deeper, it does eventually get boring and even concerning that you maybe seriously unwell.

G

 

[FunkyBassMan]

Don’t you even know what “to mask” means? How do you hide or conceal a frequency that’s already completely hidden/concealed because it’s inaudible? What planet are you on?

Huh? So fast linear phase anti-image filters will produce artefacts that can’t be measured in the audible band, do not register in the brain stem or auditory cortex, cannot be discerned in a DBT, produce no subconscious behavioural responses but you can still hear them and that’s not magic?

You’re walking proof that’s not the case. Either that or you actually enjoy “feelings of embarrassment or shame”. You were shown the proof and since then your argument is about masking things that are already inaudible and filter artefacts on subtle transients that don’t register in the human brain but apparently you can hear them anyway. lol. And your reliable supporting evidence is ..…. (crickets). No cognitive dissonance there whatsoever. Have you never heard of a mirror?

While it’s entertaining watching you make a fool of yourself and just keep digging that hole deeper, it does eventually get boring and even concerning that you maybe seriously unwell.

G

you're totally oversimplifyign auditory masking and missing the point. Masking isn’t just about "hiding" one tone behind another. it’s about how sounds interact and how subtle transient changes can actually influence our perception. You keep acting like frequencies just stay neatly separate in their own little boxes. Sorry, but audio doesn't work that way in the real world with real human beings.

Linear-phase filter ringing (especially near nyquist) can and often does produce subtle time-domain artifacts. Even if these artifacts themselves measure outside the audible band as steady-state tones, the transient efects and slight timing shifts can still affect perceptoin—human hearing can be surprisingly sensitive to timing details. There's plenty of solid scientific evidence on transient perception if you're willing to actualy read up on it.

As for cognitive dissonance—if your best comeback is insults and sarcastic comments about someone's mental health instead of adressing the actual points, maybe that "mirror" you're suggesting would be useful for yourself. If you're genuinly bored watching someone dig a hole, why not step back from personal attacks and try engaging with actual science for a change? Just a thought.

 

[gregorio]

Masking isn’t just about "hiding" one tone behind another.

So that’s a “yes” then, you don’t know what “to mask” means or what auditory masking is. How many seconds does it take to look that verbal up in a dictionary and how many minutes to look up auditory masking on Wikipedia? Yet you argue before doing so!

Linear-phase filter ringing (especially near nyquist) can and often does produce subtle time-domain artifacts.

Huh, linear phase filter ringing is the “subtle time-domain artefact”, it doesn’t produce them.h

Even if these artifacts themselves measure outside the audible band as steady-state tones, the transient efects and slight timing shifts can still affect perceptoin—human hearing can be surprisingly sensitive to timing details.

Again, these transient effects and slight timing shifts are called “ringing” and anything “outside the audible band” is by definition inaudible and therefore cannot still affect perception. And, this is proven in the cited papers!

There's plenty of solid scientific evidence on transient perception if you're willing to actualy read up on it.

Yes, there are plenty of papers on transient perception, none that demonstrate ultrasonic ringing components of transients affect perception though. While we do have some papers (cited previously) demonstrating they do not and cannot affect perception.

As for cognitive dissonance—if your best comeback is insults and sarcastic comments

No, my best comeback was categorical, compelling scientific evidence that ultrasonic filter ringing and other ultrasonic content did not even register anywhere in the auditory pathway, let alone was actually audible. But as you just ignored the science, in a science discussion forum, and just kept repeating how wrong I am and that you can hear the inaudible, with no reliable evidence at all, then sarcasm seemed justified where basic logic, demonstrated fact and rational reasoning had so comprehensively and miserably failed!!

G

 

[FunkyBassMan]

So that’s a “yes” then, you don’t know what “to mask” means or what auditory masking is. How many seconds does it take to look that verbal up in a dictionary and how many minutes to look up auditory masking on Wikipedia? Yet you argue before doing so!

Huh, linear phase filter ringing is the “subtle time-domain artefact”, it doesn’t produce them.h

Again, these transient effects and slight timing shifts are called “ringing” and anything “outside the audible band” is by definition inaudible and therefore cannot still affect perception. And, this is proven in the cited papers!

Yes, there are plenty of papers on transient perception, none that demonstrate ultrasonic ringing components of transients affect perception though. While we do have some papers (cited previously) demonstrating they do not and cannot affect perception.

No, my best comeback was categorical, compelling scientific evidence that ultrasonic filter ringing and other ultrasonic content did not even register anywhere in the auditory pathway, let alone was actually audible. But as you just ignored the science, in a science discussion forum, and just kept repeating how wrong I am and that you can hear the inaudible, with no reliable evidence at all, then sarcasm seemed justified where basic logic, demonstrated fact and rational reasoning had so comprehensively and miserably failed!!

G

your entire arguement rests on a total oversimplificaton of psychoacustics...and repeating it louder wont make it true. auditory masking isnt just "tone A hides tone B," its about complx interactions between signals, timing, and perception. If you'd actually bothered to read beyond first two sentences on wikipedia you'd know that by now.

Linear-phase ringing isnt just some isolated ultrasonic phenomenon, its inherently tied to transient response IN the audible range. Those sublte timing shifts and transient details absolutely CAN (and often do) afect perception even if you personally refuse to acknowlege it. You keep insisting confidently these phenomema are "proven inaudible" but the papers you keep citing simply dont demonstrate what you say—they tested steady-state ultrasonic content, not subtle time-domain transient effects within the audible band.

Ironically your confidence is based on exactly the kind of selective reading and misunderstanding youre accusing everyone else of. if you actualy cared about "science in a science forum" you'd stop oversimplfying psychoacoustics into neat little boxes that fit your narrow viewpoint. Sarcasm isn't helping your case here either—maybe try engaging with what the science actually says, instead of what you want it to say?

 

[gregorio]

your entire arguement rests on a total oversimplificaton of psychoacustics

Absolutely! It rests on the simple argument that something that is inaudible in theory and has been demonstrated to be inaudible in practice can’t be heard. It doesn’t get any simpler than that!

Those sublte timing shifts and transient details absolutely CAN (and often do) afect perception even if you personally refuse to acknowlege it.

As there’s plenty of science that contradicts this assertion (some of which I’ve cited) and as I’m not aware of any science that supports it (and you have provided none), then logic dictates a refusal to accept it!

You keep insisting confidently these phenomema are "proven inaudible" but the papers you keep citing simply dont demonstrate what you say—they tested steady-state ultrasonic content, not subtle time-domain transient effects within the audible band.

So, you can’t even read!!
From the 2020 paper: “Moreover, the high-cut sounds examined in this study contained a temporal distortion (i.e. blurred onset and offset), which might be detected in the conventional auditory pathway. The lack of the corresponding electrocortical responses suggests that the auditory sensory memory cannot register the characteristics of a high-resolution grade sound (both high-frequency components and sharp onset and offset)” - “blurred onset and offset” refers to pre and post linear phase filter ringing.

The 2021 paper “… exposed listeners to a white-noise burst with a 50-ms duration every 500 ms.” - So that is obviously neither steady state nor solely ultrasonic content!

The parrot is dead for god’s sake! How many times in one day are you going to make a fool of yourself?

G

 

[FunkyBassMan]

Ahh...for whom does the dead parrot squawk?—it squawks for thee...

 

[bigshot]

Inaudible frequencies are inaudible. Masked sound is masked.

 

[VNandor]

Yep, auditory masking absolutely requires that the masked frequency be audible itself, right? Because everyone knows the human auditory system checks frequencies individually on a spreadsheet before deciding what to mask—duh! Nice strawman though, top-notch work there..

Let me invite you to read through the fundamentals of how masking works, and please quote the part you misunderstood that makes you think that 20kHz signals with amplitudes about -60dB and even lower existing for microseconds could mask the rest of the signal that is both sustained for orders of magnitude longer and is higher in level in orders of magnitude. Just because masking is complex it doesn't mean that anything could be masked by anything and it also doesn't mean that you get to invent your own BS about it and pass that off as how masking works.