What audio system do you have, or plan on getting?

[drogulus]

So a new marketing trick had to be devised; stick in a silly looking transformer/power supply and revert to a less accurate, more expensive, superseded technology (R2R and NOS designs) but market it as "better", "more musical", "more refined", "more analog" and a host of other nonsense claims. They did look different though, if you open the case, they look more like the DAC sections of CD players back in the early days of digital audio that no one uses anymore. A bit like opening a car bonnet and seeing a bunch of gold plated Weber downdraft carburettors and thinking it looks different, more expensive and performs better than modern fuel injected turbo engines.

    Over at ASR there's some discussion of the Fiio R2R DACs. I wonder about these products because though they are inferior measurement wise they are inexpensive as these devices go.

[Daverz]

After reading through many more posts in this thread, it's obvious that at least three of you need serious psychological help. The amount of idiocy and vitriol being spewed forth is utterly incomprehensible. I'm sorry that I ever made an account, and I'm correcting that error immediately.

This is just a thread on a music forum.  It's not an audiophile safe space.  There are plenty of forums where audiophiles are protected from any kind of skeptical vibes (What's Best, Audiogon, Audiophile Style, Steve Hoffman, Audiocircle, etc.)

(Edited to remove judgmental adjectives.) 

[StudioGuy]

Over at ASR there's some discussion of the Fiio R2R DACs. I wonder about these products because though they are inferior measurement wise they are inexpensive as these devices go.

The audiophile DAC market is really quite absurd these days. Essentially, the process of converting from digital to analogue was perfected (beyond the limits of human audibility) just a few years after digital audio was released to consumers, by the use of delta-sigma oversampling. By the mid 1990's that level of performance started to become available even in cheap converters and by the 2000's it was difficult/near impossible to find any DAC that didn't achieve that. Of course though, we're talking about digital technology so the only two options were either a race to the lowest price possible or to just to keep creating "upgrades"; more features/options and levels of fidelity ever further beyond audibility. So on the one hand we have something like the Apple Dongle, which costs just $9 and is audibly transparent/perfect (actually well beyond thresholds of hearing) and on the other we have something like the Topping D10B which has artefacts (distortion/noise) so tiny they can't even exist as sound, let alone be audible. And this absurd level of fidelity (which cannot be fully resolved into sound without breaking the laws of physics) still only costs around $110!

As mentioned, some audiophile DAC manufacturers, around 15-20 years ago, decided to revert to the superceeded technology (R2R instead of ds oversampling) purely for marketing purposes. Performance was terrible but falsely marketed as "more musical", etc., and because the R2R technology was intrinsically more expensive to implement, they had an apparently justifiable reason to charge silly money. Many audiophiles tend to trust audiophile marketing, especially if it's presented by a reviewer they like, so the manufacturers succeeded in creating an R2R DAC market but of course, that didn't end competition, it just shifted it to an inferior technology/topology. So now, 15-20 years later, we have a bunch of different R2R DACs; at one end we have DACs that perform similarly to the Topping D10B (EG. Such high fidelity it cannot be resolved into sound) but cost around $3-4k rather than ~$110 (the Holo Spring 3 for example) and at the other, we have relatively budget R2R DACs for those with less disposable income who still want to jump on the R2R bandwagon. For example the FiiO DAC, which at around $170 is obviously only a fraction of the cost of say the Holo Spring 3 but has relatively atrocious performance. Significantly poorer than an Apple Dongle, hugely poorer than say the Topping and around the same performance as a cheap delta-sigma converter from the mid 1990's.

What is so commonly omitted in the audiophile world is what audible difference these relatively massive differences in performance actually make. In the case of say the FiiO, even though it has atrocious performance compared to modern DACs, the difference is still only just within the thresholds of audibility and therefore probably only audibly different to a minority of consumers under fairly extreme listening conditions. 

[71 dB]

It's not an audiophile safe space. 

To me it is crazy to even expect to have safe spaces online. Why should a group of people with certain opinions enjoy the protection of safe spaces when everybody else don't?

[DavidUK]

It's very annoying chip manufacturers started providing user selectable filters around a decade or so ago. Prior to that DAC chips just applied a single optimal filter, job done. Unfortunately, the audiophile industry started to apply their own (non-optimal) filters and the never ending cycle of providing upgraded features eventually resulted in the chip manufacturers adding various built-in filter options to cater to audiophile manufacturers. This is annoying because a DAC manufacturer can name the filters anything they want, don't have to explain exactly what they do and sometimes it can be difficult to identify which is the original/standard optimal filter.

On the fortunate side, the original/optimal filter type is typically the default filter and, the differences between many/most of the filter choices are inaudible anyway but sometimes they're not all inaudible and the optimal filter is not always the default.

The optimal type is a fast roll-off (starting around 20kHz), linear phase filter.
The slow roll-off type will typically be inaudible but it depends of course where the roll-off starts. Often they start at 16kHz and will therefore be inaudible (to adults) but sometimes they start at 10kHz or even lower, which maybe audible.
Minimum phase is not ideal, you either have to have a slow roll-off or relatively large phase shifts, typically this is not audible but again depends on where the roll-off starts.
Apodising filters are typically the same as a minimum phase with a slightly lower than optimal roll-off and some phase shift. Almost never an audible difference with an apodising filter though.
NOS, "ideal transient" or some other name that indicates no filter ringing in response to a Dirac Pulse (which don't exist in music) are typically the worst filter type, little/no filtering above the Nyquist frequency (22.05kHz in the case of CD) and a roll-off starting at around 2kHz, which can/will be audible.

The Audiolab 8300 manual is not very clear and makes various incorrect assertions. The optimal (fast roll-off, linear phase) filter appears to be the one they call the "Optimal Spectrum" filter but it's not clear if that's the default option. If highest fidelity is what you consider "best" (as most would) then you should choose this filter option. However, if you prefer a little less high treble than your classical recordings actually contain, the slow roll-off option might be "best" for you (depending on your age/hearing and where the roll-off actually starts).

That's great. Many thanks for the response.

[Daverz]

As I recall it, the pre-ringing of linear-phase anti-aliasing filters was pounced on as an issue, so manufacturers starting using "slow" minimum phase filters in their DACs, even though the pre-ringing is not audible and the slow filters allow more aliasing into the potentially audible range.  Archimago had a post on this:

https://archimago.blogspot.com/2018/01/audiophile-myth-260-detestable-digital.html

It's a case where measurements -- like those impulse response measurements of DACanti-aliasing filters you see in Stereophile -- mattered to audiophiles because they could misinterpret their significance to create an exclusive marketing point.  Now every Chinese DAC has a set of selectable filters.

[Kalevala]

My streamer quit today. No audio output anymore, even after rebooting it. It had a few other issues before then, warning me that its days were numbered. Since the software has gone dramatically downhill since I bought that Bluesound Node, I'm jumping to another brand AGAIN. I've ordered a WiiM Ultra. I just hope it has Spotify Connect, which has become essential for me.

A couple of things that I've been wondering about streamers is how well and long do they last regarding updates and how hard are they to update?  I am tempted!

After reading through many more posts in this thread, it's obvious that at least three of you need serious psychological help. The amount of idiocy and vitriol being spewed forth is utterly incomprehensible. I'm sorry that I ever made an account, and I'm correcting that error immediately.

I'm so sorry that you are upset and decided to leave; I was looking forward to reading more of your posts--particularly about music.   

K

[DavidW]

A couple of things that I've been wondering about streamers is how well and long do they last regarding updates and how hard are they to update?  I am tempted!

I've used a couple of streamers for years without either one reaching end of life on updates. All three streamers I've owned were dead simple to update. The app (you need to download and use an app on either a phone or tablet to control the streamer) tells me when it is time to update, and I click and wait a few minutes while the device downloads the firmware, installs it, and reboots.

[Brian]

My aunt just told us a funny story. She has finally replaced her 40-year-old stereo, including a 40-year-old CD player (!). She says the new gear plays CDs slower. As in, the music all sounds slower. Is it possible that 40-year-old equipment would perhaps become uncalibrated and be playing music at the wrong speed?

[71 dB]

My aunt just told us a funny story. She has finally replaced her 40-year-old stereo, including a 40-year-old CD player (!). She says the new gear plays CDs slower. As in, the music all sounds slower. Is it possible that 40-year-old equipment would perhaps become uncalibrated and be playing music at the wrong speed?

Should not happen. I think the new gear just "feels" different and appears slower for that reason.

[Daverz]

My aunt just told us a funny story. She has finally replaced her 40-year-old stereo, including a 40-year-old CD player (!). She says the new gear plays CDs slower. As in, the music all sounds slower. Is it possible that 40-year-old equipment would perhaps become uncalibrated and be playing music at the wrong speed?

It's not common, but could the old CD player have had a speed adjust?

I suppose you could time a track with your phone just to make sure.

You say she replaced her stereo, so the new gear may be exciting bass frequencies differently in the room, and a room bass mode with a lot of reverberation can make music sound "slow" by dulling rhythmic attacks.  Listen for things like kick drum hits that are too "fat" and reverberate too much.  Pulling the speakers further out into the room may help.

[Spotted Horses]

My aunt just told us a funny story. She has finally replaced her 40-year-old stereo, including a 40-year-old CD player (!). She says the new gear plays CDs slower. As in, the music all sounds slower. Is it possible that 40-year-old equipment would perhaps become uncalibrated and be playing music at the wrong speed?

The timing is probably determined by a quartz oscillator, which should be very stable over long times, with drift measured in parts per million per year. It is hard to imagine a noticeable change over 40 years. Is the old CD player still around? Just play a track and time it with a stopwatch to see if there is a discrepancy between the time readout and real time.

[Fëanor]

My aunt just told us a funny story. She has finally replaced her 40-year-old stereo, including a 40-year-old CD player (!). She says the new gear plays CDs slower. As in, the music all sounds slower. Is it possible that 40-year-old equipment would perhaps become uncalibrated and be playing music at the wrong speed?

I agree with others that that is certainly not the case. Though while the CDP's time won't be the problem, some older equipment might have degrade in some ways.

I'd speculate that the newer equipment is objectively better, i.e. lower in various distortions, hence sounds smoother and "easier going" to your aunt's ear.

[drogulus]

   
     There is your traitor, there is your terrorist!

[Daverz]

There is your traitor, there is your terrorist!

That one of his other videos being promoted is "The Nyquist Theorem is a lie" does not fill me with confidence.

[drogulus]

That one of his other videos being promoted is "The Nyquist Theorem is a lie" does not fill me with confidence.

     I don't know that one. It might be a case like all the vids that are blurbed "They are lying to you" where a legit point is made but no actual lies are involved.

     Also, the actual title was "The "Nyquist theorem" isn't what you were taught (why digital used to suck)". Putting a more clickbaity and misleading blurb on the thumbnail is par for the course.

[71 dB]

Putting a more clickbaity and misleading blurb on the thumbnail is par for the course.

Everyone should assume 100 % of Youtube videos have clickbaity titles/thumbnails.

[DavidW]

Also, the actual title was "The "Nyquist theorem" isn't what you were taught (why digital used to suck)". Putting a more clickbaity and misleading blurb on the thumbnail is par for the course.

Ah, yes... I don't even have to watch the video. I know what misinformation he is debunking. People misperceive the discretization of audio like this:



Mostly due to damned lies from the industry itself, case in point, this is taken from Cambridge Audio, and I also had Sony as a choice, and Sound and Vision. Boy, would they be surprised to learn that a 1 kHz sine wave is perfectly captured.

Edit: or this one from Qobuz

[Todd]

This graph appears to be from someone trained in marketing rather than engineering. 

We're what, more than a half century out from Denon's first digital recordings, and more than a century out from earliest mathematical work in the field.  My understanding is that work was being pursued before the Great War at Western Electric (ie, AT&T), so well before Nyquist - also at AT&T, surprisingly enough, or not - published his work in 1928.  I have a sneaking suspicion people doing work in the field probably knew what they were doing.  Their employer demanded it.

I've listened to 70s era Denon digital recordings, and they sound fine, if perhaps not up to the best recordings of the intervening half century.  I don't have any early 80s CD transfers that supposedly sound awful, though my understanding is that there are some that sound awful, and that was due to poor mastering.  CD players added oversampling pretty early on, eliminating the bad sound reproduction in the mid-80s or so.  My first CD player, ca 1990, boasted 16x oversampling (if memory serves) and it sounded pretty good.  It would telling to do A/B comparisons with CDs using modern tech and early 80s tech to hear how bad early units sounded.  Otherwise, all comparisons rely on aged memories and anecdotes and are therefore subject to all manner of biases and exaggerations.

[71 dB]

Ah, yes... I don't even have to watch the video. I know what misinformation he is debunking. People misperceive the discretization of audio like this:

Stupid picture.

1) As if analog sound was 100 % perfect. No distortion? No noise...? It is not.
2) Those "staircases" DO NOT EXIST AS SOUND! They are simply a graphical way to plot the data.
3) Those "staircases" would be "microscopic" for the ear* if they actually existed.
4) Apparently these people have never heard of dithering.

* Just as we can't see microscopic objects with our bare eyes, we can't hear the quantization error in 16 bit or more audio. These signals shown in these picture would be very quiet. It would be hard to hear the sounds let alone quantization noise. Most of the time signals are about 100-1000 times bigger, but then even 16 bit digital audio would "look" fine on paper. That's why they "zoom in" a lot, but if you zoom in this much with analog sound there should be even more noise than signal!

Mostly due to damned lies from the industry itself, case in point, this is taken from Cambridge Audio, and I also had Sony as a choice, and Sound and Vision. Boy, would they be surprised to learn that a 1 kHz sine wave is perfectly captured.

Edit: or this one from Qobuz

More stupid graphs... ...that's a sine wave. If you compress it into mp3 at 320 kbps, the algorithm would easily allocate enough bits for the signal at ONE FREQUENCY. The staircases wouldn't look this bad and those staircases are just a way to show the date graphically. After the reconstruction filtering in DAC the signal is as smooth as it should be.