Is Streaming HIFI really HIFI?

So I recently decided to join a discord server with audio engineers and audiphiles however I was completely dismissed and labeled crazy essentially... My question is about whether anyone has had similar experiences to me with HIFI audio, a while ago I purchased FLAC files from sources like Qobuz, 7Digital, and Bandcamp however I have noticed that they sound better then what Apple Music and Tidal sound like, now the reason I don't believe it is a mastering difference is because I have had the exact same conclusion with a bunch of songs and I doubt they would have a special master for those who buy FLACs compared to what Tidal is offered. In terms of sound I noticed differences in presentation, texture, spatial depth, and dynamics on multiple setups. FLAC sounded a lot more better. (now before anyone says it's issues with my setups or placeabo, I promise you it is not.)

*EDIT: more details can be found in the comments.

 

Hello, new person. You should really follow the directions and post in the introductions thread. Doing so will make it seem like you’re an actual human being who wants to engage with a topic as opposed to someone trying to stir up controversy. You should also provide actual details on the setups in question and what variables you’ve controlled for. Supposing I take you at your word, there’s too many possible explanations for this to even point to a specific cause. I could speculate, especially given what I know of Apple Music’s behavior on various platforms, but that wouldn’t be helpful to anybody. Without those details, I could also just as easily speculate that you’re actively looking for an example of crazy audiophiles and threw out a bogus question to get the results you want.

Suffice it to say that there are specific, identifiable, and replicable causes which might result in this effect depending on what each of the signal chains look like. However, given the evident lack of thoroughness in your post, you could also just as well have neglected to turn off Sound Check in Apple Music. If you genuinely want to go through this methodically to build your understanding, you’re going to have to go post that introduction, then come back here with more details.

 

Hi, sorry im a bit new to this. I don't posses any DAWs but I do have the files in audacity with images on how the waveform varies, I can also list what systems I have tried it with. Would you like me to delete this and where is the introductions thread?

 

Matt, you are fine continuing this thread. Continuing your introduction is highly recommended, contex and other things.

You are definitely asking one of those mysteries of audio *files*. Why do services sound different? I like the fables of WAV files reducing the noise of systems. Though I’ve never seen a measurement to quantify the change. only cd transport enthusiasts tell me it’s better.

On Apple Music sometimes they offer the Dolby Atmos mixes, so they do offer both and it isn’t obvious how you select but the icons do change near the title in the program, and on the phone app too. So maybe you are hearing that, at least that would be the simple answer.

 

I’d need information on the various playback systems and mechanisms to comment further. Apple Music is really only going to give you bit-perfect output on iOS, and potentially on the Mac if you match your system sample rate to the content. Outside of that, resampling is involved in some form, and potentially an ASRC if you’re using an Apple TV. Sample rate conversion is the most likely culprit if other differences, like Sound Check, have been eliminated. I can’t speak to Tidal since I’m not a subscriber to that.

 

This is a vaguely related tangent that's gonna have me on the loony train again but heck, seems even different browsers can end up having an audible effect on streaming the same YouTube video. This is with all enhancements off so is basically down to how each browser handles the incoming data. I wasn't the one who generated these files so pardon the slight difference in filesize:

Different Browsers (Linux)

There was also a thing where I struggled with why FLAC and WAV file downloads were making a (small but discernible) difference straight off of bandcamp some time ago. That made no sense to me either and the fact that I couldn't tell any difference comparing FLAC I converted myself from the source WAV file to that source WAV file confounded things further.

https://www.superbestaudiofriends.o...-music-file-format-and-why.14126/#post-423552

I can't seem to find the files I made before so might have to redo this experiment. Would be curious to see if things are the same lmao. That said I'm reasonably sure a big part of this *could* have been foobar2k messing up somehow or other... but it's bit-perfect playback either way so shouldn't matter then right? But then why does using JRMC or Audirvana allegedly make for an audible difference even with all DSP disabled?

All told, "digital audio is weird as hell" is my main takeaway at this point. If it sounds good enough then that should be good enough; an excess of nervosa will only ruin your ability to enjoy music.

 

Woah, hello everyone!

So I have been an Audiophile enthusiast for about 3 years now and have done extensive research into audio (from a consumer perspective). I don't have the most fancy setups on the planet or professional software but I got my ears and heart.

For testing or should I say more what systems (all of mine) that I have noticed local FLAC sounding better.

1. Macbook, Apple Music and Tidal and FLAC via Audirvana. It was on a marantz 2238 and boston acoustics a70's. The main things I noticed was the directionality and instrument serperation of FLAC was superior.

2. PC, Tidal, Audirvana, MusicBee. I used a DX5 via usb and balanced cables to a Hifiman Sundara. On these headphones I noticed that the soundstage and layering of sounds were more open on FLAC files compared to streaming (all bit perfect, no eq.).

3. PC, Tidal Audirvana, MusicBee. I used the Audeze Maxwell via usb/ dongle mode. Bit-perfect, no eq of course! I ended up observing similar things to the boston acoustics a70's.

Oh I have also tried seeing if the files are different through audacity, and they were however the tests werent perfect, unfortunately I can't get it perfect on my setup. unless it is and there is a difference XD.

Yes I observed different things in each setup due to the tuning and sound signature of the different equipment however one silver lining was that local FLAC always sounded better, when using Tidal or Apple Music on the marantz and boston speakers I could go "oh I hear the sound to my left and it would fizzle and rotate around me" but with FLAC the sound was so clear I could point to where I could hear that sound, I felt truely surrounded by music.

if you guys would like anymore specifcs please ask!

 

Hello Matt. Glad you turned out to be a an actual human music enthusiast not an AI bot or a troll . Welcome!

I'm an old, rather deaf ex-'phile who just hangs around here (I abandoned other audio forums as being either too full of 'phool bullshit or no longer relevant to a guy who shifted his spending to photography). But I first connected my computer to my hifi over 25 years ago, back in the day when built-in sound really was crap and Soundblaster was true to its name. I progressed from a 5-GBP add-in card to a 50-GBP card to a bargain-basement discontinued RME card quite quickly, and each gave noticeable improvements. In fact my RME card sounded better than my Cyrus CD player and I stuck with it long into the days when dogma denied that internal cards can possibly sound decent. I moved from Windows audio to Linux audio. I had a firewire interface, which meant, in Linux, using JACK and I still do just... because.

After all that waffle I conclude: two different instances of playing the same source via a computer will not necessarily sound the same!

I don't use any of the streaming services. I don't listen to enough music at home to make that worthwhile (I get most of my music live) and, audiophile-shock-horror, Youtube is mostly good enough for my bad ears. Sometimes one has to pick and chose the listenable copy.

And sometimes even that isn't listenable. But it is after a reboot!

WHY? Well,I'm an ex-techie too, although I never dealt with internals, but there lie the answers. PCs are not real-time systems. Realtime systems will do just one thing at a time and won't interrupt it for anything else. Your audio stream is being interrupted (actual technical term) all the time, because it is almost certainly at a lower priority than lots of other stuff like outputting to the screen.

At its worst, there will be actual breaks and dropouts, but at less-than-worst there may be distortion. Sometimes very slight, but audible, even to my ears.

I am something of an objectionable objectivist. Bits are bits. But your computer doesn't necessarily treat them all equally.

This doesn't address the question of why you should hear consistent differences according to source. It addresses our expectation of a kind of repeatability from computer systems.

As to the difference in sources. I strongly suspect that the sources of the sources were not identical. Have a look at the earlier conversation that
@Lyander mentioned.

 

Yes I totally get you, that's why I use exclusive mode to priortise the audio. (Also I don't rip any CD's or anything I buy FLACs as close to the master ie Qobuz, 7Digital, Bandcamp.)

 

And that’s your difference. Audirvana supports exclusive mode, everything else is going through the system mixer. To try to minimize this, did you use Audio MIDI Setup to change the sample rate of your audio interface to match the content while playing in Apple Music? Also, can you please go through the preferences and double check to make sure all the audio-munging stuff like Sound Check is off? Even once that’s all done, Audirvana is taking over the audio interface directly and Apple Music isn’t.

Hopefully - hopefully - if you match the sample rate and set Apple Music’s volume to 100% you’ll get the same bits coming out, but there’s no guarantee of that, nor any guarantee of real-time processing. I can’t speak to Tidal or PC stuff, but you should look for similar causes there.

IF the same bits are going out to the DAC, and the system is keeping up with everything and not silently dropping samples, then you’ll get essentially* the same audio out the other end. If not, well… it’s not a valid experiment because there are factors not controlled for. That’s actual science as she is done.

* Within a degree of approximation. If one of these players is inserting a glitch, even a momentary one and before the actual audio playback starts, the impulse response of that will continue to ring indefinitely through any analog domain filter or physical transducer. Indefinite doesn’t mean infinitely in practice, but frequency-domain analysis alone won’t capture this effect, and linear analysis won’t capture how that impulse response interacts with the nonlinearities in any real playback system. The tools we use to characterize systems are simplifications that make analysis tractable, not actual models of the entire electrical and physical response of a real signal chain. That would be an interesting exercise though!

 

When on windows I do use exclusive mode no matter what (I have to), Tidal, MusicBee, Audirvana, all of it. On mac that isin't the case so when testing on mac I just used Audirvana because it's the only way I could think of properly playing FLAC files on Apple. (when using apple music I use apple dongle and iphone whereas I used Mac and Audirvana for FLAC)

 

Last time I went down this road I compared Qobuz w/ local files (from various online retailers such as Bandcamp) as at the time Tidal was MQA & I don't trust Apple's and other big tech not to process and 'enhance' files (it's just software to them) in the background even when they claim "bit perfect".

When I was reasonably (never fur sur - info is always limited somewhere) confident it was an apples to apples comparison (e.g. same mastering, nothing processing files in between and all software/hardware settings along chain were properly set up I found no significant difference, as in anything I thought I heard different was still within placebo range.

@as darmok points out, there are just too many variables/gotcha's in a comparison like this, and even otherwise competent computer/audio folks miss stuff.

 

In my limited experience with streaming is that I find the sources sound identical (local or Tidal of the same master) if delivered exactly the same way to the DAC.

Streaming service apps have a different way of delivering that file to DAC that goes to different pathway (I.e. for Windows OS and Audirvana, music is preloaded to RAM and is unpacked at RAM level before it's sent to WASAPI layer) while Foobar2000 and native streaming uses the SSD/HDD only for unpacking FLAC or ALAC then gets sent to WASAPI layer. Apple music and Spotify have the longest pathway which goes to Windows mixer layer before WASAPI layer.

The RAM and SSD have different performance which can be easily measured by Random IOPS. The differences in performance between SSD and RAM falls in the placebo realm obviously when comparing Audirvana FLAC (KS or WASAPI) playback vs Tidal WASAPI output

 

I'm not sure I agree that X is not hifi simply because Y is "more hifi". something is always "more hifi"

 

Especially when adorned with the atomicknob golden volume knob option as seen here

 

@MattKim6 you might find it interesting to read this post

 

I forgot this thread existed! Can we re-pin it to the front page because damn if this oughtn't be mandatory reading for lots of folks.

 

Hi, everyone.

The reason for the choice of title was purely for attention and to cause curiosity.

I'm not claiming lossless streaming isin't hifi. It is, I can hear it. However it's just from what my research and other people have said over the years I thought "lossless" was lossless but this enigma of sound I have discovered goes against that, I thought It would be good to talk about it. (somewhere more respectful than discord it seems...)

Now I'm not claiming that the sound is necessarily higher quality; to me the sound between FLAC and Streaming FLAC is very similar/ inextinguishable but what I am asking from you guys is to please give it another go, properly set it up for yourselfs and listen. Hopefully you can notice what I did. Instead of listening for how high the peaks are or how low it can go, listen to where the sounds come from, can your ears identify each layer of sound perfectly, try to visualise the listening experience and like me maybe you will notice something odd. Streaming FLAC/ ALAC sounds for some reason slightly more flat, slightly more mushed together. I don't get beautiful transitions and transients that I can hear on FLAC, the texture just the whole visceral feeling of the music, that perfectly tapered whispering around me, It all goes away.

This is just my experiences and as far as im aware my testing is proper. Yes this could be placeabo but the reasons I think it isin't is because it is a consitently repeatable difference for me (even at lower volumes). And because my uncle (45) can hear these differences as well.

Long story short, I came here to see if anyone could notice what I can. Don't try to over-analyse it because you will end up missing it, just listen and feel because that's the difference I noticed not something that can simply be shown on a graph, and my conclusion for if you do hear it is some sort of very minimal or slight streaming compression because I doubt labels are handing around a slightly worse master/ different master to every platform out there (you would think majority of the time it would be the same master).

I would also like to add I noticed this difference by accident. I originally bought FLAC because one of my albums wasn't available on Tidal anymore but it's when I pressed play and listened to it, that I noticed it sounded different then what I was used to, leading me to buy and try more songs directly.

 

Who knows, some people here claimed to hear differences between let's say Foobar 2000 and JRiver. So there's that, maybe the player does influence things (i.e. your online Tidal player vs whatever else you're using offline like Foobar 2000). Me personally, I always listen to CDs over Foobar 2000. I wasn't able to differentiate it from VLC or the Windows Media player. What I can hear though are obvious differences between Flac files I ripped myself and Spotify MP3 quality. On certain live recordings the differences are there and mostly it's soundstage/imaging related.