Lossy vs Lossless Audio Explained!
ฝัง
- เผยแพร่เมื่อ 28 ก.ย. 2024
- Lossless audio streaming is going to get a boost as Apple prepares to launch its own Apple music lossless streaming service.
But what is lossless audio, and is lossy audio all that bad? Let's get into it! Oh and sorry for consistently mispronouncing "WAV"......
========================
Thanks for Watching!
Please don't forget to subscribe to the channel - let's build a great community of audio nerds who don't take themselves too seriously!
Music from those great folks at EPIDEMIC SOUND
Come hang out on the discord server: / discord
Follow me on twitter: @AudioFixated - this is the best way to contact me!
========================
#AudioFixation #Lossless #Audio
I think it's harder to hear the difference between a good lossy file and a lossless file if you aren't familiar with the song, hence why I only get about 60% of the tests correct. However, when I test my favourite songs, which I know very well, I can easily tell the difference.
You should download foobar with the abx plugin. I did and tested that theory a long time ago. With my absolute favorite songs, I got up to 224kbps aac before failing (yes I passed 192kbps). And I used high quality unforgiving headphones with a good dac and amp. I had some of my friend AB test 224kbps aac against the original source. They all said they can hear a difference. Then I bet them they can't when done in ABX. I made plenty of money from them. They were mad at me for hustling them. lol
It will be noticeable only if you have a good audio system.
But Hey.. enjoy the music not the other nuances
I listen to FLAC because of FOMO and the placebo. Unlike others I don't lie to myself and am being honest. I want the best quality no matter if there's an audible difference. If a song is under 25mb it's low quality for me.
@@l4kr For me, my reason for ripping to FLAC is that I don't really have a reason not to. Storage is cheap, I have a 2 terabyte hard drive in my system that is basicaly empty.
Might be hard for you. Lossless is way more important than any speaker
I’m exactly the same, can’t hear a difference above 192kbps MP3. I stopped being an ‘audiophile’ once I realised that a pair of Grado SR80’s plugged into my phone listening to 192kbps MP3 or Spotify high quality was more than satisfactory. Better things to spend your money on than high end audio. Deposit for a house for instance!
you have been cursed.
320kbs sounds the best, no artifacts in sound.
You are deaf.
Lossless, such as CD quality audio, is not really lossless because they have stripped out the ultrasonic frequencies that most musical instruments emit. So that may be why people can't tell the difference. It's because lossless and lossy are both crap without the ultrasonics.
@@dtz1000You mean frequencies that are beyond human hearing. Frequencies never recorded in the first place 20hz-20khz. Frequencies above 20khz that are cut out in the recording process to stop aliasing artifacts. Mate, I've got an invisible suit to sell you. Ultrasonics, smh.
It’s also possible that the reason why some people maintain that they can hear the difference is because there is a difference, but not because of the sample rate or bit depth, but because the track has been remixed with lossless in mind. If an audio engineeer is tasked with mixing a track or an album he knows is going to be heard on systems designed for it, it would make sense to mix it in such a way that substantially changes the sound. Apart from that, the fact that Apple are selling this feature for $0, if people want to use it and then play it via nice amps and speakers, then I’m happy for them to do so. I like nice things too.
There is an audio test on NPR. It provides several songs each encoded at 128K MP3, 320K MP3, and uncompressed WAV files. With my audio set up. I was able to reliably determine which track was the 128K MP3, but I could never determine the 320K MP3 and the WAV. It was always a guess.
In addition to the usual cut off above 20kH for mp3 and other lossy formats, you also don't hear the beefy details of the various instruments playing in a soundtrack. This is evident with Rock, Electronic and Trance music where a 256 or 320kbps file would just give you noise when the hi-hats and/or arps playing in the background along with the beat and guitar riffs.
can you name the specific songs you are referring to so we can test?
@@Viewer13128 My friend showed it to me on his planar headphones, I think there were a couple of SUM41, and Trust Company songs plus Sean Tyas tracks that I listened to.
there is a chance this weren't true 256 or 320 kbps.
there are sometimes relic files from the early days of compression between 64 and 192 kbps, and there the noise, ringing, muffling and bloops become very apparent.
If you are listening to music through speakers the thing that makes the most difference is the room. I have a KEF LS 50 wireless system that I have tuned to my room. I've heard the same system in showrooms and in other people's houses at they all sound different.
I actually have experience with sound back in the 2000s with radio. Heck in 1999, (yes, I am that young) we had a CRT TV. It sounded great! I’ve also listened to tapes. They sounded great as well. Then I listened to Apple Music lossless. I can not hear the difference. But like you and my mom said, it depends on the song. If they want all instruments to sound clear, well they will sound nice and clear. I originally thought that lossless audio made ALL songs high quality. But I was wrong I guess. PICKLES AND PRUN JUICE
To do this test, I really believe you still need to have a good DAC that can produce lossless because without it even the lossless its still going to be not producing the lossless file.
The biggest use of losless files is archiving and you can easylie reencode ito somthing lossy (i prefere ogg.). But I agree that lossy is a-okay for everyday use :) I still prefere losless, but thats just my justification for my (relativly) expensive setups. I have heard some difference in the past with some really complex music genres like classical and big band, but still no real justification :D Keep up the good work.
I sometimes do mixes or just split tracks. If i did that with lossy files they would degrade each time.
I have done my own tests before and came to the same results, likewise found mqa to be no better when you compensated for the placebo from the louder volume (though mqa is not lossless). I do notice a difference in the car between Spotify from my phone over Bluetooth compared to my iPod plugged directly in, it sounds clearer even with engine and road noise. The music on the iPod is in V0 MP3. Thanks for the video, really enjoyed it.
I spent a lot of time listening to Flac files, thinking that I could hear the difference. Tried the website you suggested, and I can’t hear it. I feel so bad right now hahaha
*WAV = BEST AUDIO FILE, I AM GONNA USE IT EVERYTIME NO MATTER WHAT, I DON'T REALLY CARE ABOUT DISK SPACE!*
*WAV FILE ROCKS (no really bro/sis it is the best for audio, I mean you could use any other one but I am gonna stick with WAV)*
why not flac though
I was one of those Napster users 😆 majority of the Mp3 were at 128kbps!
They tend to sound like there is a towel covering the speakers?!
It’s good to know 192kbps is indistinguishable from lossless audio, I always rip my cd’s on that bit rate.
now I understand Lossy and Lossless different .thank you.
320 mp3 is as good as any flac or wav I've ever listened to. I was Flac guy for years but now it's all 320 mp3. I have a fantastic system with very hi end speakers in a sound treated room specifically for deep listening. I have ABX on many many songs and can't tell the difference. If I guess correct it's because it's a guess, not because I can hear a difference. Yes you can visually see what’s been removed in all these wav comparison audio programs to “show” you what you’re missing, but can you hear it ? not likely. It broke my heart to have to go to 320mp3 but totally worth it after months of re ripping all cds to 320mp3. Don't ever let any audio snob, as I once was, tell you differently or make you feel small for not hearing a difference. The man behind the curtain lies. Trust your ears and not some audio snobs delusional dream of self importance.
could you maybe do an episode on how amplification (from underpowered amp to appropriately powered amp to overpowered amp) affects your speakers or headphones? i recently had an awful time finding out how my apparently insufficient amp was affecting my 600 ohm 990s and the stuff coming out of them
good idea, sounds like an idea for a future audio explained episode!
I honestly can't tell the difference between apple music, tidal and yt music even after using a dac and sony's headphones
I probably can’t tell the difference but why not use lossless with storage being so inexpensive?
Compression rate needed Depends on the complexity of the music and the quality of the DAP and of couse the earphones.
with my testing, I could encode a 192kbps MP3 file using Razorlame with -no filtering and 'forced joint stereo' which made it impossible to distinguish from a lossless file regardless of system. I've heard one guy can distinguish in 9 out of 10 on his 10k worth system, but I have my doubts ;)
An MP3 at 256 kBit/s retains all the frequencies like the WAVs. So the "very high frequencies" are >20 kHz > Ultra sonic.
Compressed music suffers more in the high frequencies, the compression ends up creating more distortion and artifacts up there, perhaps for you there is not that big of s difference because you can barely hear anything above 15kh
Thank you for the video. Very informative and easy to understand.
There is another consideration that can be easily confused with the final music delivery formats like FLAC or CD or MP3. It is the quality of the source recording. We all get exited about new HIFI gear like DACs, Headphones, Amps and speaker. However IMO we don't know much about the quality of the music source material - master tapes/recording/mixing. Looks to me that there are music industry dirty secrets about the original master recording of the artist and how they are used and preserved. For example how many of you know that a fire has destroyed 118,000 to 175,000 audio master tapes belonging to Universal Music Group (UMG). (according to Wikipedia)? We don't know what was lost (artists, albums). What backups did they have , if any? So it is possible that we are listening to a copy of a copy of a copy of an analog tape from 1970s digitized to CD quality or better, with HiFI Music golden logo through the streaming service. On the other hand I found many recordings from 1980s or later with excellent quality in CD format. My bottom line : to achieve high quality of the music experience the entire chain of the master recording/mixing -> media production-> distribution,->delivery to users in different formats -> user consumption must be consistent on high level of quality.
So… except for Apple and Amazon, the rest of the industry has been charging folks extras for something most people can’t tell??
Yup! And in the case of Tidal with their “MQX” crap, can actually make the end result WORSE
*Can, doesn’t mean “always does”
Apple states on their music support page that anything above their 256 kbps is “virtually indistinguishable” to the human ear.
Personally, I still have all my music possible in lossless or even hi-res lossless because I LOVE audio and want to be assured I’m hearing the best I can get.
I have a black Dragonfly DAC for my car stereo but it seems my iPhone no longer likes that DAC as it “requires too much power.” So I recently switched to the charging cable that I use for CarPlay, and to my surprise, the music via USB still sounded phenomenal and is crystal clear! After researching, I’ve come to find out Apple has their own built-in “MFi” DAC in all their cables. Which as you stated in a different video, supports 24-bit/48 kHz audio(lossless).
Interestingly enough, the Apple DAC actually seems to produce clearer audio to my ears, but the Dragonfly DAC does add more low end punch. Though not much…
I came across your video because I was trying to figure out exactly what bitrate Apple’s DAC supports! The more you know!
I can detect a major difference in my 2015 BMW using lossless Apple Music on USB. My theory is that cars have less than stellar audio DACs to suck out details from highly compressed music. These parlour brand stereos in cars do much better with lossless CD Quality. Less distortion, louder, deeper warmer bass, more separation. I can actually use the full spectrum of the volume bar without the treble tearing my ears up.
Interesting!
@@AudioFixation ok so right now I’m trying my Steelseries Arctis Pro Wireless wired in to the iPhone with the apple lighting dong. The stereo separation, lows and mid tones are so much more alive and have a more signature fuller sound. The audio is much more roomy. I used the same headset earlier on Bluetooth and it was much less punchy or spacious. I’m not an audiophile, this is not audiophile gear, it’s my pc gaming headset though it is Hi Res certified lossless in its wired form.
@@roadhouseguy good for you, you clearly have golden ears 😀. Would be interested if you could tell the difference in a blind abx test.
@@AudioFixation I’d like to give it a try though I can’t guarantee the sampling rate through a browser test. I highly doubt iOS pushes Hi Res Lossless through a web browser. On Windows, I don’t have an external DAC pushing that sampling rate without some sort of downsampling.
I finally achieved quality sound in my autos with a small red dragonfly dac, cable connected to my smartphone which has a sd card storing 320K mp3 files. I then use a short quality cable to feed the 3.5 mm aux jack of the car system. The dragonfly also acts as a small preamp to further help the auto’s amp. I used to think the speakers were the big problem, but I was wrong.
192kbps is not enough for me, because sometimes the cymbals sound "floaty" to me, but 256 works pretty well.
I tried that abx test and the sample songs were terrible. Tons of dynamic compression and distortion just from the mastering.
i think a big thing in the future will be algorythmic audio reconstruction.
where a local neural net or set of algorythms get the lossy information and guess how the lossless file must have sounded.
it won't be the original, but a music theory following interpretation.
bad and useless idea.
@@kjererrrt2381 not useless. its good for restauration and to get better sound from limited bitrate streaming through local dsp
@@Chrisspru useless and bad. and inaccurate and unreliable and hard to implement in real time and many many other reasons for you to forget about it kid. but most importantly it doesn't make any sense even if it was a plausible idea.
@@kjererrrt2381 it can only work if the source has enough data for music theoretical reconstruction.
and using local processor power to restorate audio when bandwith is limited makes sense.
fir audio listening, a sn
small latency is no issue
@@Chrisspruforget about it once and for all, use lossless instead. tis crazy shit will completely ruin what's been left of music. you don't listen to music with real instruments anyway. it may work for you kid. why do you care about compression anyway.
I agree, can’t hear any difference. But then why do CDs sound better?
Of course lossless is better than lossy, it goes without saying but if you can't hear it, then don't worry about it.
this'll be helpful!
Hires Trick Marketing only
This is not right approach , Apple losses music files goes up to 24bit
/192 Kz. … and then you can easy hear difference between Aac lossy file and 24 bit lossless file.
This man compare AAc and CD quality what is 16 bit/44 Khz And yes … you don’t need “golden” ears, just hi-end dac, amp and hi-end spaakers.
Diifference is more noticeable with quality recordings, eg… acoustic quartet in venue with nice natural reverb , jazz quartets ..etc. Listening EDM music from phone in noisy subway is not the reference point, and soundstage in audio is important part of experience.
Hello audiophiles, of course all will disagree.
not worth it. end of story.
For me, I've known for about a year i can't really hear the difference, but it's more of an assurance thing, i know that even if i can't tell, I'm still getting the best experience, it also stops corruption and stuff which is great
I think you mean generational loss.
i put just about my entire CD collection onto a 64gb card in AAC 320 KBPS close to 4000 songs , and i am very happy with the sound quality. and the convenience of it.
7:56 that's actually a good reason to have/stream lossless, because you'll still end up with better quality than transcoding an already lossy file
This is true. Apart from if tie listening to Apple Music on AirPods in which case the AAC file is passed over Bluetooth without transcoding I believe. I think that’s why Apple Music sounds better than Spotify on AirPods
So, when am I listening to any music with my AirPods Pro , pick the lossless or the lossy one?
(I like to get the most from the music, but if it doesn’t matter to my ears so I stick with the lossy one)
@@ohadrahum1384 if you're on an apple device and it actually does pass thru aac files, then I guess it doesn't matter, other than that I'd pick lossless
@@AudioFixation it's actually *unlikely* that it doesn't transcode, the audio format would need to be encoded in a way the exact hardware decoder can decode it in case of AAC (since there're multiple AAC libraries available). I.E. it might work with Apple Music's AAC, but it might not work with TH-cam AAC due to using different libraries (with different royalty fees attached as well 😖).
In short, AAC isn't as bad as MP3, but it's still a mess compared to OGG Vorbis / OPUS (both royalty free & with only 1 library namely libvorbis & libopus which are compatible with a single decoder).
Of course you *could* include all libraries, but you'd obviously pass all the fees onto the consumer which would position your headphone in an unfavourable position cost wise.
Just because we don’t hear 👂 a thing, doesn’t mean we cannot hear the same thing. Can we hear things on a subconscious level? Do people bodies, are parts of there bodies including things people wear interact with inaudible sound frequencies. Just because we don’t test for resonance, does not mean it’s existence is in question. There are other questions that need to be answered before arriving at your conclusion. Moreover, apart of the gift artist, musician, and sound engineers 🧑🏾💻 have given is being taken away without informed consent from the gifter and the gifted. Most gifts are useless, it is the spiritual connections which are formed, and the creation of relationships engaged in for a collective experience to come into enlightenment is what is important. What do they say, it is not the size that counts, for sex happens in the the mind of a person. Blessings and one love 🙏 Can you hear, taste , see, or breath love. Yet the whole world 🌎 believe in its existence!
Ssserrt5a
When doing ABX tests before I could tell the difference up to 256 and sometimes 320kbps. In the test with The Killers' Flesh and Bone, I can particularly hear the compression in the cymbals, I'm a drummer so I pick up that part of the song more by default. This ignores the fact that when I listen to music, I do just that - I don't try and pick out detail. But storage is cheap and so I put 320kbps on my portable devices (modded iPod's are my poison of choice) which I transcode from my FLACs kept on my hard drive, I figure if I'm gonna spend the time tagging all my music, I may as well so it once and do it well!
Great video as always, thanks!
Yeah the killers track and hotel California are the 2 I could most easily spot. Thanks for watching 😀
And to add to the cake, there are the lesser known hybrid codecs eg. WavPack Hybrid, LossyWAV, OptimFROG Dualstream, mp3HD, MPEG-4 SLS (these last two are nearly impossible to come by nowadays). WavPack Hybrid is my favourite. Around 4bits / sample (~384kbps) it produces an excellent lossy base layer, but with the correction file you can restore the full lossless version. You can archive both, if you are in need of disk space you can delete the correction file and the lossy base functions alone. Very neat, and you have to only add metadata and covers once.
I have done the abx test on an actual DJ set, and I did tell the difference immediately bwe6een 320kbps mp3 audio file, and a vinyl/CDs. therefore. I honesty believe that lossy audio is a backword format, like cassette tapes and vinyl but much worse, the consumer are just being robbed for buying into these crappy lossy formats. A 320kbps mp3 audio file has 78% compression, which means the listener is only getting 22% of its original quality and they are paying the full price for it.
Refreshingly honest video, thank you. I too doubt anyone can hear a difference between lossless and 320Kb/s MP3 or Ogg Vorbif. If a difference can actually be heard there will be some other explanation such a different master, different volume or EQ and nothing to do with the format. I will try Spotify Hifi when available but I am not expecting any audible improvement.
Vorbis
I’ve done a series of A/B/C tests, and confirmed I can sort of hear the difference. Still, the lossy files didn’t sound as bad as I was expecting them to do. My rule of thumb is to always go with lidless CD quality audio (16bit/44.1khz). That makes me sure that the format I am listening to has all the possible detail capable of being caught by the human ear.
I have Spotify & did an a/b test with a free trial of Tidal HiFi. Tidal sounded better with more pristine highs & wider stereo separation, but I can't help but wonder if that was due to them tweaking the sound like a studio engineer.🤔
I’ve had a near identical experience with downloaded FLAC files as well as Apple Music which I use right now.
I’m just trying to find ways to justify Spotify because everything is better than Apple Music. But the sound quality is just better.
In my opinion how a recoding is mastered and what hardware it’s playing it on are more important than if it’s a lossy or lossless format.
I think you are almost correct but I think some people can hear very little bit difference in lossy and lossless with using ultra high end listening headphones 🎧 and DACs. But most probably your opinion is fits most of people. 👍👍. other people are actually Batman 😂
Bottom line, stream what you like.
I just came across your channel and man I've never been so wrong my whole life 😂 now I see why some of my mp3 and flac files sound exactly the same, but there are some songs in my library I've compared in different formats that are slightly louder/better in my ears than the lossy ones. Just want to know what's your input on that. BRB checking your other vids 👍🏻
I tried this the old fashioned way years ago, converting one FLAC file into a ton of bit rates and I couldn't tell the difference until I went from 192 to 128 as well. Lossy audio is some pretty amazing stuff, really. Keeping everybody but the super nerds happy until just recently is impressive. I archive in FLAC just in case one day I can hear the difference, since not long ago I couldn't hear the difference between FLAC and gravel, so poor was my setup. Still, I have a healthy scepticism of those who purport to hear the quality in FLAC; again, those who created these standards weren't fools by any means.
Yeah I archive in FLAC too. Just because storage is cheap and peace of mind in case I lose all my CDs!
I listen to lossless, but I just find lossy audio fascinating and just tried some tests with OPUS about a year ago, just to see how far I could take it before some songs started to fall apart. Surprisingly some "songs" held up all the way down to 64kbps... 😅 (with there being some difference of course, but without critical listening I was easily fooled).
Tidal doesn't even do lossless. MQA is not exactly what it originally promised.
Good point
Agreed that this marketing language "Authenticated" is very dubious. However, I have been impressed with some of the "file format = .MQA" sonic impressions!
I had big plans to use lossless compression when phones had enough storage then I found out Bluetooth can't play those files. Bluetooth won't be HiFi for many years if ever. At home I use an Apple Airport Express as a network attached DAC and play uncompressed .wav files from iTunes over AirPlay. That uses my network not Bluetooth so there is plenty enough bandwidth. I then use the iTunes "remote" app on my phone as a remote control for the laptop that's playing the files. It's a little trouble to set up but the sound really is CD quality and fantastic. The big down side to .wav is that you can't have album art but the quality is worth it when I'm at home and ready to just kick back and listen. It's a shame Apple doesn't sell the AirPort Express anymore it's an amazing product. You can use Apple TV to AirPlay but the new one only had digital out so you have to have an external DAC or playback system that can accept optical in or extract audio from HDMI.
You didn’t touch how stereo image is lost in lossy formats. WAV forever! It’s good btw let’s fool the masses and keep share of pie for ourselves 😉
I’ve watched two of your videos this afternoon. Both have put me at ease and made me very happy with the setup I already have. It’s very old, but it’s GOOD. I keep seeing videos about the new equipment, DACs, streamers, tube pre-amps, etc., and it makes me feel inferior with my home-brew decommissioned iPhone8 and iPad-based library and streaming. A good deal of my MP3 files actually DID come from Napster, and for most 1950-1985 popular and/or country music, 128kbps IS JUST FINE! Classical and jazz is better above 192, and I try to always rip at 320 if it’s the good stuff. Thank you!
The reason lossy formats were introduced years ago was to reduce file sizes and thereby storage requirements. These days, gigabytes of storage is so cheap that the need for lossy formats has pretty much been removed. Given that CD standard is all our ears can appreciate ( Hi Res formats are just a scam to resell the same music in a different ‘wrapper’) there seems little reason not to just accept the 16/44 standard as the one to settle for. Sure, the differences between 16/44 and good quality lossy versions can be vanishingly small ( and much more important is the mastering etc) but why throw any data away if there is no need - esp. for those with good enough ears and systems that can appreciate the last degree in fidelity.
People get high with words, tend to believe that they are really having the experience as projected by the sound technicians, forgetting that the projection is based on theoretical results and has got nothing to do with real time listening..
I’m no exception, I love and defend flac, have all of my cd collection converted to flac and look for a flac format whenever possible, but I confess that I have not done the blind listening test. I’ve always thought and taken it for granted that flac is better than mp3 or aac.
Thanks once again for the video.. I’m feeling an urgency to take the blind test.. ha ha
👍😄
I'd be interested to hear how you get on with the blind test.
Sure.. 👍
@@AudioFixation Hi..
i did go for the blind test at abx.com but felt that the song(s) provided there were not my favorites. it was a hard test and not entertaining, so instead..
i searched in my collection and took out the following songs,
1. another brick in the wall by Pink Floyd (WAV 1411 kbps)
2. smooth operator by Sade (WAV 1411 kbps)
3. hotel California by Eagles (WAV 1411 kbps)
4. London town by Paul McCartney (WAV 1411 kbps)
5. evil ways by Santana (WAV 1411 kbps)
6. enter sandman by Metallica (WAV 1411 kbps)
and,
7. jeremy by Pearl Jam (flac 3200 kbps and WAV 4608 kbps)
and a song by the doors which has an old recording (an old CD, AAD) which has a reasonable amount of tape-hiss and friction sound. i chose it to see the change in the underlying hissing sound and the subtle disturbance.
'hello i love you' (WAV 1411 kbps)
next i converted all of these to mp3 (highest bitrate of 320 kbps) using foobar 2000 v1.54
(here, i'd like to inform you that i had a listening test and noticed that my hearing starts just below at 20 Hz and goes all the way upto 17.5 kHz. i am 54 and i found that result to be better than many of my friends)
Ok, going back to the topic,
i choses JRiver for playback and passed the sound through USB to the Cambridge Audio DAC MagicPlus and listened through the SONY headphone MDR-HW300K (though this is a wireless headphone, i used the supplied wire to use it as a conventional wired headphones). i repeatedly played them all at random, listening to different segments of the songs, paying special attention to the lower notes (for bass) and higher notes (for treble) resonances and echoes as well as the overall change in melody and listening pleasure
conclusion : i must admit that the blind test brought no result as i failed to distinguish between two different formats, viz, wav and mp3. it was shocking, but i admit the defeat with humility.
next i listened to the songs categorically to separate any noticeable difference by looking at the bits and pieces. surely there was no big difference but at few places the WAV files had more bass punch/resonance and cleaner treble shine. this was more evident in the wav file of 4608 kbps.
overall response was that the mp3 sounded more muddled up and various high pitched notes (of around 10kHz) sounded as if smeared against each other. this was not a finding of a blind test and i am not sure if i would be able to tell these difference in a blind test.
anyway, this has been an eye-opener to me.. i still cant believe that the 1411 kbps WAV files sound same as a 320 kbps mp3 (i have always hated mp3)
thanks once again for the video, my friend..!!
@@jagannathghimire3039 Thanks for reporting back Jagannath and for posting your honest results - it's amazing, isn't it how hard it is to tell between them!
@@AudioFixation indeed.. 😇
Even more so, as I always believed that I could tell the difference easily.. no doubt I could tell the difference between a good recording and a bad one, but never thought what would be the scenario if both the versions were from the same source.. ha ha
Honestly, the overall clarity of human voice, bass guitar resonance and the subtle vibration of the cymbals still sound better in a flac or wav file, though I’m unable to explain it in simple words.. mp3, as always sounds tiring if you listen to it for a long time, again I cannot explain why this is so..
I feel like attributing all of such experiences and results to the mediocre quality of my music system, and feel that the music system has to be A-one to hear and feel the differences..
Thanks for your response and hopefully we will have more of such videos in the future as well.. 👍👍
I had the opposite experience back in probably 2005... I swore that high quality MP3's were just as good as the CDs I was ripping from. Ripped the same song as 320 and as WAV because FLAC was not much of a thing yet. On my humble Boston Acoustics computer speakers, the WAV was far better and I could hear it. Then it happened again with Tidal. I didn't expect any real difference, but did the trial which was all MQA at the time and got addicted. Odd twist is that sometimes, I've actually preferred the *lower* quality audio because the max quality sounded *too good!*
Hi, I have a Yamaha RN803 Amp for my home audio where i stream music from apple music through airplay. It has a ESS Sabre DAC built in. I was thinking of buying a DDC for connecting it to my PC to to the Amp via the coaxial input in the amp for playing lossless audio. Do you think it will make any difference?
High bitrate is useful in a mixing environment mostly for volume control. In certain situations.
Let’s say you have an amplifier with no preamp, in that case, you are using the digital source as a volume control. If you are going to do that, your 16bit audio will be compressed in its dynamic range.
If you are using a pc and it’s a 32bit desktop environment, you can control volume digitally with minimal degradation.
If you have a full volume output and an analog preamp (proper system) then it doesn’t matter.
I’m using Apple Music and set the quality to hi res lossless on all scenarios. Even in car stereo with Bluetooth connection, when switched to lossless, I’m getting an improvement in loudness with an additional punch for beats when hi res lossless song is playing. Can’t hear much difference with with just lossless over Bluetooth.
You could just use the volume knob to achieve the same effect...
So, do external DACs matter to you, or are the DACs in the phone or computer good enough? Another question is, if convenience is more important at the moment, how do you rank the BlueTooth CODECs:
-APTX HD
- APTX
- LDAC
- SBC (Samsung)
- LC3 (auracast)
I would take OPUS over AAC - takes much less space, needs much less bitrate sounds as good as AAC... if we're going lossy route, otherwise - flac it is.
I'm 99% sure I can't hear the difference between lossless and good quality lossy. I buy CDs and CD quality downloads because, just like you, I like to have a bit perfect copy of the music. I also play that music lossless on my DAP and via my PC in FLAC format simply because I have the space for it now that storage is much cheaper. Also simply the fact of playing bit perfect FLAC files just makes me happy even if I can't hear the difference. I'm also content with just streaming lossy music from services
Another benefit of having lossless audio files is that you can convert them to any existing or future lossy or lossles format. If I were to convert a lossy file to another lossy format, the new format will have less of the audio information to create a good file with. I'm not sure how much of a difference this makes in practice, but I'd rather convert to lossy from lossless file.
This video got me so excited to try it out and then it hit me i have yet to recieve a wired DAC to try flac
Lossless audio is great for audiophiles... Who can really tell the difference and want to experience their music in its purest form. Casual listeners who play their music while doing other stuff will probably be fine with mp3 quantity audio. It's good for when you don't have an unlimited data plan.
I accept a lot of what you are saying but, of course, it depends upon how the MP3 was recorded (I have a lot at 128). Now, I still will never accept that FLAC is not better than MP3 when I am listening on my Arcam system.
I kept getting 60%. I was using sony wh1000xm2 headphones plugged into a macbook pro. The first song had a lot of busy noise going on and for a song like that it's hard to tell. With less busy music with real instruments and vocals that are clear, I think it's easier to tell the difference especially if the original recording is high quality.
Isn't the Sony ldac codec lossless? I'm fine with aptx though. The real kicker is how muddy the headphones sound, which I despise. I can recommend the edifier tws1 pro for the absolute best bang 4 buck quality. Paying more is a waste.
This appears to be an endless rabbit hole the Industry is going down to have people buying gear and music that they can't even hear or tell the difference. Although science may have to continue to pursue perfection in audio I think there's a point where it's meaningless .
This is a philosophical question really. Why "make do" with lossy when you can "reach for the stars" with a better digital representation of an analog sound which is what we are truly talking about? High-Fidelity has always strived for the nearest approximation to the analog. Why stifle innovation? Promoting High-Res and the likes (e.g. MQA) will encourage people to invest their mind space to inventing a breakthrough in the future. I am personally impressed with the sonic experience (not a purist) that I get from most recently recorded MQA stuff. However, I must admit, the older stuff can sometimes sound terrible. Eyes on the future...
btw I share your appreciation of PMC speakers. I have a pair of 20-year-old FB1s that sound pretty awesome to me still!
Thanks for watching and commenting. Looove the FB1s. Great pair of floor standing speakers
Nothing hits like 32 bit floating point files when i buy my personal anechoic chamber.
...given i dont accidentally destroy the universe trying to play that file with my 1400 db future speakers.
Buying a CD feels good to me, because it is collectible. I could buy a cheaper MP3 music that pressed on CD than expensive audio CD instead!
Cassettes had better sound quality than MP3s because they recorded directly from CDs. And I only use wired headphones so far. 🙂
How do lossy formats usually play back? Is 128kpbs still in a 16 bit format? Would 128 be the final bitrate, or in a decoded state is it effectively more? If sections have been removed, how is it sent out so that the timing isn’t off?
wait, what's the difference between soft/hard sounds compared to quiet/loud sounds???
I'm sorry, but the only way you're going to get the best out of your music is to listen to CD's and Albums from your home stereo system. That way you don't have to worry about any of this lossy/lossless nonsense. And by the way, have you guys ever heard of this crazy thing called RADIO??
Apple lossless is good but the codec and the hardware which is non existent in the iPhone is an absolute shame.
And yes lossless audio matters to anyone who can hear a difference.
Most of what I listen to these days, only just recently came live on streaming services. I have figured a good way to buy HD digital tracks from Japan, so that's where I go. May not need the 192/24 files but since the price is the same if I pick that or CD quality... I just go for the higher quality files these days.
And while I don't think I can tell a difference on all tracks, I can tell a difference on enough to really dislike most streaming services. If HD got us away from that terrible quality compression, at the same price, sounds good to me.
Since Apple is offering this free as part of the service, here is hoping that other formats follow the leader. Of course, since I do TH-cam Music (free with the TH-cam Premium subscription I already pay for) and they've stated they give no craps about upgrading, guess I will continue to use my streaming to sample things I am debating on buying.
Jriver Media Center offers the AB/X option now as well, which I've used to figure out I can't really tell much difference of the 192/256 lossy vs lossless FLAC. On MOST titles/tracks at least.
Listening recommendations please!
Can I tell the difference between hi-res and lossy files? Sometimes. The issue is I'm no longer listening to the music for enjoyment and now trying to analyze the song which defeats the purpose of listening to music in the first place.
My choice:
For Archive 》Flac
For portable listening 》Aac/m4a
Excellent science and explanation thank you. I must ask though, not being an audiophile, why all the fancy amp, DACs, and headphones then? I can't even make use of the higher quality music with my air pods pro or Sony wh-1000mx3s, as they're stuck with lossy Bluetooth. But if a person can't tell the difference, why the expensive equipment?
Hi Victor
I didn’t say you can’t tell the difference between equipment. Regardless of the file you are playing, it will sound better on more expensive equipment (up to a point). So for example, the £100 moondrop starfields sound far superior to the Apple wired buds even when listening to Spotify compressed music, and my shure se846 make those starfileds sound lousy. Once you get to a certain price point though, you are into the law of diminishing returns, with only marginal improvements. That’s why for most people, I think “mid-fi” equipment is what they should invest in.
@@AudioFixation thank you for the explanation. I've recently bought some Samson sr850 and I'm probably grabbing a BTR5 to drive them as I have no headphone jack in my phone
@@victorvannatter312 I love the BTR5. Versatile little device
The only reason I rip to lossless these days is I have no reason not to, and see zero reason to use uncompressed in a world where lossless exists.
I'm old guy so I prefer enjoying music on my old cd player than modern streaming way.
I agree, I have a really nice hi fi stereo system and listen to cd quality or listen to my my vinyl!!!
Just an hour or two ago I listen to go slowly on both Apple Music and title. I purposely try to listen to specific effects to see if they were different on one from another. The title was richer with more produced affects than the Apple Music so there is a difference but if you’re not looking for it so what, I am peace
For me its all situational, I went between apple music and amazon the other night through my dac and hd 650. I was completely shocked about the difference. I felt like I'd been listening to mud for ages. Through speakers through hifi though or in the car not a chance I'm telling the difference though. So its defiantly situational for me. I've done the test a while ago. I could tell the difference between AAC and CD but after that nothing. To me there is a weird distortion I get through compression which only listening through headphones shows up.
A lot of the perceived differences are due to volumes being not quite matched, which is why it’s so important to use a proper abx testing programme. But if you can tell the difference between 256 AAC and WAV then I salute you 😃
@@AudioFixation You are correct. Increase the gain of one sample by a couple of decibels and they will choose that as the premium file. Though I think sometimes I can hear a difference in acoustic female vocals, but the difference is tiny. Then again some audiophiles can spend $20,000 just to get a 1% improvement.
maybe they werent sourced from the same file. Who knows what a digital music file (even lossless) went through before it hit your ears.
This is the best video I have seen regarding the high quality music
Really enjoy your presentation. Very clear, full of useful info. 2 thumbs up!!
Thanks for the great explanation. I have been telling others they will almost certainly not hear any difference between Apple lossy and the new lossless. I think Apple do such a good job of their lossy codec, but it will be good if only psychologically to know you are getting the full lossless information.
I am using Apple Music on the iPad to airplay to my 3rd gen Apple TV box using its optical output into my Cyrus amp which has a great built in DAC. I’m not sure if I can get lossless via that set up. I think from my understanding I should be able to, at CD quality, not high res, but whether I can or not I’m not sure as yet. I don’t know any other way to get it into my amp, at least not without hard wiring the iPad to the amp some way. Maybe just maybe Apple will make a dedicated streamer we can use for this.
That’s a complex setup you’ve got (but good choice on the Cyrus gear) - yeah not sure that will provide bitperfect audio from the atv box. You can get USB to optical SPDIF converters, i think Darko has mentioned them on his channel. I have an old (2012) Mac mini that I use to run an optical out into my Cyrus DAC-X which outputs bit perfectly, and that’s what I’m hoping to use for ALAC streaming - it’s currently running Audirvana for local FLAC files.
@@AudioFixation I have been watching some videos including Darko's on hard wiring such as lightning to USB then some sort of cable to convert the USB to digital RCA but a lot of that goes over my head. I might contact my Audio specialist later in the month to see if they can set me up with what I need. I only want CD quality lossless, my hearing is not going to be up to high res quality distinction. But if the older Apple TV box will do it then problem solved. The newer ones only have HDMI out so that’s not much good to me. I’m inclined to think a HDMI audio extractor will be not good for the audio quality.
@@AudioFixation I have my Airport Express connected via digital Mini TOSLINK cable to my receiver/speakers. The Airport Express supposedly has an inbuilt decent DAC. It can output decent lossless up to 24bit/44hz. Since the lossless streaming commenced on Apple Music I've been streaming lossless (where available). I detect a perhaps slightly fuller, richer sound but it could be placebo (I'm old!).
Bluetooth audio is not lossless? Do you hear about Aptx(Aptx-HD, Aptx adaptive) and about LDAC. But, mostly, video is very informative. Thank you.
Thanks for watching Alex. Sadly even LDAC is not lossless (it’s the best of the bunch though). Bluetooth just didn’t have the bandwidth to transmit bit perfect lossless audio.
All Bluetooth audio codecs are lossy, some come somewhat close to lossless & depending on the song might actually fall within the bitrate needed for a lossless flac (LDAC could for example fit at least one song in my music library within the available bandwidth of the latest Bluetooth standard, it's slightly higher than 500kbps compressed which is actually quite impressive).
Excellent review, just come across your channel so going through the various vids. The differentiator then above 44.1/48khz at 16 bit mainly is the quality of the recording and the mastering and its distribution via a streaming service or a download retail outlet. If i listen to the same album on say Apple lossless and Amazon hires at the same sample and bitrate they do not sound the same quite often and i have abx’d this. This implies either the master they receive isn’t the same of something else maybe the player itself is different. With Apple trying to rapidly build such a large library of lossless music I do wonder what QA they have in place for the masters they receive and their ingestion. One of the things with MQA which apparently has been discredited although I’m unsure that it is was the control over the masters quality was it not?
Welcome aboard Adrian and thanks for the comments. Yes, I too wonder why the Amazon and Apple lossless files sound so different…as for mqa, there’s a lot we don’t know about the process as it’s far from transparent.
@@AudioFixation Indeed as are the processes of most companies that rely on their IP to differentiate, the trashing of MQA and TIDAL was done at a very fortuitous time for the launch of Apple Music, I don't use MQA personally any more but I used to use TIDAL and found the audio enjoyment on a par with Amazon but no better, I really miss the purple lights on some of my DAC'S though 😊
Thanks, that was really very informative. Loved the presentation too 👍
Ohh I can't wait, this comment section will be interesting
very clear explanation. thank you for this
sony LDAC can send lossless bluetooth audio
Actually it can't, unless it is transmitting at 990KBPS mode in optimum circumstances, where it is probably equivalent (but not exactly the same as) CD quality. However it often defaults to 660KBPS or worse 330 which are definitely not capable of lossless transmission.
@@AudioFixation but it is possible, maybe LHDC is better, but I never tried it
Gotta do the ABX test on a proper FOH PA system, there the difference should be audible.
It actually does get very hard to tell at 192kbps MP3, although a modern codec like OPUS can easily push that down to 160kbps without there really being too many obvious "killer samples" like lossy audio used to have.
I believe only harpsichord is still a bit of a challenge for OPUS needing about 200kbps to make that transparent last I checked.
The fact that people genuinely *think* the difference should be audible, but in reality just isn't audible to the majority of listeners is a good indication of how far lossy audio codecs (and psycho acoustic models) have come. MP3 used to be pretty bad, but just a more modern psychoacoustics model makes a gigantic difference in that instance.
@E. O. I honestly don't know, I don't have harpsichord tracks in my music library & since I have the storage space most of it is lossless. Pretty sure you can download some samples from forums that discus OPUS & it's killer samples though.
This is an excellent review and lesson on lossless music. Thanks
Great video! What might the explanation be for my lossless songs on Apple Music being louder and more “crisp” sounding than non-lossless songs? Those normal songs often sound quieter at the same volume level and I can’t make out subtle sounds as easily.
@E. O. Yes indeed, I meant audio tracks. Lossless tracks sound *much* more crisp whenever I use my AirPod Pros or especially the AirPods Max. I’m pretty naive when it comes to audio-related technical jargon (your average Apple Music listener- but I love fancy tech, hence my purchase of the AirPods Max). Anywho, if there is truly no audible difference between lossy vs lossless on Apple Music, how come lossless songs are louder at quieter volumes (which allows you to hear those musical subtleties)? I have done ‘blind tests’ with my family and friends, and they choose the lossless track because they say it has “more detail”.
Fundamentally how are lossless formats compressed? Does flac just cut it in half? What about MLP/TrueHD for 2 channel, or even ALAC? I get it’s complex, but like video is complex it can be oversimplified in a way I want to understand.
I really appreciate you going into the science behind this stuff. Very easy to understand from that perspective
Thanks!
An interesting and informative video, and I agree wholeheartedly in your description of the sounds the human ear can perceive, however...the reason the majority of people are unable to discriminate between lossless and lossy music is due to their equipment not their hearing abilities. All home loudspeakers/hi-fi and audiophile systems alike, irrespective of the cost of your investment, 'colour' the sound. They are designed to 'improve the sound quality' of any and all music formats played. Lossless music is the default of recording studios for a reason--the same reason adding extra salt to a renowned chef's signature dish may well induce a pulmonary embolism in said chef--the sound an artist and their engineers produce is the sound they want you to hear, not a 'coloured' version. It's true the sensitivity of the human ear is limited to sounds (vibrations) in the range of 20Hz to 20kHz, and whilst it is indeed the primary organ for sound detection and location, it is not the only sound (vibration) receptor in the body. For example, Meissner corpuscles in the skin detect vibrations as low as 5Hz. The human body, as a whole, is capable of detecting vibrations (sound) in the range of 1-100kHz, far exceeding the range of our ears. The combination of these other receptors are the reason people with profound hearing loss may still enjoy music, the reason we feel bass notes deep in our gut and high pitched sounds in our skull. The online tests/home tests to discriminate between Lossless and lossy music are fundamentally floored. If I served up two similar dishes, then poured gravy over both, it would be difficult (if not impossible) to distinguish between them--your loudspeakers are the gravy, they colour the sound. If you want to enjoy lossless music, the true nuances of the recording sessions, you need to rethink the way you reproduce the sound: Replace 'audiophile' speakers with a pair of high quality powered studio monitors and sub (these will not 'colour' the sound) position the main monitors to form an equilateral triangle with your listening position. Connect your speakers to a 'stand alone' digital audio player (with balanced leads) and do NOT use any eq or effects. The position of your setup within a room will have a marked effect on the sound, change its position, change rooms--but do NOT add eq/effects. Recording studios are acoustically balanced, so if you don't wish/can't afford to spend mega bucks on balancing your room, you will have to find a room/position that is the best compromise. With this setup not only is the difference between lossless and lossy acutely apparent, the latter becomes painful to listen to. I have carried out several 'double blind tests' with the above setup and, with the exception of one young lady (who is no longer on my Christmas card list) everyone, without fail, chose lossless as the 'best quality sound'.