The FLAC format is basically a wrapper that holds the PCM data within. Similar to how when you zip a file, you end up with a .zip file, but your original file is contained within. Once unzipped, you will have back your original file, down to the last byte. Nothing will be different. The zipping process does compress your file, but that compression is lossless. The FLAC format compresses the PCM content, and does so losslessly. In addition to the benefit of saving storage space (due to the compression), the FLAC format allows for nearly endless metadata to be tagged to your music file (such as "Artist", "Album", etc, and you can make up any number of custom tags, too). If you do not need the extra tagging that FLAC offers, you can go with the Windows .wav standard. It offers only basic tagging, but it is uncompressed. The PCM data within is identical to the PCM data that is in a FLAC file. So for anyone that is curious about whether or not the additional CPU cycles that are required to uncompress a FLAC during playback, you can convert your FLAC file to a .wav file, and listen for any differences. Unless you are using ancient hardware (or driving a late model Honda), you should not hear any difference between a .wav file and a .flac file. My Honda Accord has massive jitter issues with FLAC files. So I converted my library of FLAC files to .wav files on a USB flash drive, and now they play in my Accord with no issues (except that the stereo kind of sucks). I am not, however, able to search through my .wav files (tagging issue). With FLAC files I was able to search, but the songs had massive jitter. Way to go Honda! They used a Panasonic front-end for the stereo, and Pioneer speakers. The stereo is not worth listening to, anyway. Honda dropped the ball on this.
@@DWHarper62 All: David Harper called Paul McGowan an idiot. That is why I replied to him that that is what an eight-year-old would do. Rather than David Harper apologizing for her insult, she deleted her "Paul is an idiot" comment. That, too, is what an eight-year-old would do.
It's not a huge difference and it's fine to covert wav to flac or alac. It's close enough. I'm just saying there is a slight differnce that i can hear and my digital meteres measure that difference.
@@380stroker No, they don't, because no such difference exists. They are literally identical. Let me know when you can replicate the result with a double-blind study using your equipment.
Something I've learned with these "high res" recordings. a lot of these services will sell you 24-bit versions whether it was recorded that way or not, if you really like an artist or a label. check with that record label to see what is the highest recording format they actually use. Save your money.
That’s not easy…. It’s not the recording format…it is the master you have to see… recording, mixing then mastering may all use different quality technically. Lots of audio may still be 16/44.1k but master may be higher ( why ? they put to the mix through analogue gear, compressors, limiters etc ) , so the crappy mix can be bumped up if the mastering engineer puts it through a $80,000 piece of vintage Abbey Road gear…..then the master decides the resolution. Recording may be 16 bit waves but the master may be 24bit.
With modern internet connections there is no need to deal with lossy codecs for audio streaming. Forget about MP3, MQA and other lossy codecs if you are serious as audiophile. It doesn't take much processing for a modern CPU or SOC to play highest quality lossless audio e.g. 192kHz 24 bits PCM. Digital audio streaming is already top notch today with master quality bit perfect capability but of course if you live in a ghost town in the Nevada desert or on your own island in the Pacific, you will not be able to enjoy it.
@@ford1546 Well, one CD can contain up to 680MB. 300 CDs will occupy less than 204GB. A Western Digital SSD with 250GB cost around US$40. Your CD collection is not "a lot of space" anymore. Ages ago I got my first PC with a 20MB hard drive and your CD collection would have been an insane amount of data :-)
@@Canadian_Eh_I Speaker distortion is usually much worse than any other distortion in your system. But of course compression distortion can become the worst in an otherwise great system if you run your audio through some low bitrate decoder e.g. 128kbps or less MP3 is audibly really bad.
@@ThinkingBetter Yeah I notice anything below 320 kbps. But to be honest, even with a very revealing system 44khz sounds great to me. 16 bit is just enough. Ive been playing around with active speakers and mods for along time and I finally came to the conclusion that the speakers are 90+% of what you hear. With very good speakers (proper waveguides) even the room doesnt matter that much. CHeers.
Even though FLAC and AlAC deliver PCM losslessly, there is a time variable involved that may cause jitter or a temporary interruption of that stream. So they are the same assuming that there is no interruption during the unpacking or decoding process, which as you say is trivial on most modern computers. However jitter can still be a problem, even when transmitting data from a streamer to say a DAC due to electrical or EMF interference, since that digital data is still being sent through an analogue carrier wave. Caching of that data and verification of that data after each step could mitigate that jitter, so that your DAC doesn’t start decoding a signal and have interruptions or errors in that signal due to electrical interference. I think that’s why HDMI is the best way to send a digital signal to a DAC, followed closely by USB, and then followed by the worst being SPIDIF. This is due to the time variable component and subsequent electrical noise that can interfere with the digital signal. As for DSD being the “best” that’s only true if all of your source material being recorded is acoustical in nature, as in sticking a microphone or microphones in a physical room to capture a live performance. However, this DSD information is then converted into PCM for mixing and mastering purposes and if you are using digital instruments, like synthesizers or drum machines, that are PCM based, then that difference is lost. Not to mention that DSD DACs employ oversampling and filtering as a matter of functional necessity, which may skew the sound from the original source recording, where as an R2R ladder DAC, has the option, and typically works best without any oversampling (at least internally). So DSD is only “better” than PCM in very special use cases, which in today’s world of how recordings are mixed and mastered only rarely give a better result than say a very high resolution…say 768k PCM recording format. An example where DSD might be better than PCM is in recording a live orchestra, where you are capturing all the instruments and the acoustical signature of the recording hall and the entire “mix” happens in the recording itself simultaneously. DSD was originally invented by the same engineers at Sony and Phillips, that also invented the PCM CD format, and DSD due to it’s insane file size, and inability to mix and master it, was designed to be an archival format for super high resolution digital audio. That’s why it’s not very useful in most modern recording studio situations. Not to mention the file sizes…OMG , for example one song of 96KHZ FLAC can be saved in say 70mb, where as the exact same song in 512 DSD would be 900mb! So file sizes, even in the current day of multi-terabyte hard drives are still very much an issue and a hinderance to having all of your music in high resolution DSD format.
You can just change the player to decode (in advance) the whole song in memory before beginning playing it. It would be extremely fast on modern machines. Playing it will then be 100% equivalent with WAV.
Paul speaks highly of DSD encoded music. I am streaming music through a PLEX media streamer, and the vast majority of it is "uncompressed" FLAC. It sounds very good to me, so I am assuming "uncompressed" FLAC is essntially identical with the WAV files found on my CDs. Some of the music I stream is high res--from 24/96 to 24/192. My PLEX handles it without problems. So how would I get DSD format tracks? Would PLEX handle them? Wouldmy DAC handle them? I am intrigued by what Paul is saying, but I cannot determine how DSD is relvant to me? Any insights?
@Any RD The vast majority of DACs on the market are delta-sigma DACs, in which case they only handle DSD natively, and if you feed them PCM they have a converter that will do on the fly conversion from PCM to DSD. So no it's never "downgrading" to PCM. If you have an R2R DAC then it handels PCM natively. The only case is where DSD is converted to PCM is on R2R DACs that have the conversion circuit, many don't. But you really don't want to feed DSD to an R2R DAC. DSD isn't 'better'. It's better for delta-sigma DACs, because that's their native format, as where PCM is better (sometimes necessary) for R2R DACs. What's bad is the on the fly conversion.
But PCM puts a higher load on the network by simply being larger and is thus more sensitive for network loads. Not very likely to cause issues for those on good connection but as likely as the increased load on the CPU for handling FLAC.
And any computer that’s still safe to have online (i.e, still gets security patches) is going to be able to decompress the audio without even shrugging it’s metaphorical shoulders.
@@Harald_Reindl Yes, now I'm streaming a 192kHz 24 bits track from Amazon HD Music and it occupies less than 0.1% of my available 1Gbps optical fiber bandwidth.
@@ThinkingBetter 0.01% :-) besides that in most cases it is buffered anyways - if my storage would be a network device mpd would load the next track and uncompress it completely before start playing - hence all the nonsense about CPU usage for decode FLAC golden ears could hear is exactly that: nonsense I really wish audiophiles would have a brain and technical understanding - imagination is for the sound stage when listen to music :-)
Good thing you didn't go to that PCM vs DSD comparison. Contrary to speculation DSD is not a better format and more time than DSD files are converted to PCM for editing (because there isn't enough available software to edit DSD) so if there were any gain it would be lost in the conversion process.
@@xanderguldie A ‘standard’ DSD file- often referred to as DSD64 is roughly equivalent to a sample rate of 24/88.2kHz. ‘Double DSD’ or DSD128 samples that single bit of information 5.6 million times a second to give you a signal equivalent to 24/176.2kHz.
I find PCM much more detailed and in-depth than flac of the same track. Maybeby the process of unpacking does something to the sound and make it a little brittle...
FLAC. has previously been incompatible with many players such as car players and older mp3 players. This is due to the flac. requires more processor power and then uses more batteries on the mp3 player. win7 and win8.1 win10 I do not think support flac. before installing software or used winamp. all new mp3 players and most mp3 players on ebay now support flac. Do not know with players in newer cars as they are a few generations behind car players you can buy in the store. I only use flac. and you do not save 50% on the file size when you put the flac on the recommended compression. If you want to be 100% sure that everything can play the music then you should use wav but it takes up a lot more storage space. pcm. and flac. has the same sound during playback. here it depends on the software that plays the file and the codec being used. When you need to transfer wav. through a network it requires more capacity since wav is larger files than flac is
The „varying degrees“ of compression only effect the size of the file, not the quality. The level of compression simply determines how much work the encoder is allowed to do in order to find an economic packaging of the data stream. On the producer’s side, a higher level of compression comes with a longer processing time and a higher energy consumption. On the consumer’s side, the level of compression doesn’t matter in any way.
CAPS 2 C NOT SHOUTING…. PCM VS FLAC, IS JUST 1/2 THE STORY… WHEN I PLAY MUSIC THRU MY APPLE TV I GET PCM, SOUNDS TERRIBLE. WHEN I LISTEN TO AMAZON MUSIC MOST LIKELY FLAC SOUNDS BETTER… BUT ITS NOT THE FLAC , ITS RUNNING THRU MY YAMAHA A6A… BETTER DAC…
The FLAC format is basically a wrapper that holds the PCM data within.
Similar to how when you zip a file, you end up with a .zip file, but your original file is contained within. Once unzipped, you will have back your original file, down to the last byte. Nothing will be different.
The zipping process does compress your file, but that compression is lossless.
The FLAC format compresses the PCM content, and does so losslessly.
In addition to the benefit of saving storage space (due to the compression), the FLAC format allows for nearly endless metadata to be tagged to your music file (such as "Artist", "Album", etc, and you can make up any number of custom tags, too).
If you do not need the extra tagging that FLAC offers, you can go with the Windows .wav standard. It offers only basic tagging, but it is uncompressed.
The PCM data within is identical to the PCM data that is in a FLAC file.
So for anyone that is curious about whether or not the additional CPU cycles that are required to uncompress a FLAC during playback, you can convert your FLAC file to a .wav file, and listen for any differences. Unless you are using ancient hardware (or driving a late model Honda), you should not hear any difference between a .wav file and a .flac file.
My Honda Accord has massive jitter issues with FLAC files.
So I converted my library of FLAC files to .wav files on a USB flash drive, and now they play in my Accord with no issues (except that the stereo kind of sucks).
I am not, however, able to search through my .wav files (tagging issue). With FLAC files I was able to search, but the songs had massive jitter.
Way to go Honda! They used a Panasonic front-end for the stereo, and Pioneer speakers. The stereo is not worth listening to, anyway. Honda dropped the ball on this.
audiophiles don't understand the L in FLAC
@@DWHarper62 Calling our host an idiot is what an eight-year-old would do.
@@DWHarper62 All:
David Harper called Paul McGowan an idiot. That is why I replied to him that that is what an eight-year-old would do.
Rather than David Harper apologizing for her insult, she deleted her "Paul is an idiot" comment.
That, too, is what an eight-year-old would do.
Yes, you can, and some do.
It's the classic convenience vs. quality trade-off.
@@VideoArchiveGuy the tradeoff is laughable bad in 2021 compared to 2001
Your videos very informative love from India 😍
I just received an e-mail debuting your new Aspen speakers. When will you debut them on this channel? They look awesome.
Forgive me if I’m getting the speakers confused, but didn’t show them off about 2 months ago?
ruclips.net/video/bJAYzqCmcsc/видео.html
Oh! I never knew how it all worked. Now, I can breathe easier knowing that FLAC and ALAC are not a big compromise.
It's not a huge difference and it's fine to covert wav to flac or alac. It's close enough. I'm just saying there is a slight differnce that i can hear and my digital meteres measure that difference.
@@380stroker No, they don't, because no such difference exists. They are literally identical. Let me know when you can replicate the result with a double-blind study using your equipment.
Something I've learned with these "high res" recordings. a lot of these services will sell you 24-bit versions whether it was recorded that way or not, if you really like an artist or a label. check with that record label to see what is the highest recording format they actually use. Save your money.
It's all remastered bullshit - otherwise you won't hear the difference over RUclips - the same mastering would sound identical with 44100/16
That’s not easy…. It’s not the recording format…it is the master you have to see… recording, mixing then mastering may all use different quality technically. Lots of audio may still be 16/44.1k but master may be higher ( why ? they put to the mix through analogue gear, compressors, limiters etc ) , so the crappy mix can be bumped up if the mastering engineer puts it through a $80,000 piece of vintage Abbey Road gear…..then the master decides the resolution. Recording may be 16 bit waves but the master may be 24bit.
Hope you guys don't stream music over RUclips? LOL
@@ChiefExecutiveOrbiter …oh I love my mp3s
I like watching these vids, I not a big audiophile but do have interest in sound quality.
With modern internet connections there is no need to deal with lossy codecs for audio streaming. Forget about MP3, MQA and other lossy codecs if you are serious as audiophile. It doesn't take much processing for a modern CPU or SOC to play highest quality lossless audio e.g. 192kHz 24 bits PCM. Digital audio streaming is already top notch today with master quality bit perfect capability but of course if you live in a ghost town in the Nevada desert or on your own island in the Pacific, you will not be able to enjoy it.
pcm takes up a lot of space on the hard disk. I have almost 300 original CDs I have taken on my computer
@@ford1546 Well, one CD can contain up to 680MB. 300 CDs will occupy less than 204GB. A Western Digital SSD with 250GB cost around US$40. Your CD collection is not "a lot of space" anymore. Ages ago I got my first PC with a 20MB hard drive and your CD collection would have been an insane amount of data :-)
Yup, I think the difference isnt very audible. Speaker distortion is vastly higher than compression loss by a factor of over 100x I would think.
@@Canadian_Eh_I Speaker distortion is usually much worse than any other distortion in your system. But of course compression distortion can become the worst in an otherwise great system if you run your audio through some low bitrate decoder e.g. 128kbps or less MP3 is audibly really bad.
@@ThinkingBetter Yeah I notice anything below 320 kbps. But to be honest, even with a very revealing system 44khz sounds great to me. 16 bit is just enough. Ive been playing around with active speakers and mods for along time and I finally came to the conclusion that the speakers are 90+% of what you hear. With very good speakers (proper waveguides) even the room doesnt matter that much. CHeers.
For a second I thought those circles in the background were tortillas
Hello Master. Greetings from Colombia.
unpack the flac, yeah yeah yeah! un pack! the flac!
I’m sure the original questioner meant some format like WAV, or AIFF.
Even though FLAC and AlAC deliver PCM losslessly, there is a time variable involved that may cause jitter or a temporary interruption of that stream. So they are the same assuming that there is no interruption during the unpacking or decoding process, which as you say is trivial on most modern computers. However jitter can still be a problem, even when transmitting data from a streamer to say a DAC due to electrical or EMF interference, since that digital data is still being sent through an analogue carrier wave. Caching of that data and verification of that data after each step could mitigate that jitter, so that your DAC doesn’t start decoding a signal and have interruptions or errors in that signal due to electrical interference. I think that’s why HDMI is the best way to send a digital signal to a DAC, followed closely by USB, and then followed by the worst being SPIDIF. This is due to the time variable component and subsequent electrical noise that can interfere with the digital signal. As for DSD being the “best” that’s only true if all of your source material being recorded is acoustical in nature, as in sticking a microphone or microphones in a physical room to capture a live performance. However, this DSD information is then converted into PCM for mixing and mastering purposes and if you are using digital instruments, like synthesizers or drum machines, that are PCM based, then that difference is lost. Not to mention that DSD DACs employ oversampling and filtering as a matter of functional necessity, which may skew the sound from the original source recording, where as an R2R ladder DAC, has the option, and typically works best without any oversampling (at least internally). So DSD is only “better” than PCM in very special use cases, which in today’s world of how recordings are mixed and mastered only rarely give a better result than say a very high resolution…say 768k PCM recording format. An example where DSD might be better than PCM is in recording a live orchestra, where you are capturing all the instruments and the acoustical signature of the recording hall and the entire “mix” happens in the recording itself simultaneously. DSD was originally invented by the same engineers at Sony and Phillips, that also invented the PCM CD format, and DSD due to it’s insane file size, and inability to mix and master it, was designed to be an archival format for super high resolution digital audio. That’s why it’s not very useful in most modern recording studio situations. Not to mention the file sizes…OMG , for example one song of 96KHZ FLAC can be saved in say 70mb, where as the exact same song in 512 DSD would be 900mb! So file sizes, even in the current day of multi-terabyte hard drives are still very much an issue and a hinderance to having all of your music in high resolution DSD format.
Most modern DACs have ways of handling jitter over USB. They reclock the signal internally. This hasn't been a real problem for 15 years or so.
You can just change the player to decode (in advance) the whole song in memory before beginning playing it. It would be extremely fast on modern machines. Playing it will then be 100% equivalent with WAV.
Paul speaks highly of DSD encoded music. I am streaming music through a PLEX media streamer, and the vast majority of it is "uncompressed" FLAC. It sounds very good to me, so I am assuming "uncompressed" FLAC is essntially identical with the WAV files found on my CDs. Some of the music I stream is high res--from 24/96 to 24/192. My PLEX handles it without problems. So how would I get DSD format tracks? Would PLEX handle them? Wouldmy DAC handle them? I am intrigued by what Paul is saying, but I cannot determine how DSD is relvant to me? Any insights?
@Any RD The vast majority of DACs on the market are delta-sigma DACs, in which case they only handle DSD natively, and if you feed them PCM they have a converter that will do on the fly conversion from PCM to DSD. So no it's never "downgrading" to PCM.
If you have an R2R DAC then it handels PCM natively. The only case is where DSD is converted to PCM is on R2R DACs that have the conversion circuit, many don't. But you really don't want to feed DSD to an R2R DAC. DSD isn't 'better'.
It's better for delta-sigma DACs, because that's their native format, as where PCM is better (sometimes necessary) for R2R DACs.
What's bad is the on the fly conversion.
If you have a DSD capable DAC, then yes; it will play the DSD without conversion.
Software support, on the other hand, is going to be your issue.
@@G3rain1 Modern DACs are using multi-bit delta-sigma. Many don't support DSD input at all.
But PCM puts a higher load on the network by simply being larger and is thus more sensitive for network loads. Not very likely to cause issues for those on good connection but as likely as the increased load on the CPU for handling FLAC.
Those reading this article already should have bandwidth fast enough to stream videos and are likely not having issues with FLAC streaming.
laughable - any network of the past 15 years don't give a shit about some audio stream
And any computer that’s still safe to have online (i.e, still gets security patches) is going to be able to decompress the audio without even shrugging it’s metaphorical shoulders.
@@Harald_Reindl Yes, now I'm streaming a 192kHz 24 bits track from Amazon HD Music and it occupies less than 0.1% of my available 1Gbps optical fiber bandwidth.
@@ThinkingBetter 0.01% :-) besides that in most cases it is buffered anyways - if my storage would be a network device mpd would load the next track and uncompress it completely before start playing - hence all the nonsense about CPU usage for decode FLAC golden ears could hear is exactly that: nonsense
I really wish audiophiles would have a brain and technical understanding - imagination is for the sound stage when listen to music :-)
Nice video!
Whoever uses a high grade audiophile system should not even touch mp3, wma or any other compressed format.
I really want to try DSD audio files but I don't know where to get/buy them
Blue coast records should have free demo songs
@@dagnisnierlins188 I'll check them out. Thanks for the suggestion
Good thing you didn't go to that PCM vs DSD comparison.
Contrary to speculation DSD is not a better format and more time than DSD files are converted to PCM for editing (because there isn't enough available software to edit DSD) so if there were any gain it would be lost in the conversion process.
It depends on what you define as a "better format." PCM is often easier to handle, but when it comes down to pure sound quality, DSD is superior.
the problem w DSD is that you can't edit it without losing information
@@xanderguldie A ‘standard’ DSD file- often referred to as DSD64 is roughly equivalent to a sample rate of 24/88.2kHz. ‘Double DSD’ or DSD128 samples that single bit of information 5.6 million times a second to give you a signal equivalent to 24/176.2kHz.
@@johnholmes912 You can. Pyramix Merging will losslessly edit DSD. Tascam Hi-Res Editor edits DSD without loss.
I think Paul should invent his own codec "P-CMFLAC"
everyday lecture, thank you sir
I find PCM much more detailed and in-depth than flac of the same track.
Maybeby the process of unpacking does something to the sound and make it a little brittle...
No, it doesn’t.
I use Windows Media Audio Lossless. It works.
What's the point of different FLAC compressions (ie. uncompressed FLAC)? If it's lossless why bother with the option of a bigger file size?
the same point as compression levels for zip, bzip, lzma or zstd - space versus CPU load
The new speakers look beautiful. I hope I can afford them some day
FLAC. has previously been incompatible with many players such as car players and older mp3 players. This is due to the flac. requires more processor power and then uses more batteries on the mp3 player.
win7 and win8.1 win10 I do not think support flac. before installing software or used winamp.
all new mp3 players and most mp3 players on ebay now support flac.
Do not know with players in newer cars as they are a few generations behind car players you can buy in the store.
I only use flac. and you do not save 50% on the file size when you put the flac on the recommended compression.
If you want to be 100% sure that everything can play the music then you should use wav but it takes up a lot more storage space.
pcm. and flac. has the same sound during playback. here it depends on the software that plays the file and the codec being used.
When you need to transfer wav. through a network it requires more capacity since wav is larger files than flac is
the network don't give a shit about laughable tiny audio transmission
FLAC is great, using it for years. However, FLAC CAN be compressed to varying degrees and I've seen compressed songs being streamed.
It is still lossless. It just require a little more power from the CPU to unpack. Sound quality is the same.
The „varying degrees“ of compression only effect the size of the file, not the quality. The level of compression simply determines how much work the encoder is allowed to do in order to find an economic packaging of the data stream. On the producer’s side, a higher level of compression comes with a longer processing time and a higher energy consumption. On the consumer’s side, the level of compression doesn’t matter in any way.
I use to like this guy until he started pushing expensive power cables.
CAPS 2 C NOT SHOUTING…. PCM VS FLAC, IS JUST 1/2 THE STORY… WHEN I PLAY MUSIC THRU MY APPLE TV I GET PCM, SOUNDS TERRIBLE. WHEN I LISTEN TO AMAZON MUSIC MOST LIKELY FLAC SOUNDS BETTER… BUT ITS NOT THE FLAC , ITS RUNNING THRU MY YAMAHA A6A… BETTER DAC…