Can't tell difference between FLAC and high quality MP3/AAC

Rip to flac. You can always make a top-quality mp3 from the flac. You can never make a top quality flac if you rip to mp3.

 

Transcoding from a loss less audio codec to a lossy audio codec can be quite consuming if you have a large music library. It's better, faster, and cheaper just to rip and encode to a lossy audio codec and be done with it. If we're talking about 16 bit 44.1 kHz source material, then it doesn't make much of a difference in sound quality between loss less and lossy audio codecs. Loss less audio codecs should be used for high resolution 24 bit 48 - 382 or DSD 64/128/256 music albums because the higher sound quality will be preserved in tact. This is what I have done.

 

Keep going with FLAC. I hear day and night differences between FLAC and MP3, even 320kbs, on every song.

 

If you can't tell the difference, so much the better. Saves you money. Disk space. General aggravation. What's not to like?

 

FLAC and ALAC are perfect copies of your CDs, but compressed to save space. If something happens to one of your CDs, you can burn a replacement with FLAC or ALAC. Lossy formats, no matter how good, can never create a perfect replacement CD. I used to do this all the time to create "car copies" of various CDs (the original CDs are in storage). If my car copy got damaged, which they usually did, it was easy to make a new one.

Now I have a car that will play lossy files on a USB stick, so I generate AAC files from my FLACs. In the car, they sound as good as the CD and I can fit a ton of albums on one USB stick. Best of both worlds.

 

I'm just gonna parrot what everyone else here is saying, and rip to both: FLAC for storage, and V0 MP3 for on the go and limited space things. In my testing, at the right volume, some albums do sound slightly better in FLAC, but V0 MP3 gets the job done very well for casual listening.

 

I think most people miss the fundamental issue.

If the material is a brickwall limited source master to sound as "LOUD" as any other CD, MP3 or a commercially distributed download, a 96/24 vesrion is not going to sound better than the versions mentioned previously. You might as well buy the cheapest.

However, if the record company does distinct masters (that is masters that preserve the dynamic range of the music,, and publish at high resolution, I think the audible differences can be significant.

The problem is that the consumer really has to dig around to know how the master was created for the particular listening format that they buy.

There's no transparency, which is why there are forums like this to help in that regard.

 

Hi uncredited,

Nothing to be ashamed of here. Many people can't tell the difference between the two. And of those who say they can I'm pretty sure a certain percentage just think they can because they expect to. The way how we perceive audio is unfortunately very much influenced by our expectations. And that is what is troubling you right now. You feel that your MP3s sound good enough, but you expect them to not sound as good (because of what you read on this forum probably) and so that diminishes your enjoyment because you start having doubts.

Modern-day MP3s use very sophisticated coding algorithms based on psychoacoustics. A 320 MP3 sounds great and don't let anybody fool you who says it doesn't. AAC is even better than MP3. The difference between lossless and MP3 cannot be easily explained in words, but you get less exact treble with MP3s and a lessening of the sense of continuity of the music. But they do carry 95% of the musical information that our brains need for casual listening (and definitely not 5% as Neil Young wants to make you believe). In order to detect the differences you actually need to train your ear (your brain actually) to know what to look for and where. A good training for that purpose is encoding MP3s with very low bitrates. The artifacts are very easily discernible with those kinds of MP3s (even without the lossless source as comparison), so you'll learn what the areas are where the codec is doing the heavy lifting. If you then go up bitrate by bitrate, those artifacts will slowly become less, but you will now have an idea of what the "sound signature" of MP3 data compression is.

But I have to agree with everyone else here: ALAC is just future proof, hard disk space is cheap. Most importantly, it will make you worry less

 

Just go with ALAC or FLAC for exact replicas of the original audio file. Then you can convert it to whichever format you want as the lossy formats/codecs get updated.

 

Loss less written out as two words drives me nuts. That implies that there IS loss just less of it while LOSSLESS implies there is no loss at all. It's lossless.

 

You lost me at "VBR".

 

Variable BitRate (as opposed to Constant)...

 

When I'm listening at home on my main system, I prefer to think I'm listening to lossless. It's a much more engaging, enjoyable experience.

 

Yep. Doing that very thing right now.

Taking baby steps with JRiver MC19 and FLAC files in my main system. Mose Allison sounds great.

 

One trick for getting faster rips is to rip in burst mode and verify the rip with AccurateRip. If you have a good ripping drive and your discs are in good condition (not all scratched up) you will find that most discs will rip clean in burst mode. For the discs that don't rip clean in burst mode you'll need to switch to secure mode ripping. EAC can be set to do burst mode. If I've got a big stack of CDs to rip I use burst mode. The ripping goes much faster.

You can rip to ALAC on the PC using either dBpoweramp (it costs money) or CUETools (it's free). CUETools includes a ripper that can do secure rips. It verifies rips against AccurateRip and its own CUETools database. It's a good ripper. Not quite as good as EAC for scratched discs, but does very well. And since it verifies the rips with AccurateRip and CTDB it'll let you know if a rip is bad. For the bad discs you can resort to using EAC.

Ripping to ALAC will likely be more convenient for you if you're doing a mix of Mac and PC for playback and file management.

 

This ^.

I cannot hear the difference but for archival, future-proofing, etc. purposes I rip everything to lossless (ALAC in my case).

 

Shameless plug for jriver.
Since switching to jriver I dont even worry about mp3's anymore. Just store a lossless and let jriver stream a converted on the fly mp3 to your phone. Then blue tooth or headphone it to where ever you are.

One file, lossless, and done. No mp3 player required.

 

The way I feel is there are tracks that I "care" about, then there are some that mean less to me. FLAC for the ones that are "special" to me, mp3 (320) for the ones that mean less to me.

 

Actually, with dBPowerAmp, you can set up a batch mode that will rip a CD to Lossless, then immediately make a lower-res MP3 out of the Lossless and store that in a separate folder. All in one step, all very fast. The advantage of having a backup Lossless copy of your CD collection is: if a natural disaster happens, if the bomb goes off, if there's a fire or some other huge disaster... you can grab your hard drives under one arm and your spouse/partner in the other arm and head out the door. It's much harder to grab hundreds or thousands of CDs.

I just moved, put my CDs in storage for the moment, but have more than 6000 of them with me as Lossless files, spread across three 2TB drives. The drives only cost $225; the CDs cost more than $60,000. And I backed those drives up to two more sets of drives at another $450. So for 1% of the cost of my collection, I have three duplicates. Very handy.

That's very wise advice. I figured out back in 2004, when I embarked on my mass-ripping adventure (12,000 CDs and counting), that if I ripped a CD at 96K, all the horrible stuff I heard was still there at 128K, only not as bad. When I went up to 192kbps, the artifacts were reduced even less. By the time I went to 256kbps, they were almost gone. I made many tests between WAV, FLAC, Apple Lossless, MP3, and AAC, over a period of a couple of weeks, using solo piano music, solo strings, stuff with bells and cymbals, some real torture tests. (The song that the Fraunhofer Institute used as a test for MP3 was Suzanne Vega's "Tom's Diner," which for some reason had the most audible sound problems.)

I finally came to the conclusion that ALAC worked best for me as an archival format, that AAC sounded a little better than MP3, and that even though the artifacts became invisible at 256kbps, I'd go to 320K just because. The guys on the Hydrogen Audio Forum insist that nobody has ever reliably detected the difference between 192kbps and WAV files in a proper A/B/X test. The AES has conducted several tests with large groups of people, and essentially said the differences are trivial but detectable by some at 256kbps.

 

A couple of comments about the future. It's impossible to know exactly what is going to happen in the future in terms of audio. I didn't expect myself to have the gear I have now, or that gear could sound like this. Back in the early 2000s I was ripping some of my CDs to MP3. Hard drives were too expensive back then to rip to lossless, and my gear at the time was not worth it. I wouldn't have heard the difference. I ripped a small part of my collection to MP3. But mostly I mostly I listened to CDs spinning in the CD player. Not a lot of computer playback listening. Computer as source was still in its infancy. Getting gapless playback was still rare. Getting bitperfect playback was even rarer. Spinning a CD was the easiest and most reliable way to get both gapless and bitperfect playback. Unfortunately I kept with MP3 ripping for more years than I ought to of. Eventually I ended up reripping everying to FLAC. Should have done that sooner once the price of hard drives became reasonable.

As for headphone listening in the future, I expect we'll be getting some digital processing advancements over the next decade. Advancements that will make you wish that you had ripped everything to lossless. I'm expecting some headphone audio processing advancements due to virtual reality headsets advancing. The virtual reality headsets are going to provoke the advancement of better HRTF (head releated transfer function), which is a very fancy form of headphone crossfeed. That sort of digital processing isn't likely to be kind to lossy audio. That sort of advanced processing is very likely to make the inadequacies of lossy compression more obvious compared to lossless audio, and high-res lossless audio. With good headphones and good headphone amps and good DACs people are going to be regretting choices to have lossy music libraries. The more advanced HRTF processing to match with virtual reality headsets is going to make MP3 even more obvious than it is now. Even now with basic headphone crossfeed processing the issues with lossy compression can become evident. With more advanced processing possible in the future, especially to match advancements in virtual reality headsets, the limitations of current lossy perceptual encoding are going to become even more evident.

It is really really shortsighted to commit to lossy now. Moreso than it ever was. The future of high quality audio processing, especially headphone listening, is not going to be kind to lossy codecs.

Friends don't let friends do lossy given current and future technology. Ripping to lossy, or buying lossy music, is very shortsighted today.

 

Yikes, that was a word salad. I just finished a growler of stout and am about to head to bed. Late. I'm not at my most coherent right now.

The point I am trying to make is that virtual reality headsets are going to provoke the advancement of better 3D HRTF processing for headphones. That HRTF processing is not likely to be kind to current lossy perceptual encoding, since current perceptual encoding doesn't take into account the full virtual reality 3D audio effect. Future HRTF headphone processing is quite likely to break current lossy audio perceptual encoding models. Doing lossless will future proof you from those advancements.

 

That might as well be vaporware to my ears. Listen, if you're comparing CDs to loss less to lossy, then there are no real differences in sound quality so just choose whatever works best for you. I chose MP3s because I can't tell any differences myself and I like to store a large variety of music. If I had ripped and encoded everything to FLAC, then I'd run out of storage capacity a long time ago. I would also have to keep purchasing larger and faster hard disk drives every year or so just to keep up with my music library. I have always been of the mindset that lossy is the way to go because MP3, AAC, and Vorbis are gaining popularity and the psychoacoustics are mature and very sophisticated to minimize any sound quality differences. What I clearly see is that loss less is way more expensive for virtually no detectable differences. People want peace of mind and I've told my family members, relatives, and friends the same thing and they sure as heck can't tell any differences in their own listening tests. Have you ever tried to backup your loss less music library to a cloud storage service provider? It takes forever. Heck, backing up a large MP3 music library takes weeks if you've got several hundreds of gigabytes like I do. There's only a handful of merchants where you can purchase and download loss less music albums and most artists, bands, and groups don't bother to provide anything more than MP3 or AAC versions of their latest music. That's the other limitation that nobody seems to care about. I just don't see high resolution taking off like a rocket either.

Do what sounds best to your ears. For me, I've given up on loss less completely. I don't want to keep buying the same music in newer formats over again and I don't want to keep buying bigger and faster hard disk drives every year. I'm completely satisfied with MP3s. I still have most of my CDs stored in my apartment and that's my loss less music library.

 

If I had a chance to do it all over again, then I would not choose MP3s. I would have chosen AAC or Vorbis in an OGG container. MP3 is mature and it is very old. I would have chosen AAC VBR at around 224 kbps or Vorbis VBR at 192 kbps to save more disk space compared to MP3s. That said, I don't plan to re-rip and re-encode all of my CDs to FLAC or another loss less audio codec. It takes too much time and I don't like to sit and babysit my computer during the process.

Lossy audio codecs strike the right balance for me and I can tolerate backing up my large MP3 music library to SpiderOak. I can backup 60+ GB per day to SpiderOak which means that it takes one week for me to backup my most recent MP3 albums. That's doable for me. My friend Robert Warren is visiting me tomorrow afternoon and he's bringing his Western Digital My Essential 2 TB USB 3.0 desktop hard disk drive. I plan to spend a couple of hours backing up my MP3 music library to his hard disk drive and I'll return it to him. He's my off-site data backup and he lives 15 minutes away from my apartment. He can't tell the differences between FLAC and MP3 so I told him that I've chosen MP3 because he doesn't want to keep buying bigger hard disk drives every year.

 

For the remainder of this week, I've got my Sennheiser HD-800 and I'm borrowing a Sennheiser HDVD 800. I just ripped one of my most favorite music albums entitled Ann Hampton Callaway's Slow to both FLAC and MP3. I used the maximum level of compression for both audio codecs. I still can't tell the differences between the FLAC and MP3 and I'm not doing a double blind listening test either. The FLAC version takes up roughly 403 MB. The MP3 version takes up 129 MB. My listening sample size is limited to just me, but I still can't hear a difference. I'm sticking with the MP3 version. So, there it is.

 

It's ironic because I've been thinking in my head for the longest time, "Do I grab the wife and my Steve Hoffman Gold Discs or do I grab the wife and my computer?" ..., it really is a tough call and in the end, she would most likely grab me and a couple bottles of good wine!