6 ms·

Don’t be fooled. 24-bit will not fix computer audio

72 points by bensummers 16y ago

ugh 16y ago
We have reached the limits of human hearing, to believe that tweaking the sampling rate or bit depth will improve audio quality is foolish. We should use the 16 bit we have better – that’s the real problem. 24 bit won’t help with that.
- Bud 16y ago
  We did reach the limits, or nearly, but then we backed way off from the limits, and stayed back. 128kbps MP3 is most certainly NOT reaching the limits of human hearing, for instance. So to the extent that a push for 24-bit files will result in higher sampling rates in the kinds of files most folks are listening to, yeah, it could be a big help. Especially if you are talking about my field, classical music, where the dynamic range involved is much wider and the variety of sounds involved is much greater. Will it be the 24-bitness that causes the improvement? Not directly, no.
  - ugh 16y ago
    I was talking about 16 bit, 44.1 kHz, not 128 kbps MP3 [0]. Oh, and the sampling rate has nothing to do with the bit depth. 44 kHz gives you frequencies up to 22 kHz, that’s at least two kHz better than great human hearing. I don’t see any possible reason why a higher sampling rate would be necessary. I also would like to know what you mean by 24 bit causing an improvement indirectly. How so? [0] I attended a university lecture with one of its main inventors and he told us in no uncertain terms that 128 kbps MP3 is definitely not CD quality and also that MP3 has certain artifacts (castagnettes are its enemy) that cannot be remedied by simply picking a higher bit rate. Lossy audio compression is still extremely clever and cool.
    - Bud 16y ago
      Yeah, but my point was, most of us are no longer listening to 16/44.1 CDs. We're listening to MP3, AAC, etc. If we were still listening to CDs directly, there wouldn't be an issue. What I mean by the push for 24-bit causing an improvement indirectly is, Apple (and others) aren't stupid. They aren't going to just up the size of the sample to 24 bits and leave the sampling rate and overall quality alone. They're doing this to offer much larger files, probably 24/96 files, with either lossless compression or much less lossy compression, so they can a) sell those files for more money and b) justify folks purchasing new audio players and huge amounts of storage to store all their new shiny files. My point is therefore that 24-bit will drive an overall improvement in quality, hopefully to something greater than CD quality.
      - ugh 16y ago
        Ah, ok, that’s understandable. I would, however, be much happier about lossless music that doesn’t needlessly waste storage space. Mobile storage space is still limited, especially after everyone switched to flash memory. (I’m personally happily buying 256 kbps AAC files. I did a blind test before I started investing money and couldn’t hear the difference. Buying and storing lossless files would be kind of pointless for me personally.)
    - Derbasti 16y ago
      Note that the anti-aliasing filters necessary to cut off audio at 22 kHz distort the audible range, too. Also, what they don't cut off gets introduced into the lowest frequencies as noise (this effect is called aliasing). With a higher sampling rate, you can use less steep anti-aliasing filter slopes that introduce less distortion and noise. Furthermore, that distortion can be limited to frequency ranges above the human hearing threshold. Hence, higher sampling rates can make a huge difference. There is no such thing as a perfect, distortion-free filter.
      - ScottBurson 16y ago
        Right. Which is why it's a shame that CDs aren't 16/48 rather than 16/44 -- it allows a filter with half as steep a slope.
        dedward 16y ago
        This is where oversampling comes in, and you get CD players with 196khz 1 bit DACs - we mathematically shift samples to a much higher bitrate (at lower sample width), and then run it through a dac - that allows for a much, much gentler filter, as artifacts are shifted way higher in the spectrum. This is already a solved issue......(even if sales guys a decade or so ago tried to claim it increased resolution - oversampling was all about gentler filters and cleaner sound, not increased resolution, as we know that's impossible)
        Derbasti 16y ago
        Oversampling only improves DACs (CD-plyers, sound cards), it does not solve anything for ADCs (mastering). When mastering a CD, the antialiasing filter still needs to shut off at 22.050 Hz and there is nothing oversampling can do to improve this. You can use really long, linear-phase look-ahead FIR filters that are way better than the analogue IIR filters of the olden days, but the filter slopes still have to cut off between 18-22 kHz, which means some ~60dB+ per octave. A filter like this will always produce audible artifacts. This is pure physics and there is nothing we can do about this.
  - sp332 16y ago
    Is there a major music store that only sells 128-bit MP3 (or equivalent AAC or ATRAC etc.)? Amazon and iTunes sell 320kbps mp3's, and most smaller sites I know of make WAV or FLAC versions available.
- lautis 16y ago
  Digital audio has had the capability to reach the limits of human hearing, this capability has been rarely used. The main problem is, as the article pointed out, that consumers prefer convenience over quality. Making records sound "louder" makes them more suitable for playback in non-ideal environments. While 24-bit audio won't directly help with bad mastering practices, it might indirectly encourage producing of records with higher dynamic range. And, 24-bits should also be enough to reach limits of human hearing in dynamic range, although I personally don't feel the need of reaching the threshold of pain.
  - ugh 16y ago
    I’m not a fan of audio files that needlessly waste space just to encourage stubborn producers.
rythie 16y ago
Does anyone distribute 24bit content anyway? iTunes, Amazon, Spotify etc. I don't think any of them do.
- Bud 16y ago
  Not yet. That's the idea; they will begin to. Apple is apparently already talking about doing so. http://www.macrumors.com/2011/02/22/apple-reportedly-looking-to-offer-24-bit-music-files-in-itunes-store/ http://www.macrumors.com/2011/02/22/apple-reportedly-looking... http://venturebeat.com/2011/02/22/apple-24-bit-music/ http://venturebeat.com/2011/02/22/apple-24-bit-music/ http://articles.cnn.com/2011-02-22/tech/24.bit.music_1_higher-quality-music-online-music-stores-music-catalog?_s=PM:TECH http://articles.cnn.com/2011-02-22/tech/24.bit.music_1_highe...
  - rythie 16y ago
    Interesting, the hype around 24bit could kill off the CD, since CDs can't really compete with that and DVD-Audio and SACD failed to reach the mass market. As a bonus to Apple they will take more space, especially if they use FLAC, so people will buy bigger iPods and iPhones.
    - wladimir 16y ago
      SACD failed to reach the mass market, partly, because most music customers didn't really care about 24 bit audio. At least, not enough to pay more for it. I don't think it's different this time. Music quality is not an issue for most people. If it was, low-quality mp3 and players with crappy headphones, and crappy computer speakers wouldn't be that popular.
      - rythie 16y ago
        It doesn't have to reach critical mass this time though, people who care about quality can download the better version, people don't care don't have to.
      - Bud 16y ago
zandorg 16y ago
I master my songs to 24-bit 192khz - a single master and every multitracked track. Even with 20 tracks at that resolution, each song fits on a DVD-R. As to why I need that resolution, my ancient analogue synth can sound good at that high resolution. As for distribution, I just post FLAC files to my website.
- mambodog 16y ago
  24-bit 192khz No-one will tell the difference between that and Redbook CD quality in a double blind test. For recording, and intermediary files, sure, knock yourself out. For the final master 16/44.1 is plenty.
  - justincormack 16y ago
    Yes, as the article points out, if you don't go mad and apply far too much compression like so much recorded music does. Over compression is hideous...
    - zandorg 16y ago
      I compress individual drum/synth tracks through 'insert' channels. All my mixing is done in the analogue world, and only finally goes into my Emu 0202. I used to compress finished tracks, but I have the attitude that people can apply their own compression. But me compressing individual tracks is fine.
  - JonnieCache 16y ago
    It is nice to have a final mastered render in that high bit/sample rate around for future reference. In this day and age where we regularly resample, re-edit and generally stretch, bend and chop finished tracks, particularly during live performance, it can make a difference. There is a need for it, albeit quite a specific one. EDIT: Also any masters you send to cutting/pressing houses will be re-mastered per medium. So the cutting house will do a vinyl master, a CD master and so on. Vinyl has very specific needs[1] and so there will be many layers of processing AFTER the artist has finished their mastering process. This will benefit from a high bit/sample rate input. Yes, obviously vinyl introduces all of its own (delicious) noise and distortion. However we don't need the vinyl distortion and aliasing of the digital waveform as well. The latter is avoidable. [1] https://secure.wikimedia.org/wikipedia/en/wiki/RIAA_curve https://secure.wikimedia.org/wikipedia/en/wiki/RIAA_curve
EgeBamyasi 16y ago
"The speakers built into a portable computer are most likely a bit hopeless – and it may well be that HPs are better than most – but that is easily overcome by plugging in powered speakers, or using an external digital to analog converter (DAC)" I bought a low end recording audio interface( E-MU Tracker Pre) whitch features a pretty decent DAC and boy oh boy does it make the music pop! Comparing a 192kbps MP3 featuring a big dynamic range with headphones connected directly to the on board audio on my laptop vs headphones connected to my Tracker Pre there is a night and day difference. On the external sound card the MP3 still sounds a little flat, but it does it with class as opposed to boring on the onboard DAC. So before you get into the whole "OH LAWL, MUST ONLY LISTEN TO FLAC" thing start by updating your sound card, its well worth the money!
- Derbasti 16y ago
  Or just connect your digital output to your stereo system. Many people already have the necessary hardware to do this.
  - EgeBamyasi 16y ago
    Granted that the reciever has a decent DAC. :)
- juiceandjuice 16y ago
  Actually, a lot of laptops have decent DACs. My thinkpad has an AD1984: http://www.analog.com/static/imported-files/data_sheets/AD1984.pdf http://www.analog.com/static/imported-files/data_sheets/AD19... With a SNR of 96dB, you should be able to get all your resolution out of anything originating from a CD. HOWEVER, Sometimes recording equipment drives inputs much better than cheaper chips, so you get that difference in sound quality. I have some Audio Technica ATH-M40fs headphones that prefer pro stuff because of their 60 ohm impedance, but even on a laptop they sound great (Which for me means flat) http://www.audio-technica.com/cms/headphones/7c784888146c212e/index.html/ http://www.audio-technica.com/cms/headphones/7c784888146c212... They also handle up to 1.6W power (8 times that of an AKG k240 :)
  - 16y ago
PBenz 16y ago
Unless the dynamic range of the environment you're in is greater than the difference of the dynamic range between 16 and 24-bit, and unless the audio content also possesses that dynamic range, you will never hear the difference. It's still useful to record at 24-bit, though, to give yourself the extra headroom and avoid digital clipping. Recording at 24-bit is basically a no-brainer. Don't even get me started on recording using sampling rates higher than 44.1 kHz. This has been discussed ad-nausem on various audio forums. Mix Magazine (I think) ran a double-blind test to see if people from all walks of life, including experts in the audio industry, could hear the difference between CDs and super-audio CDs. They could not. I believe we have indeed reached the limits of human hearing.
- samatman 16y ago
  The reason to record at sampling rates higher than 44.1 is to be able to pitch and tempo shift with fewer annoying autotune-style artifacts. 2.8 MHz 1-bit audio, clean, may not be distinguishable from 16/44, but add enough signal processing and you'll be happy to have the extra bits around. This is similar to the 8 bits-per-channel problem with GIMP. It's not that you can see the difference; it's that when you factor in post-editing, it's nice to have the extra bit depth.
  - ugh 16y ago
    I think no one doubts that higher bit depth and higher sampling rates are useful when recording and editing, the interesting question is whether you can get away with a lower bit rate and sampling rate when selling the music. I would say that 16/44 is definitely enough if you don’t plan on editing the music in any way. What MP3 has shown is that size matters – even today. Flash memory is still expensive, storage is still not unlimited, broadband access is still not available everywhere. Why should we be needlessly wasteful in that last step from producer to consumer? (I’m a bit uncomfortable with that statement. Ideally, it would be possible for us to just buy the raw, not yet re-coded output and a plethora of other formats. Consumers can then just re-encode themselves as needed. 24/96 truly is better – humans can’t hear the difference but if our ears were better we would – there is really no harm done in selling it. But any claims that consumers really need 24/96 audio seem simply wrong to me.)
gallerytungsten 16y ago
When getting into these discussions of bit depth and sampling rate, the debate is endless among audio engineers about the differences. What really makes a difference is the specific gear you use (mics, preamps, processors, a/d converters, clocks, etc.). Debating the numbers without discussing the signal chain is somewhat ridiculous. A fantastic (great gear) 16/44 chain will sound a lot better than a crappy 24/192 chain. The concept of a signal chain "only being as strong as the weakest link" is of great importance. That is only the tracking side of the equation. Eventually, individual tracks get mixed down. Your summing bus is the flip side, and there is a lengthy debate over mixing in the box (digital summing) vs. analog summing, which can be done with a traditional mixing console, or with the recently invented category of gear known as "summing boxes." (There are of course many further variations and permutations.) On top of that, the skill of the engineer will also make a large difference. Those interested in the details may wish to visit a site such as gearslutz.com where these topics are discussed and debated in great detail.
- Derbasti 16y ago
  The artists make the music, not the gear. This is why compressed music is still great music and some people don't care about audio quality.
- juiceandjuice 16y ago
  Gearslutz is okay, but the topics are for more amateur IMO. I think best sites/forums to talk about recording audio are the tapeop forums and electrical audio tech forum. You must be warned though, they (generally) have an "analog is better but I embrace digital" mindset. http://messageboard.tapeop.com/ http://messageboard.tapeop.com/ http://www.electrical.com/phpBB3/viewforum.php?f=5 http://www.electrical.com/phpBB3/viewforum.php?f=5 (You can argue with Steve Albini on here :)
goalieca 16y ago
Computer audio is broken? News to me. I'm happy with the 44.1Khz @ 16-bit.
Derbasti 16y ago
Recently, Apple started talking about 24 bit. Now HP/Dr. Dre is talking about 24 bit. It makes a good story. But fact is, most audio bought online is in some compressed format (mp3,mp4/m4a/aac). Digital audio compression works by reducing the bitrate of certain parts of the music (frequency-time blocks). So 24 bit is nice but audio compression reduces it anyway. So, this is apparently not about compressed music then? It must be about uncompressed music. Well, we can't change the redbook CD standard. I guess these people must be talking about DVD-As and SACDs then? They clearly are not. Now there is one thing that I would actually love to see (but that these people do not seem to be talking about). I would love to buy 24 bit 96 kHz FLAC-encoded music on iTunes. Or maybe not FLAC but Apple Lossless or whatever and maybe not iTunes but Amazon or some new HP thing. I don't care. But Lossless, High-Quality Music in some major online music store. Now that would be something!
- headShrinker 16y ago
  Each song would be about 150MB at 96/24.
  - bdb 16y ago
    That's fine. Apple "sells" HD movie rentals (~3 GB) which are only useful for 24 hours.
    - headShrinker 16y ago
      Good point but I am also concerned about how many songs I can fit on my player. if each song is 150MB, I could fit 30 times less songs. I want higher quality audio if anyone does. My ears are trained and have spent 10 years mixing behind near-field monitors. I have done side by side comparisons with 50MB 44/16 audio and 5MB 256kbps AAC encode, and I can't hear the difference.
verisimilitude 16y ago
Please don't focus on bits, when the [loudness wars][1] are the primary enemy of quality sound _reproduction_ in most (not all) music recordings today. This is about radio. As you sit in your car, tune your radio to your local Top 40 station. You'll notice that, even when you turn down the dial to the lowest audible setting, you perceive a constant drone of music/noise (depending on how you feel about pop music). Now, tune to the local classical station. Little spurts of noise can be heard, punctuated by... quiet spots. The average consumer thinks: "What's wrong with this music?! I have to turn up and turn down my volume all the time!" Connoisseurs of classical music, however, encourage dealing with this high dynamic range, because [dynamics][2] are a critical part of classical music. Here's the sad part: POP MUSIC DOESN'T NEED TO HAVE ITS DYNAMIC RANGE SMASHED! Radio stations can easily take high dynamic range source material and run it through a [compressor][3] to limit the dynamic range, thus making their music more car compatible (solving the classical music 'problem'). However, consumers expect to hear the same when they download an AAC/MP3 and play it outside their car. "What's wrong with this old recording, it's so quiet", is a common complaint. Of course, when iTunes (and competing software) have features like automatic output leveling ([Sound Check][4]), compressing dynamic range at OUTPUT and not at MASTERING should be the choice producers make. Yet, the industry persists, making the music louder at the expense of eliminating its dynamic range. They're painting soundscape with a more limited palette (though, doing a surprisingly effective job, given the limitations). [1]: http://en.wikipedia.org/wiki/Loudness_war http://en.wikipedia.org/wiki/Loudness_war [2]: http://en.wikipedia.org/wiki/Dynamics_(music) http://en.wikipedia.org/wiki/Dynamics_(music) [3]: http://en.wikipedia.org/wiki/Dynamic_range_compression http://en.wikipedia.org/wiki/Dynamic_range_compression [4]: http://support.apple.com/kb/HT2425 http://support.apple.com/kb/HT2425
- headShrinker 16y ago
  As a former sound engineer, I agree with everything here. Just one correction 'sound check' doesn't add compression to a song but rather adjust (down) the volume of songs that are perceived to be louder. This is referred to as "normalization". Note: it does not adjust (up) the songs that are perceived quieter if they have peaks that are 100%... because they would clip. To get them louder would require a process of limiting and compression.
- 16y ago
mwsherman 16y ago
Indeed, bit depth is hardly the problem. Uncompressed 16/44 sounds really good, and they can offer that now. The problems are compression and end-user audio equipment. (And they are only problems if the consumer cares.) If the industry wants to tangibly move quality forward, they should simply up the standard bitrate to 320 or beyond, today.
- dedward 16y ago
  I can agree - hearing a well mastered CD on some decent (not super high end, just good) amp and speakers, in a quiet environment, shows there is plenty of detail and dynamic range. My nice amp and headphones show me that even decently encoded MP3s are often limited by things further down the playback chain (dac/amp/speakers/headphones) than the audio format itself.
ajg1977 16y ago
Really, all this talk of introducing "higher quality" digital music is the record companies looking for the next way to keep you paying for content you already own. I can even understand their fear. Until now every format has had a mass market lifespan of ~15-20 years. When people switched, huge numbers of 'library' albums would be sold providing a nice amount of income for very little work. It's hard to imagine people ever paying to replace existing mp3/aac tracks, and worse - you and I can give each of our children a perfectly reproduced copy of our entire music library.
vilhelm_s 16y ago
When the CD was designed, 44kHz at 16bits was chosen because that exceeds the limitations of human hearing. With the introduction of Blue-Ray audio, there have been claims that the added resolution makes it sound better. However, some members of the Boston Audio Society did extensive testing where they compared (A/X/B) high-quality Blue-Ray music versus the same music downsampled on the fly to 44kHz/16bit -- and even after extensive listening on very expensive equipment by expert listeners, it was impossible to tell the difference. The results are reported in: E. Brad Meyer and David R. Moran, "Audibility of a CD-Standard A/DA/A Loop Inserted into High-Resolution Audio Playback", JAES Volume 55 Issue 9 pp. 775-779; September 2007. (http://www.aes.org/e-lib/browse.cfm?elib=14195 http://www.aes.org/e-lib/browse.cfm?elib=14195 -- I read a PDF last summer, but now I can't find a non-paywalled version). At the same time, there is widespread agreement that music released on Blu-Ray Audio sounds better than CDs -- but this is not because of the extra bitrate, it's because the sounds engineers pay more attention to details, and the discs are marketed to Hi-Fi enthusiasts, so there is no pressure to e.g. destroy the dynamic range by over-compressing the sound (which makes it sound superficially better on low-end equipment).
- rdale 16y ago
  "When the CD was designed, 44kHz at 16bits was chosen because that exceeds the limitations of human hearing." No it wasn't, it was designed to be implementable given the technology of the time. Philips thought they were working on a 14 bit system, until Sony changed the spec to be 16 bits. That wasn't because Sony changed their minds about 'the limits of human hearing', it was because they thought they could implement the technology. 30 years later we can implement a bit depth of 24 bits with no problems. According to the Wikipedia article on the history of the Compact Disc (http://en.wikipedia.org/wiki/Compact_Disc http://en.wikipedia.org/wiki/Compact_Disc) the sampling rate was defined for the following reason: "the exact sampling rate of 44.1 kHz was inherited from a method of converting digital audio into an analog video signal for storage on U-matic video tape, which was the most affordable way to transfer data from the recording studio to the CD manufacturer at the time the CD specification was being developed." So again this has absolutely nothing to do with 'the limits of human hearing'. Recording engineers, such as Barry Diament of Soundkeeper Recordings, think that the sound of 24/192 digital matches the output of the live microphone input of their recording desk. Barry Diament doesn't think 16/44.1 is nearly as good. If you buy a recording at a high resolution you can always encode it as an MP3 or ACC so that it fits on your portable player. On the other hand if you buy an MP3 or AAC recording you can't bring back the lost resolution. So if I can fit my entire CD collection onto a cheap 1 TB hard drive, why do we care about how much disk space high resolution 24/96 or 24/192 audio will take up? When you don't have physical media it is trivial for a site to offer a range of resolutions according to the needs of the buyer. If I want 24/96 and someone else is only interested in 128 kps MP3s, then we can both download from the same site and only pay for the quality we need.
harshpotatoes 16y ago
So there is this recent talk about sound quality from computers, and it makes sense to me that dynamic range would be the biggest culprit against good sound quality. However, every now and then I still here people chime in that upgrading your sound card will also increase your sound quality. Being curious, I went in search of sound card reviews. Of the reviews that I can find, and which aren't ten years old, I can only find reviews of the high end cards which obviously come to the conclusion that the high end cards produce high quality sound. Of course, the numbers they show are quite meaningless, because I have nothing to compare them too. Are there reviews of onboard audio anymore? I have a feeling that technology has progressed to the point that all onboard audio is good enough assuming you have decent speakers/headphones, but does anybody else know?
- jmillikin 16y ago
  Sound cards were important a decade ago, when CPUs were not yet capable of rendering complex audio. If you wanted to (for example) remix ten mono channels to 5.1 surround in real time, the only option was a separate dedicated processor. Since then, CPUs have become massively more capable; any reasonable audio processing can be done on a budget desktop. The only real reason to get a separate sound card any more is specialized connectors, but even that's becoming less true -- even my normal, non-fancy motherboard has an actual S/PDIF jack! Of course, even if a third-party chip's output was not measurably different from an integrated chip, there would still be hordes of idiot audiophiles rushing to drop a few grand on one.
  - chipsy 16y ago
    I agree, roughly. At the audiophile level there is some improvement to be expected with S/PDIF in particular(less cable noise). But the situation today is nothing like with older on-board sound where they used the cheapest generic AC97 chips and the output sounded like a tin can. If you're recording or need low latency playback, there's still tons of room for improvement by getting a sound card. For good recordings, you need a preamp stage, you need <50ms latency, and you (often) need more than one track. And the entry price for these features is still relatively high, between $100 and $250.
atlei 16y ago
Anyone have any tips/links to convert my CD's into great quality MP3's (for listening only) ? I've been using VBR, 196, Lame etc, but this was recommended to me years ago... Which encoder/tool/settings do you recommend for Windows (and Mac) ?
- kronusaturn 16y ago
  lame --preset extreme
georgieporgie 16y ago
Never take audio quality advice from someone who sells through Best Buy.
jfm3 16y ago
"Jimmy Iovine Does Not Understand Math." Fixed that title for you.
dfox 16y ago
It may be interesting to note that probably any computer manufactured in last five years is perfectly capable of playing 24bit PCM at ridiculous sample rates (like 192kHz) as even low-end audio codec chips for Intel HDA support that.

72 points by bensummers 16y ago

ugh 16y ago

We have reached the limits of human hearing, to believe that tweaking the sampling rate or bit depth will improve audio quality is foolish. We should use the 16 bit we have better – that’s the real problem. 24 bit won’t help with that.

Bud 16y ago
We did reach the limits, or nearly, but then we backed way off from the limits, and stayed back. 128kbps MP3 is most certainly NOT reaching the limits of human hearing, for instance. So to the extent that a push for 24-bit files will result in higher sampling rates in the kinds of files most folks are listening to, yeah, it could be a big help. Especially if you are talking about my field, classical music, where the dynamic range involved is much wider and the variety of sounds involved is much greater. Will it be the 24-bitness that causes the improvement? Not directly, no.
- ugh 16y ago
  I was talking about 16 bit, 44.1 kHz, not 128 kbps MP3 [0]. Oh, and the sampling rate has nothing to do with the bit depth. 44 kHz gives you frequencies up to 22 kHz, that’s at least two kHz better than great human hearing. I don’t see any possible reason why a higher sampling rate would be necessary. I also would like to know what you mean by 24 bit causing an improvement indirectly. How so? [0] I attended a university lecture with one of its main inventors and he told us in no uncertain terms that 128 kbps MP3 is definitely not CD quality and also that MP3 has certain artifacts (castagnettes are its enemy) that cannot be remedied by simply picking a higher bit rate. Lossy audio compression is still extremely clever and cool.
  - Bud 16y ago
    Yeah, but my point was, most of us are no longer listening to 16/44.1 CDs. We're listening to MP3, AAC, etc. If we were still listening to CDs directly, there wouldn't be an issue. What I mean by the push for 24-bit causing an improvement indirectly is, Apple (and others) aren't stupid. They aren't going to just up the size of the sample to 24 bits and leave the sampling rate and overall quality alone. They're doing this to offer much larger files, probably 24/96 files, with either lossless compression or much less lossy compression, so they can a) sell those files for more money and b) justify folks purchasing new audio players and huge amounts of storage to store all their new shiny files. My point is therefore that 24-bit will drive an overall improvement in quality, hopefully to something greater than CD quality.
    - ugh 16y ago
      Ah, ok, that’s understandable. I would, however, be much happier about lossless music that doesn’t needlessly waste storage space. Mobile storage space is still limited, especially after everyone switched to flash memory. (I’m personally happily buying 256 kbps AAC files. I did a blind test before I started investing money and couldn’t hear the difference. Buying and storing lossless files would be kind of pointless for me personally.)
  - Derbasti 16y ago
    Note that the anti-aliasing filters necessary to cut off audio at 22 kHz distort the audible range, too. Also, what they don't cut off gets introduced into the lowest frequencies as noise (this effect is called aliasing). With a higher sampling rate, you can use less steep anti-aliasing filter slopes that introduce less distortion and noise. Furthermore, that distortion can be limited to frequency ranges above the human hearing threshold. Hence, higher sampling rates can make a huge difference. There is no such thing as a perfect, distortion-free filter.
    - ScottBurson 16y ago
      Right. Which is why it's a shame that CDs aren't 16/48 rather than 16/44 -- it allows a filter with half as steep a slope.
      - dedward 16y ago
        This is where oversampling comes in, and you get CD players with 196khz 1 bit DACs - we mathematically shift samples to a much higher bitrate (at lower sample width), and then run it through a dac - that allows for a much, much gentler filter, as artifacts are shifted way higher in the spectrum. This is already a solved issue......(even if sales guys a decade or so ago tried to claim it increased resolution - oversampling was all about gentler filters and cleaner sound, not increased resolution, as we know that's impossible)
        Derbasti 16y ago
        Oversampling only improves DACs (CD-plyers, sound cards), it does not solve anything for ADCs (mastering). When mastering a CD, the antialiasing filter still needs to shut off at 22.050 Hz and there is nothing oversampling can do to improve this. You can use really long, linear-phase look-ahead FIR filters that are way better than the analogue IIR filters of the olden days, but the filter slopes still have to cut off between 18-22 kHz, which means some ~60dB+ per octave. A filter like this will always produce audible artifacts. This is pure physics and there is nothing we can do about this.
- sp332 16y ago
  Is there a major music store that only sells 128-bit MP3 (or equivalent AAC or ATRAC etc.)? Amazon and iTunes sell 320kbps mp3's, and most smaller sites I know of make WAV or FLAC versions available.
lautis 16y ago
Digital audio has had the capability to reach the limits of human hearing, this capability has been rarely used. The main problem is, as the article pointed out, that consumers prefer convenience over quality. Making records sound "louder" makes them more suitable for playback in non-ideal environments. While 24-bit audio won't directly help with bad mastering practices, it might indirectly encourage producing of records with higher dynamic range. And, 24-bits should also be enough to reach limits of human hearing in dynamic range, although I personally don't feel the need of reaching the threshold of pain.
- ugh 16y ago
  I’m not a fan of audio files that needlessly waste space just to encourage stubborn producers.

rythie 16y ago

Does anyone distribute 24bit content anyway? iTunes, Amazon, Spotify etc. I don't think any of them do.

Bud 16y ago
Not yet. That's the idea; they will begin to. Apple is apparently already talking about doing so. http://www.macrumors.com/2011/02/22/apple-reportedly-looking-to-offer-24-bit-music-files-in-itunes-store/ http://www.macrumors.com/2011/02/22/apple-reportedly-looking... http://venturebeat.com/2011/02/22/apple-24-bit-music/ http://venturebeat.com/2011/02/22/apple-24-bit-music/ http://articles.cnn.com/2011-02-22/tech/24.bit.music_1_higher-quality-music-online-music-stores-music-catalog?_s=PM:TECH http://articles.cnn.com/2011-02-22/tech/24.bit.music_1_highe...
- rythie 16y ago
  Interesting, the hype around 24bit could kill off the CD, since CDs can't really compete with that and DVD-Audio and SACD failed to reach the mass market. As a bonus to Apple they will take more space, especially if they use FLAC, so people will buy bigger iPods and iPhones.
  - wladimir 16y ago
    SACD failed to reach the mass market, partly, because most music customers didn't really care about 24 bit audio. At least, not enough to pay more for it. I don't think it's different this time. Music quality is not an issue for most people. If it was, low-quality mp3 and players with crappy headphones, and crappy computer speakers wouldn't be that popular.
    - rythie 16y ago
      It doesn't have to reach critical mass this time though, people who care about quality can download the better version, people don't care don't have to.
    - Bud 16y ago

zandorg 16y ago

I master my songs to 24-bit 192khz - a single master and every multitracked track. Even with 20 tracks at that resolution, each song fits on a DVD-R. As to why I need that resolution, my ancient analogue synth can sound good at that high resolution. As for distribution, I just post FLAC files to my website.

mambodog 16y ago
24-bit 192khz No-one will tell the difference between that and Redbook CD quality in a double blind test. For recording, and intermediary files, sure, knock yourself out. For the final master 16/44.1 is plenty.
- justincormack 16y ago
  Yes, as the article points out, if you don't go mad and apply far too much compression like so much recorded music does. Over compression is hideous...
  - zandorg 16y ago
    I compress individual drum/synth tracks through 'insert' channels. All my mixing is done in the analogue world, and only finally goes into my Emu 0202. I used to compress finished tracks, but I have the attitude that people can apply their own compression. But me compressing individual tracks is fine.
- JonnieCache 16y ago
  It is nice to have a final mastered render in that high bit/sample rate around for future reference. In this day and age where we regularly resample, re-edit and generally stretch, bend and chop finished tracks, particularly during live performance, it can make a difference. There is a need for it, albeit quite a specific one. EDIT: Also any masters you send to cutting/pressing houses will be re-mastered per medium. So the cutting house will do a vinyl master, a CD master and so on. Vinyl has very specific needs[1] and so there will be many layers of processing AFTER the artist has finished their mastering process. This will benefit from a high bit/sample rate input. Yes, obviously vinyl introduces all of its own (delicious) noise and distortion. However we don't need the vinyl distortion and aliasing of the digital waveform as well. The latter is avoidable. [1] https://secure.wikimedia.org/wikipedia/en/wiki/RIAA_curve https://secure.wikimedia.org/wikipedia/en/wiki/RIAA_curve

EgeBamyasi 16y ago

"The speakers built into a portable computer are most likely a bit hopeless – and it may well be that HPs are better than most – but that is easily overcome by plugging in powered speakers, or using an external digital to analog converter (DAC)" I bought a low end recording audio interface( E-MU Tracker Pre) whitch features a pretty decent DAC and boy oh boy does it make the music pop! Comparing a 192kbps MP3 featuring a big dynamic range with headphones connected directly to the on board audio on my laptop vs headphones connected to my Tracker Pre there is a night and day difference. On the external sound card the MP3 still sounds a little flat, but it does it with class as opposed to boring on the onboard DAC. So before you get into the whole "OH LAWL, MUST ONLY LISTEN TO FLAC" thing start by updating your sound card, its well worth the money!

Derbasti 16y ago
Or just connect your digital output to your stereo system. Many people already have the necessary hardware to do this.
- EgeBamyasi 16y ago
  Granted that the reciever has a decent DAC. :)
juiceandjuice 16y ago
Actually, a lot of laptops have decent DACs. My thinkpad has an AD1984: http://www.analog.com/static/imported-files/data_sheets/AD1984.pdf http://www.analog.com/static/imported-files/data_sheets/AD19... With a SNR of 96dB, you should be able to get all your resolution out of anything originating from a CD. HOWEVER, Sometimes recording equipment drives inputs much better than cheaper chips, so you get that difference in sound quality. I have some Audio Technica ATH-M40fs headphones that prefer pro stuff because of their 60 ohm impedance, but even on a laptop they sound great (Which for me means flat) http://www.audio-technica.com/cms/headphones/7c784888146c212e/index.html/ http://www.audio-technica.com/cms/headphones/7c784888146c212... They also handle up to 1.6W power (8 times that of an AKG k240 :)
- 16y ago

PBenz 16y ago

Unless the dynamic range of the environment you're in is greater than the difference of the dynamic range between 16 and 24-bit, and unless the audio content also possesses that dynamic range, you will never hear the difference. It's still useful to record at 24-bit, though, to give yourself the extra headroom and avoid digital clipping. Recording at 24-bit is basically a no-brainer. Don't even get me started on recording using sampling rates higher than 44.1 kHz. This has been discussed ad-nausem on various audio forums. Mix Magazine (I think) ran a double-blind test to see if people from all walks of life, including experts in the audio industry, could hear the difference between CDs and super-audio CDs. They could not. I believe we have indeed reached the limits of human hearing.

samatman 16y ago
The reason to record at sampling rates higher than 44.1 is to be able to pitch and tempo shift with fewer annoying autotune-style artifacts. 2.8 MHz 1-bit audio, clean, may not be distinguishable from 16/44, but add enough signal processing and you'll be happy to have the extra bits around. This is similar to the 8 bits-per-channel problem with GIMP. It's not that you can see the difference; it's that when you factor in post-editing, it's nice to have the extra bit depth.
- ugh 16y ago
  I think no one doubts that higher bit depth and higher sampling rates are useful when recording and editing, the interesting question is whether you can get away with a lower bit rate and sampling rate when selling the music. I would say that 16/44 is definitely enough if you don’t plan on editing the music in any way. What MP3 has shown is that size matters – even today. Flash memory is still expensive, storage is still not unlimited, broadband access is still not available everywhere. Why should we be needlessly wasteful in that last step from producer to consumer? (I’m a bit uncomfortable with that statement. Ideally, it would be possible for us to just buy the raw, not yet re-coded output and a plethora of other formats. Consumers can then just re-encode themselves as needed. 24/96 truly is better – humans can’t hear the difference but if our ears were better we would – there is really no harm done in selling it. But any claims that consumers really need 24/96 audio seem simply wrong to me.)

gallerytungsten 16y ago

When getting into these discussions of bit depth and sampling rate, the debate is endless among audio engineers about the differences. What really makes a difference is the specific gear you use (mics, preamps, processors, a/d converters, clocks, etc.). Debating the numbers without discussing the signal chain is somewhat ridiculous. A fantastic (great gear) 16/44 chain will sound a lot better than a crappy 24/192 chain. The concept of a signal chain "only being as strong as the weakest link" is of great importance. That is only the tracking side of the equation. Eventually, individual tracks get mixed down. Your summing bus is the flip side, and there is a lengthy debate over mixing in the box (digital summing) vs. analog summing, which can be done with a traditional mixing console, or with the recently invented category of gear known as "summing boxes." (There are of course many further variations and permutations.) On top of that, the skill of the engineer will also make a large difference. Those interested in the details may wish to visit a site such as gearslutz.com where these topics are discussed and debated in great detail.

Derbasti 16y ago
The artists make the music, not the gear. This is why compressed music is still great music and some people don't care about audio quality.
juiceandjuice 16y ago
Gearslutz is okay, but the topics are for more amateur IMO. I think best sites/forums to talk about recording audio are the tapeop forums and electrical audio tech forum. You must be warned though, they (generally) have an "analog is better but I embrace digital" mindset. http://messageboard.tapeop.com/ http://messageboard.tapeop.com/ http://www.electrical.com/phpBB3/viewforum.php?f=5 http://www.electrical.com/phpBB3/viewforum.php?f=5 (You can argue with Steve Albini on here :)

goalieca 16y ago

Computer audio is broken? News to me. I'm happy with the 44.1Khz @ 16-bit.

Derbasti 16y ago

Recently, Apple started talking about 24 bit. Now HP/Dr. Dre is talking about 24 bit. It makes a good story. But fact is, most audio bought online is in some compressed format (mp3,mp4/m4a/aac). Digital audio compression works by reducing the bitrate of certain parts of the music (frequency-time blocks). So 24 bit is nice but audio compression reduces it anyway. So, this is apparently not about compressed music then? It must be about uncompressed music. Well, we can't change the redbook CD standard. I guess these people must be talking about DVD-As and SACDs then? They clearly are not. Now there is one thing that I would actually love to see (but that these people do not seem to be talking about). I would love to buy 24 bit 96 kHz FLAC-encoded music on iTunes. Or maybe not FLAC but Apple Lossless or whatever and maybe not iTunes but Amazon or some new HP thing. I don't care. But Lossless, High-Quality Music in some major online music store. Now that would be something!

headShrinker 16y ago
Each song would be about 150MB at 96/24.
- bdb 16y ago
  That's fine. Apple "sells" HD movie rentals (~3 GB) which are only useful for 24 hours.
  - headShrinker 16y ago
    Good point but I am also concerned about how many songs I can fit on my player. if each song is 150MB, I could fit 30 times less songs. I want higher quality audio if anyone does. My ears are trained and have spent 10 years mixing behind near-field monitors. I have done side by side comparisons with 50MB 44/16 audio and 5MB 256kbps AAC encode, and I can't hear the difference.

verisimilitude 16y ago

Please don't focus on bits, when the [loudness wars][1] are the primary enemy of quality sound _reproduction_ in most (not all) music recordings today. This is about radio. As you sit in your car, tune your radio to your local Top 40 station. You'll notice that, even when you turn down the dial to the lowest audible setting, you perceive a constant drone of music/noise (depending on how you feel about pop music). Now, tune to the local classical station. Little spurts of noise can be heard, punctuated by... quiet spots. The average consumer thinks: "What's wrong with this music?! I have to turn up and turn down my volume all the time!" Connoisseurs of classical music, however, encourage dealing with this high dynamic range, because [dynamics][2] are a critical part of classical music. Here's the sad part: POP MUSIC DOESN'T NEED TO HAVE ITS DYNAMIC RANGE SMASHED! Radio stations can easily take high dynamic range source material and run it through a [compressor][3] to limit the dynamic range, thus making their music more car compatible (solving the classical music 'problem'). However, consumers expect to hear the same when they download an AAC/MP3 and play it outside their car. "What's wrong with this old recording, it's so quiet", is a common complaint. Of course, when iTunes (and competing software) have features like automatic output leveling ([Sound Check][4]), compressing dynamic range at OUTPUT and not at MASTERING should be the choice producers make. Yet, the industry persists, making the music louder at the expense of eliminating its dynamic range. They're painting soundscape with a more limited palette (though, doing a surprisingly effective job, given the limitations). [1]: http://en.wikipedia.org/wiki/Loudness_war http://en.wikipedia.org/wiki/Loudness_war [2]: http://en.wikipedia.org/wiki/Dynamics_(music) http://en.wikipedia.org/wiki/Dynamics_(music) [3]: http://en.wikipedia.org/wiki/Dynamic_range_compression http://en.wikipedia.org/wiki/Dynamic_range_compression [4]: http://support.apple.com/kb/HT2425 http://support.apple.com/kb/HT2425

headShrinker 16y ago
As a former sound engineer, I agree with everything here. Just one correction 'sound check' doesn't add compression to a song but rather adjust (down) the volume of songs that are perceived to be louder. This is referred to as "normalization". Note: it does not adjust (up) the songs that are perceived quieter if they have peaks that are 100%... because they would clip. To get them louder would require a process of limiting and compression.
16y ago

mwsherman 16y ago

Indeed, bit depth is hardly the problem. Uncompressed 16/44 sounds really good, and they can offer that now. The problems are compression and end-user audio equipment. (And they are only problems if the consumer cares.) If the industry wants to tangibly move quality forward, they should simply up the standard bitrate to 320 or beyond, today.

dedward 16y ago
I can agree - hearing a well mastered CD on some decent (not super high end, just good) amp and speakers, in a quiet environment, shows there is plenty of detail and dynamic range. My nice amp and headphones show me that even decently encoded MP3s are often limited by things further down the playback chain (dac/amp/speakers/headphones) than the audio format itself.

ajg1977 16y ago

Really, all this talk of introducing "higher quality" digital music is the record companies looking for the next way to keep you paying for content you already own. I can even understand their fear. Until now every format has had a mass market lifespan of ~15-20 years. When people switched, huge numbers of 'library' albums would be sold providing a nice amount of income for very little work. It's hard to imagine people ever paying to replace existing mp3/aac tracks, and worse - you and I can give each of our children a perfectly reproduced copy of our entire music library.

vilhelm_s 16y ago

When the CD was designed, 44kHz at 16bits was chosen because that exceeds the limitations of human hearing. With the introduction of Blue-Ray audio, there have been claims that the added resolution makes it sound better. However, some members of the Boston Audio Society did extensive testing where they compared (A/X/B) high-quality Blue-Ray music versus the same music downsampled on the fly to 44kHz/16bit -- and even after extensive listening on very expensive equipment by expert listeners, it was impossible to tell the difference. The results are reported in: E. Brad Meyer and David R. Moran, "Audibility of a CD-Standard A/DA/A Loop Inserted into High-Resolution Audio Playback", JAES Volume 55 Issue 9 pp. 775-779; September 2007. (http://www.aes.org/e-lib/browse.cfm?elib=14195 http://www.aes.org/e-lib/browse.cfm?elib=14195 -- I read a PDF last summer, but now I can't find a non-paywalled version). At the same time, there is widespread agreement that music released on Blu-Ray Audio sounds better than CDs -- but this is not because of the extra bitrate, it's because the sounds engineers pay more attention to details, and the discs are marketed to Hi-Fi enthusiasts, so there is no pressure to e.g. destroy the dynamic range by over-compressing the sound (which makes it sound superficially better on low-end equipment).

rdale 16y ago
"When the CD was designed, 44kHz at 16bits was chosen because that exceeds the limitations of human hearing." No it wasn't, it was designed to be implementable given the technology of the time. Philips thought they were working on a 14 bit system, until Sony changed the spec to be 16 bits. That wasn't because Sony changed their minds about 'the limits of human hearing', it was because they thought they could implement the technology. 30 years later we can implement a bit depth of 24 bits with no problems. According to the Wikipedia article on the history of the Compact Disc (http://en.wikipedia.org/wiki/Compact_Disc http://en.wikipedia.org/wiki/Compact_Disc) the sampling rate was defined for the following reason: "the exact sampling rate of 44.1 kHz was inherited from a method of converting digital audio into an analog video signal for storage on U-matic video tape, which was the most affordable way to transfer data from the recording studio to the CD manufacturer at the time the CD specification was being developed." So again this has absolutely nothing to do with 'the limits of human hearing'. Recording engineers, such as Barry Diament of Soundkeeper Recordings, think that the sound of 24/192 digital matches the output of the live microphone input of their recording desk. Barry Diament doesn't think 16/44.1 is nearly as good. If you buy a recording at a high resolution you can always encode it as an MP3 or ACC so that it fits on your portable player. On the other hand if you buy an MP3 or AAC recording you can't bring back the lost resolution. So if I can fit my entire CD collection onto a cheap 1 TB hard drive, why do we care about how much disk space high resolution 24/96 or 24/192 audio will take up? When you don't have physical media it is trivial for a site to offer a range of resolutions according to the needs of the buyer. If I want 24/96 and someone else is only interested in 128 kps MP3s, then we can both download from the same site and only pay for the quality we need.

harshpotatoes 16y ago

So there is this recent talk about sound quality from computers, and it makes sense to me that dynamic range would be the biggest culprit against good sound quality. However, every now and then I still here people chime in that upgrading your sound card will also increase your sound quality. Being curious, I went in search of sound card reviews. Of the reviews that I can find, and which aren't ten years old, I can only find reviews of the high end cards which obviously come to the conclusion that the high end cards produce high quality sound. Of course, the numbers they show are quite meaningless, because I have nothing to compare them too. Are there reviews of onboard audio anymore? I have a feeling that technology has progressed to the point that all onboard audio is good enough assuming you have decent speakers/headphones, but does anybody else know?

jmillikin 16y ago
Sound cards were important a decade ago, when CPUs were not yet capable of rendering complex audio. If you wanted to (for example) remix ten mono channels to 5.1 surround in real time, the only option was a separate dedicated processor. Since then, CPUs have become massively more capable; any reasonable audio processing can be done on a budget desktop. The only real reason to get a separate sound card any more is specialized connectors, but even that's becoming less true -- even my normal, non-fancy motherboard has an actual S/PDIF jack! Of course, even if a third-party chip's output was not measurably different from an integrated chip, there would still be hordes of idiot audiophiles rushing to drop a few grand on one.
- chipsy 16y ago
  I agree, roughly. At the audiophile level there is some improvement to be expected with S/PDIF in particular(less cable noise). But the situation today is nothing like with older on-board sound where they used the cheapest generic AC97 chips and the output sounded like a tin can. If you're recording or need low latency playback, there's still tons of room for improvement by getting a sound card. For good recordings, you need a preamp stage, you need <50ms latency, and you (often) need more than one track. And the entry price for these features is still relatively high, between $100 and $250.

atlei 16y ago

Anyone have any tips/links to convert my CD's into great quality MP3's (for listening only) ? I've been using VBR, 196, Lame etc, but this was recommended to me years ago... Which encoder/tool/settings do you recommend for Windows (and Mac) ?

kronusaturn 16y ago
lame --preset extreme

georgieporgie 16y ago

Never take audio quality advice from someone who sells through Best Buy.

jfm3 16y ago

"Jimmy Iovine Does Not Understand Math." Fixed that title for you.

dfox 16y ago

It may be interesting to note that probably any computer manufactured in last five years is perfectly capable of playing 24bit PCM at ridiculous sample rates (like 192kHz) as even low-end audio codec chips for Intel HDA support that.