Are mp3s (and mp3 players / decoders) all equal?

February 5, 2016 1:32 AM

I've often heard people rate different mp3 players with statements like "the iPods have really poor bass" or "the sound from a Sandisk is really tinny" or "metal sounds awesome on my Zune" (a hypothetical - this has never happened). Obviously speakers and earphones and the ripping / recording of the original track can have a difference. Is there audio difference between players beyond that?

I'm not talking about format (mp4 vs mp3) or compression. If the mp3 standard is well defined, then decoding a track should be ... well, decoding a track. So given the same mp3 track, with the same speakers / earbuds on different players, then "weak bass" shouldn't happen. Is this just audio hoo-haa or are there subtleties in decoding?

posted by outlier to Computers & Internet (11 answers total)

Best answer: Among other components, different MP3 players have different DACs — digital-to-analog converters — that convert digital MP3 signal to audio signal that goes through the headphone output jack. There are variations in the quality and sound characteristics of the output from various DACs, which result in differences in the output. Some who are dissatisfied with the DAC in an iPod have connected the digital output to a third-party DAC, to bypass the internal conversion.
posted by a lungful of dragon at 1:47 AM on February 5, 2016 [4 favorites]

i cannot reliably hear any difference between several dacs (of decent quality) that i own (after trying quite hard with matched levels), so i assumed that an ipod (the last gen of the ones with a real disk) was likely a decent music source (because it's basically a dac) would also be pretty much flat and took one along as source to buy some new speakers.

and it was a disaster. using the ipod as a line level source, it sounded nowhere near as good as some random cd player that we switched to. we're not talking about the kind of bullshitty audiophile details that are difficult to reproduce later, but a general lack of both treble detail and bass - it just wasn't anything like flat across the audible frequency range.

now a line level input is not a headphone, so it's possible there was some issue with impedance matching and the ipod should sound better into headphones. but really headphones are harder to drive than a line input, so i don't believe that.

so my experience was that my ipod is not flat like a good quality dac (these days). in which case you'd expect to hear differences between portable players like that. but i haven't ever sat down carefully and tried.

HOWEVER, i then went and googled, and found this article. the numbers there are pretty much what i would have expected and argue there is NOT a significant difference. so now i am wondering what happened.

(so sorry, i don't have a good answer. i suspect there should not be audible differences, but i was pretty sure i heard one when comparing an ipod to a cd player).

ps this was with rockbox software on the player
posted by andrewcooke at 1:56 AM on February 5, 2016 [2 favorites]

Personal recent experience, YMMV...

I used to have a vaccum tube stereo hooked to a cd system and Klipsch Speakers - the tube stereo amplifier was so tasty and warm. Stopped enjoying music entirely about 5 years ago because that system was lost in trans continent moves and the technology changed...

I put some mp3's on a data stick and plugged those directly into my ancient car stereo system recently - no Bluetooth or iTunes in the way - HOLY TOLEDO!! I thought my system had blown speakers for 5 years! No!! It was the software!!

We recalibrated our home streaming system based on this, sound bar + subwoofer. YEP. Spotify sounds better and richer than Google Play or ITunes, even directly plugged in via the sound plug Jack that works with conventional earphones.

I am convinced now that iTunes SUCKS.

Yes, digital playback has quality issues. I'm convinced certain listening apps are sub quality. I expect to see a return to better audio, shortly. It's nice having everything available, but dropping channels and compressing audio to lower data requirements? Not so wonderful. There is a world of difference. You are correct.
posted by jbenben at 1:58 AM on February 5, 2016

andrewcooke: was it an issue of compressed files vs. uncompressed (CD)?
posted by persona au gratin at 2:05 AM on February 5, 2016

Response by poster: Thanks to all - plenty of interesting replies already. I was totally unaware of DACs and they make complete sense as the variable in the loop.
posted by outlier at 2:23 AM on February 5, 2016

The Wikipedia page on DAC includes this:

There are several DAC architectures; the suitability of a DAC for a particular application is determined by six main parameters: physical size, power consumption, resolution, speed, accuracy, cost. Due to the complexity and the need for precisely matched components, all but the most specialist DACs are implemented as integrated circuits (ICs). Digital-to-analog conversion can degrade a signal, so a DAC should be specified that has insignificant errors in terms of the application.

In other words, there is quite a lot of balancing of design constraints going on. With audio, the DAC function should weigh resolution (lots of sampled bits) over speed - but for a portable device, power consumption and cost are also important considerations - where trade offs may well be made. Finally there are commercial issues: if you spend more money on a better DAC then will people notice it? And what about if you use a rather cheap one?

Beyond that basic level of analysis you are left looking at a lot of deeply technical material concerning the mathematics and engineering involved in implementing the algorithms. And you have to contend with the equally tricky area of getting listeners to agree on what sounds best - based on variations of DAC design alone. No simple solution.
posted by rongorongo at 2:34 AM on February 5, 2016 [2 favorites]

The MP3 format doesn't give the decoder any decisions to make. If two programs sound different on the same hardware, either one is doing extra filters for fun, and/or one is buggy. But yeah, as many people have noted, there's a bunch of analog stuff between the software and the headphones/receiver, which leaves a lot of space to excel or make a mess.

I definitely have noticed issues with treating older iPods headphone outs as a line out. I even got a pocket dock thing to get a proper line out, which sounded appreciably better to me. With newer devices (or maybe it's iPod vs iPhone) I haven't notice nearly as much difference. My guess is they realized people were plugging the headphone jack into line-ins and adjusted the design accordingly.
posted by aubilenon at 2:43 AM on February 5, 2016 [1 favorite]

So given the same mp3 track, with the same speakers / earbuds on different players, then "weak bass" shouldn't happen.

Keep in mind that some mp3 players have optional EQ settings which can affect output.

My iPod Classic has about 20 pre-set EQ options (including 'Off'). Some of them are bass-light, some are bass-heavy, and some are bass-neutral.

My iPod Shuffle has no EQ options at all.

So playing an mp3 through the same speakers/headphones but on different varieties of iPods, users can get different results.
posted by paleyellowwithorange at 3:22 AM on February 5, 2016 [1 favorite]

The NwAvGuy blog has a bunch of useful articles regarding DACs, output and headphone impedance, a handy iPod/Clip+ comparison that basically reveals that both are excellent portable players, and more. Highly recommended reading.

For well-built players, which can be had for less than $50 (see Clip+ review above), the most likely causes of bad sound quality are impedance mismatches, poor headphones, or wrong playback settings (EQ, etc).
posted by Bangaioh at 4:44 AM on February 5, 2016

In addition to the DAC and impedance mismatches that have already been mentioned, the general quality of the analog circuitry that follows the DAC will make some difference. Even things like the jack for the headphones can vary in quality in ways that can cause poor connections and weird sound as a result. In most cases, most people won't notice, but it is there.

Also, even though mp3 is a defined standard, some players do proprietary tricks in their decoding or post processing to attempt to enhance the sound, so there can be variances there.
posted by Candleman at 6:53 AM on February 5, 2016 [1 favorite]

[pag - they were mp3s, but either v0 or 320, and i have not been able to hear any difference between those and flacs, so i don't think it was that. i suspect it was just volume levels plus the disdain of the salesman to me auditioning quite expensive speakers with an ipod as source.]
posted by andrewcooke at 8:13 AM on February 5, 2016 [1 favorite]