FLAC/WAV bass response vs MP3

Post Reply New Topic

Has anyone else noticed that losslessly encoded music files like FLACs and WAVs tend to have a better bass response than MP3s? I mean, it makes sense since MP3 is a lossy format, but why do MP3 encoders strip out so much of the low-end, when so many songs and entire genres place an emphasis on it? I mean, high-bitrate MP3s can sound pretty good, but a FLAC or WAV of the same song will often sound much better.

Another thing I've noticed about MP3s is that they tend to introduce a lot of intersample peaks, which often results in clipping. Most of my (admittedly unsophisticated) music collection consists of rock, metal, and electronic music dating from the 1980s to the present day. As such, most of it is mastered rather loudly, and since a large portion of it is in MP3 format, this means that a lot of it is clipped. Most of the time, the clipping isn't audibly noticable to me, but I'll see it on Foobar2000's peak metre, or in the song's waveforms on Audacity when I enable the "show clipping" feature.

Whenever I load up a FLAC or WAV file however, I almost never see any sort of clipping, maybe some peak limiting, but almost never outright 0dBFS clipping. (It would be rather stupid to see in a lossless file, as it would be indicative of poor encoding, re-encoding from a lossy format, or simply bad mastering, but I'm pretty sure I've seen it on some of the CDs I've ripped ).

Have any other audio geeks here noticed anything similar?

i noticed, this is becouse MP3 simply chops off a few % on both extremes of the range, based on the assumption that most ears don't notice this loss (true for the high-end), and that playback devices can't produce these sounds anyway (which was true back when MP3 was developed).

personally, i don't listen to any music where this becomes an issue, and i enjoy the benefits of only needing half the storage for the same amount of hours (like storing twice the hours on the same memory card)

I agree, for the high-end it's definitely harder to notice, even though I tested my own hearing a while ago and found that I can hear into the 18kHz range. IIRC, MP3 cuts things out around the 20kHz range. The low-end, on the other hand, is noticably different to me, particularly on songs that use synth bass or drum machines, which tend to be heavy in the lower end of the bass spectrum. Now, I think newer MP3 encoders may do a better job of handling bass, because I have listened to some where the bass is pretty good, but I still wouldn't consider it a "transparent" alternative to FLAC.

Just curious, but what kind of music do you typically listen to? I've noticed that a lot of rock and metal has the bass poorly mixed, to the point where it's barely noticeable, though the reason why you may not notice as much of a difference could have to do with the hardware you have as well. I don't have an overly expensive set of headphones, mine cost about $40, but they're an over-ear set that seem to place an emphasis on bass frequencies. At first this kind of annoyed me, since they didn't have as crisp of a sound as my prior set which I wore out the cable on, but they kinda grew on me and now I like the sound. My desktop PC doesn't have the greatest audio setup though, as it has a poor-quality DAC with a poorly shielded connection on the front jacks, so one of these days I want to get an external DAC/amplifier setup that I can plug into the optical audio connection at the back.

my main music is environmental and game/movie music; those dont really emphasise basses, but are more around mid-range.
and indeed, my hardware is not top-line either, i use $30 in-ears, but that is deliberate since i rely on my hearing a lot while in traffic (cycling, walking, driving a car), and i want to be able to keep doing that even with music on.


for the record; i do actually notice the high-end MP3 cutoff, on sufficiently high-end hardware, i just listed the reasoning behind the cutoff based on the general populace.

Most low-end audio hardware focuses on reproducing the mid-range, which makes sense since it's easier than reproducing highs or lows, and it works effectively enough as a compromise for casual listeners.

I've found that if I try listening to music while I'm walking around, I lose a large amount of my perception, and I tend to bump into things more easily, cross the street at unsafe times, etc. I'm not blind, but having my ears overwhelmed by music when I need to be aware of what's going on around me is almost like wearing a blindfold. For this reason, I usually avoid listening to music when I'm walking or cycling, as much as I'd really like to be able to.

I don't think I hear highs as well as I did when I was younger, which is strange because I know I was very sensitive to them at one time. Then again, people tend to lose a certain amount of hearing in that area as they get into their adult years.

i only use flac on my computer and even on my phone, causing me to only upload the essential stuff. i can hear the difference and that is enough for me. I'm also a member of the Steve Hoffman Audiophile Forum, which, if you can get past some things, is a great place for discussing music.

I think it also depends on how they were converted to mp3s. if you use a variable bit rate it's going to have more loss than if you use a 320kbps constant bitrate. I can't tell a difference between flac and 320kbps constant but I also don't have a great set of headphones either just a good set!

where are you getting these songs with the low ranges cut off? Are you converting them yourself or getting them off like itunes? I know itunes usually only sells songs that are 256kbps and usually at 198kbps for older albums and songs.

No specific sources really, I don't buy from iTunes or other online music stores. *shot* I do have a fairly large CD collection, and I used to just rip them to 320kbps CBR with Windows Media Player, but nowadays I rip them straight to WAV or FLAC. I still have to go back and re-rip a bunch of my albums though.

I've heard that VBR is supposed to be superior in some ways, since it can use higher bitrates for certain sections of a song that have more detail. IIRC 320kbps CBR really just means a constant 160kbps for each stereo channel, which isn't exactly ideal for music that uses a lot of stereo panning.

I don't exactly have a great set of headphones either, and it can be hard to tell the difference between a FLAC and a good MP3 at times, but it just seems like FLACs have a nicer bass response to me.

Ogg Vorbis is actually smaller than the average MP3 and is lossless. If you want superior sound you need to actually buy vinyl or flac files.Superior audio is imo 32bits and 88200 khz. CD Quality is thus itself lossy because of the diminished bit depth and sample rate

I definitely agree about the low end too btw. FLAC's can be a bit much though. There's an electronic album I have in flac and I prefer it in MP3 to be honest because all the phase cancellation and the deficiencies in the synth programming are really obvious at the higher audio rate

Ogg Vorbis isn't lossless. http://en.wikipedia.org/wiki/Vorbis

I've heard the argument before that vinyl is a superior format, and while it may have a nice sound to it, technically CDs are superior due to their much wider dynamic range. In an ironic twist however, the way CDs are mastered nowadays, the dynamics are compressed to s**t and the volume levels are pushed to a near constant 0dbfs in order to sound louder. Vinyl can't cope with this outrageous abuse of dynamics, as such mastering would make a vinyl record unplayable, so they use more conservative mastering that just happens to end up sounding better. As far as the analog vs digital side of things, vinyl may better represent things when the production is purely analog, but most production nowadays involves some form of digital technology. Of course, recording studios work with sample and bitrates much higher than what CDs use, but it's rare that audio of this quality is made available for public consumption; you either have to deal with the dynamic range limitations of vinyl, or the hard digital limitations of CDs, and lossless FLACs made from pre-CD masters are hard to find for most things.

I've never encountered that problem, though to be fair, my audio equipment isn't exactly the best. I'm generally tolerant of music that has "warts" to it though.

MP3 in general is terrible.

What sucks is that it's the standard. The DRC of most commercial music is completely squashed.

And yeah Vinyl is good but lacks dynamic range and has a fair bit of a 'flawed' feel to it.

I don't get why MP3 lives on myself. It may have been a useful standard back in the 90s and early 2000s when bandwidth was limited, but now it's almost completely unnecessary.

What really annoys me is when older albums are "remastered", removing much of the original sound and dynamics. It doesn't bother me as much for newer music, because excessive DRC is just par for the course nowadays, but for older albums I often prefer to hunt down the original versions with intact dynamics.

I think the main reasons why people like vinyl so much are because of the ritual that goes with playing it, and the very specific type of distortion it introduces that gives it a "warmer" sound. It may be a technically inferior format, but in a way, it's more enjoyable.

You can compress a flac datastream and contain it within an Ogg file. Ogg Vorbis is generally lossy though