Lossless
The audio is kept intact. FLAC saves space without throwing away any of the original sound information.
LISTENING LAB / MUSIC FORMATS
The same music, saved different ways. One keeps the original digital sound intact. MP3 makes the file smaller. Can your ears tell them apart?
No technical knowledge needed. Your phone, computer or headphones are enough to try it. Headphones can make small differences easier to notice.
We use one fixed recording for each song, so the music itself doesn't change between formats.
The numbers show bitrate, or how much sound data an MP3 uses each second. “kbps” means kilobits per second. A bigger number usually keeps more information, but it may not sound different to you.
Want to switch formats without a pause? Prepare all three versions of this one song first. This uses more data and memory. Regular Play remains available without preparing anything.
Every MP3 was made from this same lossless recording. The music was not remixed or remastered. We make a tiny playback-volume adjustment to keep the comparison fair.
Our current controls compare two everyday MP3 settings. A much lower bitrate could exaggerate the difference, but it wouldn't represent how most people listen today.
WHAT DO THOSE WORDS MEAN?
The audio is kept intact. FLAC saves space without throwing away any of the original sound information.
The audio is made smaller by removing or simplifying some information. A good MP3 may still sound almost exactly the same.
They can contain exactly the same sound. FLAC simply packs the digital information more efficiently.
The real question isn't which file has the bigger number. It's whether the difference matters to what you hear.
Both MP3 examples are derived from the same 44.1 kHz lossless source, using LAME constant-bitrate encoding at 128 and 320 kbps. There was no separate mixing, EQ or mastering. The files were checked for timing alignment and sample levels. At 128 kbps (kilobits per second), the decoded sample RMS (an average signal-level measure) was about 0.44 dB lower than the lossless source, so we slightly lower the reference and 320 kbps playback to match the 128 kbps sample level. This is an approximate RMS-matched guided comparison, not a blinded scientific test. Regular switching moves to approximately the same point in the music. Optional instant switching prepares three recordings together and keeps them in step, but takes extra data and may not work in every browser.
One short listening test cannot prove that everyone will hear or prefer the same format. Listening conditions and the recording itself can change the result.
Try the separate ABX Lab with lossless versus MP3 listening challenges. Repeat A, B and X as often as you like.
Open the ABX Lab →THE RECORDINGS
These are audio examples from The Sound Question's existing Listening Labs, using recordings drawn from Extraordinary by Tom McKenzie. The songs and musical performances are real; the mixes or mastering used here are our educational versions, not the artist's official released masters.
Source: AlbumDB research dataset by Thomas McKenzie, Eloi Moliner Juanpere and Alec Wright, listed under Creative Commons Attribution 4.0. The Sound Question independently prepared these educational audio adaptations. No endorsement or partnership is implied. See the AlbumDB research paper (AES Paper 10278) for the original research. These are excerpt-based educational format comparisons, not standalone releases of the source recordings.