The Loudness War, Explained

Updated August 2026 · by Loopin

The “loudness war” was decades of mastering tracks ever louder to stand out on radio and CD — crushing dynamics in the process. Streaming loudness normalization largely ended it: a hyper-loud master just gets turned down to about −14 LUFS, so you lose dynamics for no loudness gain. The modern move is to master for dynamics and translation — and you can hit −14 LUFS / −1 dBTP free in Loopin.

What the loudness war was

For decades, engineers mastered records louder and louder so their track would jump out against the one before it — on radio, on CD, in a playlist. Louder felt better and grabbed attention, so everyone chased it.

The problem: you can only get louder by squashing the peaks and reducing the difference between quiet and loud — the dynamics. So the arms race steadily crushed the life out of masters in exchange for raw volume.

By the peak of the era, some records were so heavily limited that the waveform looked like a solid brick with almost no transients left. Drums lost their punch, and the ear grew tired quickly because there was no light and shade for it to rest on.

Goal: stand out by being louder than the next track
Battleground: radio and CD, where nothing turned you down
Cost: crushed dynamics and fatiguing, flat-sounding masters
Arms race: everyone pushed louder to keep up

Why louder felt like winning

There’s a simple psychoacoustic reason the war happened: at a first listen, louder is often perceived as ‘better’. A track that’s a notch louder than its neighbor grabs attention before you consciously judge it.

On formats with no automatic level control — radio, CD — that perceived edge was real and permanent, so pushing louder made competitive sense at the time.

Louder = more attention on first listen
No normalization on radio or CD to level the field
Perceived punch won A/B comparisons
Competitive pressure to not be the quiet one

How streaming ended it

Streaming changed the rules. Spotify, YouTube, and Tidal apply loudness normalization, playing everything back at a consistent level around −14 LUFS. The whole reason to be loud — standing out — disappeared.

Now a hyper-loud master simply gets turned down to match everything else. You crushed your dynamics, and the platform gave the loudness right back to the quiet track next to you. There’s no prize for it anymore.

Normalization measures loudness over the whole track, not just the loudest moment, so it’s the average energy that gets matched. A dynamic master and a brick-walled one that both average −14 LUFS end up at the same perceived volume — but the dynamic one still has peaks that breathe, and the brick-walled one has thrown them away for nothing.

Spotify, YouTube, Tidal: normalize to about −14 LUFS
Loud masters get turned down to match
No loudness advantage survives normalization
You keep the downside — the crushed dynamics

The trap: you lose dynamics for no gain

Here’s the modern reality in one line: master too loud and you lose dynamics for zero loudness gain. The platform pulls a −9 LUFS master down to about −14 LUFS just like it leaves a −14 LUFS master alone.

So the over-compressed version and the dynamic version play back at the same volume — but only one of them still breathes. Chasing loudness past the target is pure downside now.

Over-loud master: turned down to −14 LUFS, dynamics gone
Dynamic master: plays at −14 LUFS, dynamics intact
Same playback volume either way
Only the dynamic one keeps punch and life

Modern best practice

The winning move today is to master for dynamics and translation, not raw volume. Hit the streaming target cleanly and let normalization do the leveling.

Translation means the master holds up everywhere — earbuds, laptop speakers, a car, a club rig. A dynamic master with controlled peaks travels across all of them, while a crushed one that sounded ‘competitive’ in the studio often turns harsh and fatiguing on real-world playback. Use these targets:

Loudness: about −14 LUFS, the streaming normalization target
True peak: −1 dBTP ceiling (−2 dBTP for Amazon Music)
Priority: dynamics and translation over sheer volume
Don’t slam the limiter chasing loudness you’ll never keep

Hit the target step by step

Here’s the fast path to a modern, dynamic master that plays back at full volume without the crush:

Bounce your mix to WAV at full quality with peaks around −6 dB of headroom.
Open Loopin’s free mastering tool and drop the file in.
Let it target about −14 LUFS with two-pass loudness targeting, keeping dynamics intact.
Confirm the true peak is capped at a −1 dBTP ceiling.
A/B against your original — check it’s louder and cleaner, not just squashed.
Download the 16-bit WAV, ready for streaming normalization.

Common mistakes

Still chasing maximum loudness — streaming turns it down to −14 LUFS anyway.
Slamming the limiter to win an A/B that normalization erases.
Judging masters at unmatched volume — louder always ‘wins’ until you level-match.
Ignoring true peak — a loud master with no −1 dBTP ceiling clips after encoding.
Mistaking loud for finished — dynamics and translation matter more than a big number.

Master for dynamics, free

Loopin’s online mastering uses two-pass loudness targeting to about −14 LUFS and a true-peak limiter at −1 dBTP — so you hit the streaming target with dynamics intact instead of crushing them for a loudness edge that normalization erases. It’s free, in your browser, nothing uploaded, no watermark. Drop in your mix, A/B it at matched level, and download a 16-bit WAV.

Frequently asked questions

What was the loudness war?

It was decades of mastering tracks ever louder to stand out on radio and CD, which crushed dynamics in the process. Louder felt more competitive on formats that didn’t turn anything down, so everyone kept pushing volume higher.

Is the loudness war over?

Largely, yes. Streaming loudness normalization — Spotify, YouTube, and Tidal to about −14 LUFS — ended it. A hyper-loud master just gets turned down to match, so you lose dynamics for no loudness gain.

Why do loud masters get turned down?

Streaming platforms normalize playback to a consistent level around −14 LUFS. A master that’s much louder gets turned down to that target, so the extra loudness disappears while the crushed dynamics remain.

What loudness should I master to now?

Target about −14 LUFS with a −1 dBTP true-peak ceiling (−2 dBTP for Amazon Music). Master for dynamics and translation rather than raw volume, since normalization removes any loudness advantage.

How do I master for dynamics for free?

Run your mix through Loopin’s free browser mastering. It uses two-pass loudness targeting to about −14 LUFS and a true-peak limiter at −1 dBTP, keeping dynamics intact, and exports a 16-bit WAV with nothing uploaded.