Best equalizer settings for rock and metal

Start with the Rock preset: +6 dB at 60 Hz, +3 dB at 230 Hz, −2 dB at 910 Hz, +4 dB at 3 kHz, +5 dB at 14 kHz. Lifted ends, a small scoop in the middle. Rock and metal mixes are crowded in the mids, so you take a little out of the middle and give the kick drum and the cymbals somewhere to sit. Those are the exact numbers on the five-band equalizer, and you can hear them on your own file in about four seconds. If your track is a modern, squashed metal master, skip to the smaller numbers below — the full curve will just make it clip.

The rock curve in numbers

These are peaking filters at Q = 1.0, so each one is roughly 1.4 octaves wide. The 60 Hz band is really doing something from about 38 Hz to about 95 Hz. Nothing here is surgical, and for genre listening that's the point.

The lift is bigger than it looks. +6.02 dB is exactly a doubling of amplitude, so +6 dB at 60 Hz roughly doubles whatever low-frequency energy sits near that band. Stack +5 at 14 kHz on top and the file's peak level climbs too, which matters in a minute.

What each band is actually doing

60 Hz, +6. The punch band. A kick drum's fundamental usually lands between 50 and 80 Hz, and a drop-tuned guitar's low string is down there as well (drop C is about 65 Hz). Lifting it is why the Rock preset feels physical instead of polite. On laptop speakers it will do close to nothing, because those drivers can't move enough air that low. Headphones or a real speaker, or don't bother with this band.

230 Hz, +3. Body. This is where a snare stops being a click and becomes a drum, and where rhythm guitars get their chest. Cut it and a rock mix goes thin and papery. Push it past +6 and everything turns muddy, especially on closed-backs that already have a broad lift through the low mids.

910 Hz, −2. The small one that does the most work. With four guitar tracks panned wide, their upper-mid harmonics pile up between roughly 700 Hz and 1.5 kHz and turn into a nasal, cupped-hands honk. Taking 2 dB out separates the guitars from the vocal without sounding scooped or hollow. People leave this band flat, boost bass and treble, and then wonder why their EQ sounds worse than the original.

3 kHz, +4. Bite. Pick noise, snare crack, vocal consonants. Human hearing is at its most sensitive somewhere around 2 to 4 kHz, so +4 here reads as a much bigger change than +4 anywhere else on the curve.

14 kHz, +5. Air. Cymbal wash, room tails, the top of hi-hats. It opens up an old CD rip. It also raises tape hiss, lossy-codec artifacts and sibilance, because those live up there too.

Rock EQ for headphones and earbuds

The best EQ for rock on headphones depends on what the headphones already do. Two bands cause almost all the trouble.

If your pair is bright — most open-back studio headphones, plus a lot of budget in-ears tuned to sound "detailed" in a shop demo — the +4 at 3 kHz and +5 at 14 kHz land on top of a peak that's already there. Ten minutes in, the hi-hats spit and the S sounds hiss. Drop 3 kHz to +2, 14 kHz to +2, and take 910 Hz down to −3. Leave the low end alone. Same punch, none of the glare.

Small speakers have the opposite problem down low and the same problem up top. Most laptop speakers give up somewhere between 200 and 400 Hz, and cheap earbuds fall away below about 100 Hz, so the 60 Hz boost is mostly wasted on them. Put the energy into 230 Hz instead: +4 at 60, +5 at 230, −2 at 910, +3 at 3 k, +2 at 14 k. It fakes weight using the harmonics a small driver can actually reproduce.

The fast way to test any of this is the level-matched A/B button. It flips between processed and untouched, and it trims the processed side back to the original's peak first, so you're comparing tone instead of picking whichever is louder. Louder always wins a blind test otherwise. Take the loudness difference away and harshness becomes obvious in a second.

Metal EQ settings: less bass, more separation

Metal masters tend to arrive already scooped and already bright, so the full Rock curve on top tips into mush plus glare. Start here:

• 60 Hz: +3. The kick is usually clicky and triggered already. Big low boosts blur it into the bass guitar.
• 230 Hz: +2. Enough to stop the rhythm guitars sounding like a wasp in a jar.
• 910 Hz: −4. The deepest cut on this page. Dense down-tuned guitar layers generate more honk than anything else in music.
• 3 kHz: +3. Keeps the pick and the double-kick beater audible through the wall.
• 14 kHz: +2. Modern metal has plenty of cymbal already.

Doom, stoner and sludge want the opposite. Those records are built on low mids: try +5 at 60, +4 at 230, 0 at 910, +2 at 3 k, +1 at 14 k. If all you want is more low end and the rest of the curve doesn't interest you, the dedicated bass booster is simpler. It's one low shelf at 120 Hz, 0 to +12 dB, and everything under the shelf comes up together.

Modern masters have almost no headroom

Here's the part most EQ guides skip. A rock or metal album mastered in the last twenty-odd years typically peaks within a fraction of a dB of full scale on nearly every track. There's no room above the waveform. Add +6 dB at 60 Hz and any peak carrying low-frequency energy runs past full scale, and digital clipping sounds like gravel. It is not analogue warmth.

The booster page tells you exactly how much room you have. Smart Boost scans every sample and reports sample peak in dBFS, RMS in dBFS, integrated loudness in LUFS and headroom in dB, then names the percentage that produces zero clipped samples with 1 dB of margin. If a track comes back with 0.2 dB of headroom, you know before touching anything that a boost needs a matching cut somewhere. One precision note: it measures sample peak, not true peak (dBTP), so inter-sample peaks after a lossy decode can sit a little above the number it shows. Part of why the margin is there.

The soft limiter catches the rest. It's on by default: a tanh-knee waveshaper that does nothing measurable below 0.7 of full scale and bends smoothly toward the ceiling above that instead of chopping peaks off square. It's a limiter, not a compressor, so there's no attack or release to set. Leave it on for rock.

Live albums and older recordings need different numbers

A 1971 record cut for vinyl usually has restrained low end — deep bass had to be kept in check and summed to mono or the stylus would jump the groove — and a top end that thins out above roughly 15 kHz. There, +5 at 14 kHz mostly boosts hiss. Try +6 at 60, +4 at 230, −2 at 910, +3 at 3 k and 0 at 14 k. You get the weight back without dragging the noise floor up with it.

Live albums have their own issue. Crowd noise and room reflections sit around 500 Hz to 1 kHz and the vocal often gets buried under them. Deepen the cut to −4 at 910 and push 3 kHz to +5. The singer steps forward, the crowd stays behind the band.

Loudness-war reissues of 1970s albums are the hardest case, bright and squashed at the same time. Use the metal numbers, not the rock ones, and check headroom before you start.

Boosting a quiet concert video

Phone footage of a gig is usually quiet, mid-heavy and clipped in the bass from the mic overloading. The video volume tool loads MP4, MOV and WebM, plays them back with the audio boosted live, and exports the boosted audio as a WAV file. Be clear on what that means: it does not re-encode the video. You get a louder audio track out, and you'd have to mux it back with the picture in a video editor yourself. For the EQ side, a cut of −3 to −4 at 910 Hz plus +3 at 3 kHz does more for phone-mic audio than any bass boost, since the low end on those recordings is usually already distorted and lifting it just makes the distortion louder.

Saving a curve you like

The web equalizer is a scratchpad. Set the five bands, A/B them, then export to WAV at 16, 24 or 32-bit float, rendered offline faster than real time. Nothing leaves your machine — there's no server behind any of it, no account, and you can turn Wi-Fi off after the page loads and it still works. Export is WAV only, no MP3, so a 4-minute track at 16-bit/44.1 kHz stereo comes out around 42 MB.

For settings that follow you around, the iPhone version of Increase Volume and the Android version use the same five frequencies and the same six genre preset values, plus a library that remembers your curve per track. Volume goes to 200 % there and WAV export is included. Neither app has the soft limiter or the level-matched A/B yet, so it's worth dialling a curve in on the web first and copying the five numbers over. And to be straight about it: neither the site nor the app can raise your phone or laptop's system volume. Nothing running in a browser or an app sandbox can. What they can do is make the file itself louder and better balanced before it reaches the speaker.