Every month someone sends me a “huge” lead patch that sounds like a wasp in a tin can. Thin, fizzy, all top and no chest. And here’s the thing: building a serum 2 supersaw that sounds expensive has almost nothing to do with a secret wavetable. It’s five or six small balance decisions that most people get backwards.
We’ve shipped Serum preset packs at Monosounds since 2022, and supersaws are still the patch type customers break most often when they tweak. Same story every time: detune cranked, blend cranked, width cranked, and then a compressor trying to glue the wreckage.
So here’s the full recipe. Real numbers, real settings, and the two flavors people ask me about most right now: future rave and eurodance.
Why Most Supersaws Sound Thin
A supersaw is just detuned saw copies. The Roland JP-8000 did it with seven voices in 1996. That part is easy.
The part everyone misses is the relationship between detune and blend. Detune sets how far the unison voices spread in pitch. Blend sets how loud those detuned voices are compared to the center voice. Crank both and you’ve deleted the center voice – and the center voice is the note. That’s the thin sound: all shimmer, no fundamental, a cloud of pitch instead of a chord.
My starting values: detune between 0.10 and 0.15, blend around 75%. At blend 100% there is no anchor pitch left. At detune 0.30 your “chord” is really nine chords fighting each other.
And forget the unison 16 myth. Past seven voices you’re mostly adding CPU, not size. Serum 2 is already heavier than Serum 1 on some patches, so spend voices where they count.
The Serum 2 Supersaw Recipe: Two Oscillators, One Octave Apart
One oscillator never sounds big, no matter what you do to it. Every serum 2 supersaw I ship is built on two saws an octave apart, sometimes three.
Start from a clean init patch so old FX don’t lie to you. Then:
Oscillator A is the body. Basic Shapes wavetable, saw frame. Unison 7, detune 0.12, blend 75%, width around 80. This carries the chord.
Oscillator B is the weight. Same saw, octave down (Oct -1). Unison 5, detune 0.08, width 55. Pull its level down until it sits roughly two thirds under Osc A. If you can hear Osc B as a separate note, it’s too loud – it should feel like Osc A grew a chest.
Oscillator C is optional shimmer. Serum 2 gives you a third oscillator, so use it for a +12 layer: unison 5, detune 0.10, width 95, then high-pass it hard around 2 kHz. Automate it in only for drops.
Quick CPU note: three unison oscillators add up fast. Unison 5-7 instead of 16, sensible polyphony, and 2x oversampling keep most supersaw patches comfortable even on big chords.
Stereo Width Without Phase Mush
Width is where cheap supersaws go to die. The mistake isn’t width itself – it’s the same width everywhere, plus three widening effects stacked on top.
Give each layer its own lane. The low octave stays narrower (width 50-60) so the center of the mix stays solid. The main layer sits around 75-85. Only the high shimmer layer gets to go 95-100. Wide highs and focused lows read as “produced”. Wide everything reads as phase soup.
Then pick one widener in the FX chain, not three. Chorus or the Dimension-style expander or a wide reverb – not all of them. Serum 2’s multiple FX buses make this clean: I keep reverb and any widening on their own bus so the dry stack stays punchy underneath. I broke down that routing in detail in our Serum 2 effects guide.
Last step, always: hit mono and listen. If the patch drops 6 dB or the chord smears, your unison width and chorus are cancelling each other. Narrow the low layers first – that fixes it nine times out of ten.
High-Pass the Stack, Layer a Real Sub
Saws with unison detune are terrible at low end. The detuned voices beat against each other below 150 Hz and you get wobble, not weight. So don’t ask the stack to do that job.
High-pass the whole supersaw around 100-150 Hz, 12 or 24 dB per octave. Yes, it will sound smaller in solo. It will sound twice as big in the mix, because now the kick and bass own the bottom instead of fighting unison mud.
Then add the weight back properly: a separate sub layer. Sine or triangle, mono, unison 1, no detune, playing the root note only. A serum 2 supersaw should own 150 Hz and up; the sub owns everything below. Sidechain both to the kick and the drop suddenly sounds like a record.
If you want to reverse-engineer this instead of building it, our free 250+ Serum 2 presets include finished supersaw leads plus matching subs and 808s – open them side by side and check the filter and unison settings.
EQ Notches That Remove the Cheap
Raw saw stacks pile up energy in very predictable spots. Cutting those spots is what makes a supersaw sound expensive – not boosting the highs.
My usual notches, all gentle and wide: 2-3 dB out around 500 Hz (boxiness), 2 dB around 1.5 kHz (honk), 2-3 dB around 3.5 kHz (the harsh, ear-poking zone). If the top end still fizzes, shelve down above 12 kHz by 1-2 dB instead of fighting individual peaks.
Cut first, then judge. If it still sounds dull after the notches, a small high shelf up at 10 kHz is fine. If you want the full frequency map for this, we published our whole EQ blueprint with charts and key frequencies.
Future Rave and Eurodance Flavors
Same skeleton, very different clothes. Here’s how I re-dress the serum 2 supersaw stack for the styles people request most.
| Flavor | Core settings | Character moves |
|---|---|---|
| Future rave | Unison 5, detune 0.09, LP filter swept 400-900 Hz | Dark mids, filter carries the energy, light OTT |
| Eurodance | Unison 7, detune 0.14, filter open, fast attack | Octave stabs, pitch bend into phrases, bright top |
| Big room / anthem | Unison 7 on both saws, detune 0.12, wide +12 layer | Long reverb on its own FX bus, huge sidechain |
| Future bass chords | Unison 5, detune 0.10, OTT after the stack | 7th and 9th voicings, LFO on wavetable position |
Future rave is a supersaw pretending to be a bass. Keep it darker: low-pass around 600 Hz as the resting point, then ride the cutoff with an envelope or manual automation so the filter movement is the hook. Less width, more mono punch. Our Modern Future Rave presets are built exactly on this template if you want the finished versions with macros already mapped.
Eurodance goes the other way: bright, fast, unapologetic. More detune (0.14+), filter wide open, short amp release so the stabs stay tight at 140-150 BPM, and the classic pitch bend from an octave below into phrase starts. We packed that whole sound into our Eurodance presets for Serum 2.
For festival EDM leads somewhere between the two – big, clean, melodic – the Arcadia EDM pack is the reference sound. And if you’re chasing the softer chord version of this, the same stack with OTT and 9th chords is basically future bass; I walk through that in the future bass Serum 2 tutorial.
Serum 2 Supersaw FAQ
How many unison voices does a supersaw need?
Seven per oscillator is the sweet spot – it’s literally the JP-8000 number. Five works when you’re stacking multiple oscillators. Sixteen just burns CPU; blind-test it against 7 and you’ll struggle to hear the difference in a mix.
Why does my supersaw disappear in mono?
Too much width in too many places. Detuned voices panned hard left and right partially cancel when summed. Narrow the low octave to width 50-60, drop one of your stacked wideners, and keep blend at 75% so the center voice survives the fold-down.
Can I make supersaws in Serum 1 or Vital instead?
Yes – the core unison math is the same, and Vital’s free tier covers it. But Serum 2’s separate FX buses and third oscillator make the wide-top-narrow-bottom routing much easier. And since Serum 2 is a free upgrade for Serum 1 owners, there’s no real reason to stay behind.
My settings match yours but it still sounds cheap. What now?
Then it’s usually not the patch. Weak chord voicings, no sub layer, or a widener stacked on a widener are the usual suspects – I collected the most common ones in Serum 2 beginner mistakes. Fix the voicing first: spread chords across two octaves and the same patch grows up instantly.
Steal the Finished Versions
Everything above takes maybe twenty minutes to build once you’ve done it twice. If you’d rather start from working patches, grab our free 250+ Serum 2 presets – leads, subs, chords and more, royalty-free, no strings attached. Tear the supersaws apart, check the unison and EQ moves against this article, and then break them into something of your own.
I run Monosounds.studio and make the packs myself. Ten-plus years of producing, 97+ releases, 16,000+ downloads. Every Serum 2 preset we ship passes a CPU check on a mid-range laptop and gets used in a real track before release. If it does not survive a mix, it does not go in the pack.
More about Maxim & the studio →
