Learning how to make a riser in Serum comes down to four moving parts: a source you can brighten, an upward filter sweep, a small amount of pitch movement, and a volume curve that arrives right at the drop. The plain version takes about ten minutes inside the plugin. The part that makes it sound professional is the last two bars, where most tutorials stop.
Everything below is written for Xfer’s Serum in either version, because the oscillator, filter and LFO routing is identical in both. Serum is a VST or AU plugin, so the same patch works in FL Studio, Ableton Live, Logic or anything else you already own. No sample pack required.
If you only remember one thing: a riser is not loud, it is expectant. The drop only hits hard if the riser made the listener wait.
What You Need
You need a Serum instance in your DAW and about twenty minutes the first time. Serum ships with everything the recipe below requires, so there is nothing to buy or download.
- A Serum instance on its own instrument slot. Serum is a VST on Windows and macOS and an AU in Logic.
- The four modules that matter: the Noise oscillator, Filter 1 and Filter 2, at least one LFO, and the volume output fader.
- A tempo-synced DAW. Serum’s LFO sync and quantize read from your project tempo, so the riser keeps working when you change BPM.
- Headphones or monitors. Risers live in the top octaves and in stereo width, where laptop speakers hide problems.
- A limiter on the master only if your DAW does not already have one. Risers stack volume automation, distortion and reverb, and they clip faster than you expect.
That is genuinely it. If you want the parameter numbers up front, here is the short version, and the steps below explain why each one moves.
| Module | Parameter | Starting value | Where it goes |
|---|---|---|---|
| Oscillator B | Noise, Color White, Oct 0 | Level -6 dB | Straight to Filter 1 |
| Filter 1 | Lowpass, 24 dB/oct | Cutoff 400 Hz to 14 kHz | Automated upward, exponential |
| Filter 2 | Highpass, 24 dB/oct | 40 Hz to 300 Hz | Only for the tonal recipe |
| LFO 1 | Free, 0.5 Hz, or Sync 1/1 | Amount 15 percent | Routed to Filter 1 cutoff |
| Envelope 2 | Attack 8 ms, Decay 3.5 s | Used as LFO 1 shape | Shapes the filter curve, not volume |
| Osc A pitch | Semitones | +3 st over the build | Exponential curve, never linear |
| Output fader | Volume dB | -14 dB to -4 dB | Linear ramp, final beat pushed |
| FX Distortion | Drive | 0 to 18 percent | Ramped with the riser |
How to Make a Riser in Serum: Step by Step
Follow the five steps in order and you will have a finished riser at the end of step five. Each step states what to do and how to tell it worked.
1. Start with a Sound That Can Rise
Click the B oscillator button in the top bar and switch it from Osc to Noise. Serum gives you a Color knob for the noise type and an Oct knob for tuning it, so set Color to White and Oct to zero, then leave Oscillator A silent.
Now set the starting level low, around -6 dB on the oscillator, and pull Filter 1 down to roughly 400 Hz with the lowpass mode selected. Starting dark with plenty of headroom is what lets you open the filter later without the volume jumping.
You can tell it worked if the sound is a soft hiss rather than a hiss that already makes you reach for the volume. If you want grit instead of clean noise, load a one-shot into the noise oscillator slot rather than using the built-in White noise, and pick something with a long, noisy tail so it has material to climb through the filter.
| Riser type | Oscillator source | Filter movement | Pitch automation | Best for |
|---|---|---|---|---|
| White noise sweep | Osc B set to Noise, White | Filter 1 lowpass, 400 Hz to 14 kHz | None, or under 2 semitones | Any genre, the safe default |
| Tonal riser | Osc A saw plus a low sub layer | Filter 1 lowpass plus Filter 2 highpass rising | +3 to +5 semitones | Trance, progressive, breakdowns |
| Rhythmic riser | Plucked saw or short-decay square | Cutoff sweep with rising resonance | +3 semitones, repeating pattern | Techno, drum and bass builds |
2. Build Rising Tension with Pitch and Filter Movement
The filter is what makes a riser read as rising. On Filter 1, set the mode to lowpass at 24 dB/oct, then draw an automation line from about 400 Hz up to somewhere between 12 kHz and 16 kHz across the length of the build. Set that line to an exponential shape, because a linear sweep arrives too early and then sits there.
Add a small amount of pitch next. Automate Oscillator pitch from 0 up to about 3 semitones over the same span, also on an exponential curve. Three semitones is nearly always enough, and going further is the single most common reason a riser sounds cheap.
For the tonal recipe, turn Oscillator A on with a saw wave at low level, and sweep Filter 2 as a highpass from about 40 Hz up to 300 Hz. That removes the low end early on and restores it later, which is what makes the sound appear to gain weight as it climbs.
You can tell it worked if the riser gets brighter and taller without getting quieter. If the pitch rise is doing all the work and the filter barely moves, turn the pitch down.
3. Add Movement with Modulation
A riser with only a filter sweep sounds mechanical. Modulation is what makes it feel alive. Turn LFO 1 on, set it to Free at around 0.5 Hz for an organic wobble, or Sync to 1/1 for a slow pulse that locks to your grid, and route it to Filter 1 cutoff with the amount near 15 percent.
Use Envelope 2 as the shape for LFO 1 if you want a one-shot rise instead of a repeating one. Set Envelope 2 attack to about 8 ms and decay to roughly 3.5 seconds, then assign it as the LFO shape. Serum can use envelopes as LFO shapes, which turns the modulation into a single sweep that happens once and stops.
For the tonal recipe, route a second LFO to Oscillator A’s wavetable position. Turning on Randomize on a pitched LFO gives you a slightly different shape every time you trigger the patch, which is useful when the same riser plays in several places in a track.
Map the modulation amount to a macro so you can push the movement live. Quantize the LFO phase to the host bar if the wobble drifts against the kick.
4. Shape the Energy with Volume and Effects
Volume automation is the part that separates tension from volume. Draw a rising line on the output fader that ends a few dB below where the drop sits, roughly -4 dB against a drop at 0 dB. Push the last quarter of an extra dB into the final beat so the ear expects something.
In the Effects panel, automate Distortion drive upward through the build, ending around 15 to 20 percent. Add a Delay whose feedback rises while its time falls, and automate Reverb size upward so the tail opens out as the riser climbs. A small amount of Chorus gives stereo width that a single filtered noise source cannot.
Watch your headroom. Volume automation plus distortion plus reverb is three things adding gain at once, and a clipped riser distorts the whole mix rather than sounding intense. Sidechain the other elements instead of the riser, so the riser stays forward and the drop still has somewhere to go.
5. Create the Transition and Drop Placement
Finish with a short acceleration in the last bar. Push the pitch automation up another 2 to 4 semitones over half a bar, or speed the LFO up, then cut the volume one bar before the drop so there is a moment of silence. That gap is what makes the impact hit, and it is the step people skip.
Layer a reverse crash or a short noise swell on top for the final beat if you want more weight. Then play the transition in context with the drop, at the volume the track will actually sit at, rather than soloing the riser.
You can tell it worked if the listener winces slightly at the silence. If they do not notice the gap, shorten the riser or move the automation down a couple of decibels.
Common Mistakes
| Problem | Why it happens | Fix |
|---|---|---|
| Riser sounds harsh | Cutoff ends too high with resonance pushed up | End the sweep at 12 kHz, drop resonance under 20 percent, notch 2 to 4 kHz on an EQ |
| Riser is muddy | Noise and a low saw layer both sitting in the low mids | Highpass the riser around 200 Hz on its bus |
| Movement sounds static | Only a filter sweep with no modulation | Add LFO 1 to cutoff, then set an envelope as its shape |
| It drowns the drop | Riser level set louder than the drop and not ducked | End the volume ramp 4 to 6 dB under the drop |
| Sounds cheesy | Pitch rise of 12 or 24 semitones | Pull it back to 3 semitones and let the filter do the rising |
| Doesn’t repeat across BPMs | Free-running LFO and no quantized automation | Sync LFOs to 1/1, quantize to bar, quantize the riser clip to the grid |
Frequently Asked Questions
What is a riser in music production?
A riser is a sound that rises in pitch, volume or brightness over 8 or 16 bars right before a drop, so the listener feels tension building and wants release. In Serum it is built from a noise or tonal oscillator, an upward filter sweep, and automated pitch and volume. A sweep and a riser are usually the same thing; a faller is the reverse.
How many semitones should a riser rise?
Three to five semitones is usually enough, applied as an exponential curve so the movement accelerates toward the drop. Rising 12 or 24 semitones over a short span is the usual reason risers sound cheesy, because the ear reads it as a siren rather than as tension. Let the filter sweep do most of the rising instead of the pitch.
Why does my riser sound harsh?
Harshness usually comes from the cutoff ending too high, resonance being pushed too far, or distortion being added at full level during the whole build. End the filter sweep around 12 kHz with the lowpass at 24 dB/oct, keep resonance under 20 percent, and notch the 2 to 4 kHz range on an EQ after the fact if it is already too spiky.
Can you make a riser in Serum without white noise?
Yes. A tonal riser works well: use Oscillator A with a saw wave at a low level, sweep Filter 1 upward, add a highpass sweep on Filter 2 from around 40 Hz to 300 Hz, and automate pitch by 3 semitones. For something rhythmic, use a short-decay saw with a repeating pattern and raise the pitch in steps rather than a slide.
How do I keep a Serum riser in sync with my project?
Set the LFOs to Sync mode instead of Free, use a rate such as 1/1 or 1/2, and turn on quantize so the phase locks to the host bar. Render the riser to a clip in your DAW, then quantize the clip start to the grid. Doing both means the patch still works if you change the project BPM later.
What is the opposite of a riser?
The opposite is a downfaller, which is the same sound played backwards with the automation reversed, so energy falls instead of rising. Build it exactly like a riser in Serum, automate the filter cutoff downward and the volume downward, then reverse the rendered audio clip in your DAW. Downfallers usually work better without pitch automation, since reversed pitch bends can sound unstable.
Start flat and build upward: get the noise riser audible with nothing but a filter sweep, then add pitch, then modulation, then the volume curve. Most of the time the filter does the heavy lifting and the pitch is the finishing touch, so if a riser feels wrong, pull the pitch back before touching anything else.