> ## Content Index
> Fetch the complete content index at: https://flstudiopro.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# How to Nudge Audio by Milliseconds in FL Studio
- URL: https://flstudiopro.com/fl-studio-nudge-audio-milliseconds/
- Published: 2026-09-01T09:00:00.000Z
- Updated: 2026-09-20T00:30:33.000Z
- Description: Move one clip for feel, align a recorded source with input compensation, or correct a delayed Mixer path without applying the wrong timing fix.
- Author: Eric Norris
- Tags: Workflow

To nudge audio precisely in FL Studio, select the Audio Clip in the Playlist, place the pointer over it, then, with the default nudge setting, use Shift+Alt+mouse wheel on Windows or Shift+Option+mouse wheel on macOS. This moves the Clip in PPQ-dependent steps without changing its length; it is not a fixed one-millisecond increment. If every recording lands late by the same measured amount, fix input delay compensation instead of moving every take by hand.

“A few **milliseconds** late” can describe three different problems: one event needs feel, the recording system places every take incorrectly, or a Mixer path is delayed by processing. The correct control depends on which pattern repeats.

## Choose the timing fix before moving anything

| What is wrong                              | Use                                            | Do not use                       |
| ------------------------------------------ | ---------------------------------------------- | -------------------------------- |
| One clap, vocal word or layer              | Playlist Clip nudge                            | Global playback offset           |
| Every new recording lands late or early    | Audio input delay compensation                 | Manual edits on every take       |
| One plugin path is delayed                 | Automatic PDC first, manual PDC only if needed | Moving source audio out of place |
| The playhead display is globally displaced | Playback tracking Offset after exclusions      | A single Clip move               |
| You want groove                            | Local note/Clip timing choice                  | System latency controls          |

![Clip movement, recording input compensation and Mixer PDC shown as different timing controls](https://storage.ghost.io/c/23/d1/23d1461e-9a6f-4583-9f0a-3ee044fc8192/content/images/2026/08/nudge-methods.webp)

Choose the control for the problem and use positive or negative milliseconds according to the required direction.

## Nudge one Audio Clip with the mouse wheel

1. Zoom into the transient or word you need to align.
2. Select the Clip and confirm you are not in Stretch mode.
3. Place the pointer over the body of the Clip.
4. With the default nudge setting, hold Shift+Alt on Windows or Shift+Option on macOS. If Nudge clips with mouse wheel is enabled, hold Shift instead.
5. Roll the mouse wheel one step at a time.
6. Replay from before the edit and compare the groove.

The [Image-Line Playlist manual](https://www.image-line.com/fl-studio-learning/fl-studio-online-manual/html/playlist.htm?ref=flstudiopro.com) documents two settings. With **Playlist menu > Edit > Nudge clips with mouse wheel** deselected, the default shortcut is Shift+Alt/Option+wheel. With that option selected, use Shift+wheel instead. The pointer must be over the Clip. If the view scrolls rather than the Clip moving, check the pointer position and which shortcut the toggle selects.

## Understand PPQ before calling a move one millisecond

**Timebase (PPQ)** in Project General Settings controls timeline resolution. The [Project Settings manual](https://www.image-line.com/fl-studio-learning/fl-studio-online-manual/html/songsettings%5Fsettings.htm?ref=flstudiopro.com) lists 96 PPQ as the default; a higher value allows finer placement and more Playlist zoom, at a CPU cost. Increasing PPQ preserves existing event placement, but reducing it can move events that fall between the coarser positions. Keep a separate saved version before changing it.

For a constant tempo expressed in quarter-note BPM, one tick lasts **60,000 ÷ (BPM × PPQ) milliseconds**. At 120 BPM, 96 PPQ gives about 5.21 ms per tick; 960 PPQ gives about 0.521 ms. These are timebase calculations, not a guarantee that every wheel notch moves exactly one tick. Neither a notch nor a free drag is a universal “move exactly 1 ms” command.

## Use Snap override for a free drag

When you need to place the Clip by eye and ear, hold Alt on Windows or Option on macOS while dragging to override normal **Snap** behavior. Zooming in increases visual control. This is useful for pulling a doubled layer behind the main hit or placing a breath without changing project-wide grid settings.

Do not trim silence and move audio inside the Clip unless that is the intended edit. Slip editing changes which part of the source appears under the fixed Clip window; moving the Clip changes its position on the timeline. Watch both edges so you know which operation occurred.

## Know what Shift+Left and Shift+Right do

The Playlist also supports Shift+Left Arrow and Shift+Right Arrow to move the current selection by one snap unit. That is a grid-relative shift, not a promise of one millisecond. It is excellent for moving a phrase by a defined subdivision and the wrong tool for a tiny phase or feel adjustment.

The [FL Studio shortcut guide](https://flstudiopro.com/fl-studio-shortcuts/) keeps the broader navigation commands together. In this workflow, use arrows for musical grid steps and the modifier-plus-wheel or Snap override for finer placement.

## Measure a repeated recording offset

Use a repeatable click rather than a person trying to play on the beat. Image-Line describes [recording the FL Studio metronome through a microphone](https://www.image-line.com/fl-studio-learning/fl-studio-online-manual/html/mixer%5Ftrackprops.htm?ref=flstudiopro.com) and comparing the recorded clicks with beat starts. For that speaker-and-microphone test, set **Monitor external input to Off** to prevent feedback and keep speaker level low. Keep the microphone position fixed: the result includes the acoustic path, not just interface latency.

Record several passes through the same path and measure the same landmark on each click. For example, offsets of 6, 7, 7, 8 and 7 ms have a median of 7 ms. Those numbers are illustrative, not a correction to copy. If you cannot obtain a repeatable measurement, do not enter a global value.

If the offsets vary widely, one compensation value will not repair every take. Check the driver and test route before blaming performance. Image-Line notes that [FL Studio ASIO can introduce recording-position jitter](https://www.image-line.com/fl-studio-learning/fl-studio-online-manual/html/envsettings%5Faudio.htm?ref=flstudiopro.com); prefer a manufacturer-supplied native ASIO driver on Windows when your interface provides one. For recorded MIDI, also check input quantizing. The [quantize guide](https://flstudiopro.com/fl-studio-quantize/) covers note correction, not a measurement of audio-interface latency.

At 44.1 kHz, 1 ms is equivalent to 44.1 sample periods; at 48 kHz, it is 48\. An individual sample boundary still occurs at a whole sample index, so 44.1 is a conversion, not a promise that a file has a fractional sample to select.

![Example recording offsets of 6, 7, 7, 8 and 7 milliseconds with a median of 7](https://storage.ghost.io/c/23/d1/23d1461e-9a6f-4583-9f0a-3ee044fc8192/content/images/2026/09/measure-offset-v2.webp)

Measure your own repeatable path; disable input monitoring for a microphone-and-speaker latency test.

## Fix input delay when every recording is displaced

Check [Mixer menu > Disk recording > Latency compensation](https://www.image-line.com/fl-studio-learning/fl-studio-online-manual/html/mixer%5Fmixermenu.htm?ref=flstudiopro.com) before adding a manual offset. It compensates external input latency associated with the ASIO buffer. A remaining error can depend on driver timing; the [audio settings workflow](https://flstudiopro.com/fl-studio-audio-settings/) covers driver and buffer choices. Do not treat a delayed monitoring sound alone as proof that the recorded Clip is misplaced.

For a consistent remaining error, use the Mixer track’s [Audio INPUT delay compensation](https://www.image-line.com/fl-studio-learning/fl-studio-online-manual/html/mixer%5Ftrackprops.htm?ref=flstudiopro.com), not its output PDC control. Values can be entered in ms, samples or beats. Here, positive values move recorded audio earlier and negative values move it later. Leave **Adjust position of recorded audio** enabled normally so plugin latency in that recording path is also accounted for. Record new tests after each change; old Clips do not prove how the next take will be placed.

## Keep Playback tracking Offset global

[Audio Settings > Playback tracking > Offset](https://www.image-line.com/fl-studio-learning/fl-studio-online-manual/html/envsettings%5Faudio.htm?ref=flstudiopro.com) corrects the recorded-note or playhead reference when driver position reporting is wrong. Image-Line says to exclude input quantizing and Disk recording latency compensation first. For a real external audio loop or MIDI-out recording delay, use the individual audio input’s compensation instead.

Do not use a global Offset because one vocal word feels late. It changes the play/record reference across tracks, not just that word. Keep it for a repeatable global alignment problem after checking the driver; it is not a way to rewrite one Clip’s timing.

## Use plugin delay compensation for processing paths

Plugin delay compensation aligns Mixer paths when plugins introduce processing latency. [Automatic PDC](https://www.image-line.com/fl-studio-learning/fl-studio-online-manual/html/mixer%5Fmixermenu.htm?ref=flstudiopro.com) is the default for new projects, but its state is saved per project, so check an existing session. Manual values normally act as offsets to automatic compensation; do not add the full delay again without checking what is already compensated.

The sign rules differ from audio input compensation. For manual track PDC, positive values delay the selected track, while negative values effectively add latency elsewhere to establish alignment. Read the current [Mixer latency documentation](https://www.image-line.com/fl-studio-learning/fl-studio-online-manual/html/mixer%5Ftrackprops.htm?ref=flstudiopro.com) before entering a number.

If routing is unclear, trace it with the [Mixer routing guide](https://flstudiopro.com/fl-studio-mixer/). Moving the rendered waveform may hide one test case while the live path remains wrong.

## Align layers without destroying phase

A kick layer, doubled vocal or parallel microphone can look tighter when its peaks line up, yet sound thinner because their relative phase changed. A timing move does not itself flip polarity; those are different operations, as iZotope explains in its [phase and time-alignment guide](https://www.izotope.com/community/blog/5-ways-to-adjust-phase-after-recording?ref=flstudiopro.com). Nudge while both sources play, then check the combination in mono, listening to low-frequency weight and attack.

Do not chase identical peak shapes between different sounds. Align the perceptual event: the main click, consonant, drum attack or body. Sometimes a deliberate few-millisecond separation prevents cancellation and makes the combined hit wider or clearer.

![Listening checks after an audio nudge: play both layers, check mono and compare the original](https://storage.ghost.io/c/23/d1/23d1461e-9a6f-4583-9f0a-3ee044fc8192/content/images/2026/09/nudge-check-v2.webp)

Compare from the same playback start and at matched levels. This is a listening workflow, not a measured waveform sum.

## Nudge vocals by the phrase, not the waveform alone

For doubled vocals, align consonants that define the word before sustained vowels. Hard-aligning every cycle can remove the natural width that made the double useful. Move only the phrase or syllable that distracts, then use the [shortest clean crossfade](https://flstudiopro.com/fl-studio-fade-in-out/) at the edit boundary.

The [vocal mixing guide](https://flstudiopro.com/fl-studio-mixing-vocals/) owns tone, compression and space. This page owns placement. EQ cannot make a late consonant arrive on time.

## Keep a reversible timing reference

Before a serious micro-edit, save a separate project version and note the original position. If you duplicate the Clip or lane as a reference, keep the reference muted or isolated so it does not play along with the edited copy. Name the comparison clearly; moving the same layer repeatedly without a reference can lose the original groove.

Level-match the comparison and switch between the original and moved positions from the same playback start. If the benefit disappears when the edit is hidden, keep the original. Timing decisions become unreliable when the nudged version is also louder, processed differently or heard from a later lead-in.

In licensed FL Studio, save a separate version and reopen it only after checking any trial-plugin warning. In [Trial mode](https://www.image-line.com/fl-studio-learning/fl-studio-online-manual/html/app%5Ffeature.htm?ref=flstudiopro.com), saving is allowed but reopening requires purchasing FL Studio; render needed audio before closing the session. Even in a licensed host, unowned trial plugins can be replaced when reopening. Preserve the original project and unlock them before reopening; do not re-save over it with missing trial-plugin states.

## Avoid these micro-editing traps

- Do not change global Snap just to fix one event when a temporary override works.
- Do not enter PDC values before proving a Mixer path is the source.
- Do not fix every new recording manually when the same offset repeats.
- Do not call one performer’s late take an interface-latency measurement.
- Do not trust visual peak alignment without listening in mono.
- Do not forget that moving a Clip and slipping audio inside it are different edits.

## Run this timing check

- The problem is classified as local Clip, repeated input, global display or Mixer-path timing.
- The original position is recoverable.
- The edit is measured from a repeatable transient.
- Snap and Stretch modes are understood before dragging.
- Any compensation value was tested with multiple new passes.
- The result works with drums, bass and the full arrangement.
- Layered audio was checked in mono for phase loss.
- Creative groove remains a local musical choice.

Continue through the [Workflow hub](https://flstudiopro.com/tag/workflow/) when the timing is stable. Millisecond control is useful only when the problem has been named first.

## Frequently asked questions

### How do I nudge an Audio Clip in FL Studio?

With the default Playlist setting, hover over the selected Clip and use Shift+Alt+wheel on Windows or Shift+Option+wheel on macOS. Enabling Edit > Nudge clips with mouse wheel changes this to Shift+wheel. Movement depends on PPQ, not a fixed millisecond step. Alt/Option-drag temporarily overrides Snap.

### How many samples are in one millisecond?

One millisecond equals 44.1 sample periods at 44.1 kHz or 48 at 48 kHz. File sample boundaries are whole sample indices. This conversion is separate from the Playlist’s PPQ-based timing resolution.

### Why do my FL Studio recordings always land late?

For recorded audio, check the driver, Disk recording latency compensation and a repeatable test through the actual input path. If several new tests show the same remaining offset, adjust Audio INPUT delay compensation. Variable offsets may be jitter, not one fixed delay. Input quantizing is another check for recorded MIDI.

### Should I move the Clip or change PDC?

Move the Clip when one event needs placement. Use PDC when a Mixer processing path has repeatable latency that automatic compensation does not handle correctly.

### Why does aligned audio sound thinner?

The nudge changed the phase relationship between layers. Compare in mono, listen to the low end and attack, and keep the offset that sounds strongest rather than the one with the prettiest peak alignment.