How to Prepare Samples for the Make Noise Morphagene

Guide

The Make Noise Morphagene is one of the most beloved sampling modules in Eurorack. It’s also one of the pickiest. Drop the wrong WAV file on the SD card and the module will either ignore it, mangle it, or worst of all, play it back at subtly wrong pitch and timing.

This guide covers what you need to know to get samples loading cleanly: the exact format the Morphagene expects, how splice markers actually work, how to name and organize your reels, and the small handful of mistakes behind 90% of “why isn’t this working” questions.

The short version

The Morphagene plays back stereo WAV files at 32-bit float, 48 kHz. Each file is called a reel, lives in the root of a FAT32-formatted microSD card, and is named mg1.wav through mg9.wav, then mga.wav through mgw.wav for a maximum of 32 reels. Reels can hold up to about 2.9 minutes of audio and up to 299 splice markers each.

If you only remember three things:

  1. 32-bit float, 48 kHz, stereo WAV. Anything else and the module either won’t load it or plays it back at the wrong speed.
  2. Reels go in the root of the card, named mg<n>.wav. Subfolders are ignored.
  3. Splice markers are stored as cue points inside the WAV file. That’s how the module knows where to break the reel into slices.

The rest of this article is the why behind those rules, plus the practical workflow for prepping samples. Manually if you like to suffer, or with SampleStack if you don’t.

Why the Morphagene is so picky about format

The Morphagene is essentially a digital tape machine. Internally it streams audio off the SD card in real time and applies its granular processing (vari-speed, splice navigation, organize-knob morphing) on top of that stream. To make this work reliably with modulation rates up to audio rate, the firmware expects a very specific sample format and won’t try to convert anything on the fly.

That format is 32-bit float, 48 kHz, stereo WAV. Each piece matters:

  • 48 kHz matches the module’s internal DAC clock. Drop a 44.1 kHz file in and the module plays it back as if it were 48 kHz, meaning your audio comes out about 8.8% faster and a half-step sharp. It still “works,” but it’s not your file anymore.
  • 32-bit float gives you the headroom Morphagene’s internal processing assumes. You can record a reel from the module itself in 24-bit, but for prepared content the firmware expects 32-bit float and on some firmware versions will refuse to load 16-bit or 24-bit files.
  • Stereo is required even for mono source material. The Morphagene is a stereo module, so you have to commit your mono samples to a stereo file (with the same audio in both channels, or with one side silent) before they’ll load.
  • WAV, specifically. Not AIFF, not FLAC, not MP3. The module’s WAV parser is also strict about the chunk layout, which is part of why hand-edited files sometimes fail to load even when they look correct in your DAW.

That combination is why throwing arbitrary samples on the card almost never works on the first try.

How splice markers actually work

A “splice” on the Morphagene is just a cue point stored inside the WAV file. Standard WAV has supported cue points for decades. They’re how you mark positions inside an audio file. The Morphagene reads them on load to decide where each splice begins and ends.

A few things that aren’t obvious:

  • The first sample is always splice 1. You don’t add a cue point at position 0. The start of the file is implicitly the first splice boundary.
  • Adjacent cue points define splice ranges. Splice 1 plays from sample 0 until the first cue point. Splice 2 plays from the first cue point to the second. The last splice plays until the end of the file.
  • You can have up to 299 splices per reel. In practice, anything over 30 or 40 gets unwieldy to navigate from the front panel.
  • Cue points must be inside cue chunks, not LIST or labl chunks. Most DAWs export this correctly. Some sample editors don’t.

This is also where things get fiddly. Different DAWs and editors write slightly different WAV chunk layouts. Logic Pro, Ableton, Reaper, and Adobe Audition all produce files that work on the Morphagene without much tweaking. iZotope RX, some older versions of Pro Tools, and a few sample editors produce WAV files that the module silently rejects, or loads without splice markers because the cue chunks aren’t where the firmware looks for them.

The manual workflow

If you want to do this by hand, the path is:

  1. Convert your source audio to 32-bit float, 48 kHz, stereo WAV. That’s your starting point regardless of where the audio came from. If your source is mono, duplicate it to both channels.
  2. Trim the file to under about 2.9 minutes (174 seconds). Anything longer and the Morphagene will refuse to load the reel.
  3. Decide where your splices go. This is the creative part. Splices are how you’ll navigate the reel from the front panel, so pick boundaries that make sense as performance moments.
  4. Insert cue points at each splice boundary. Most DAWs let you do this with a keyboard shortcut while playing the file back.
  5. Export the WAV file, making sure your DAW writes cue points in the standard cue chunk format. (Reaper, Logic, Ableton: yes. Some sample-only editors: maybe.)
  6. Name the file mg1.wav through mg9.wav, then mga.wav through mgw.wav. The Morphagene loads reels in this exact order. If you only have one reel, it must be mg1.wav.
  7. Format a microSD card as FAT32. Not exFAT. Not NTFS. Cards 32 GB and under usually come pre-formatted as FAT32. Larger cards need to be reformatted.
  8. Copy your reels to the root of the card. Not into a subfolder. The module won’t recurse.
  9. Insert the card into the Morphagene with power off. Boot the module. Sweep the Reel knob and your reels should appear in order.

If a reel doesn’t load, the most likely culprits, in order: wrong sample rate, wrong bit depth, mono file, file longer than 174 seconds, file in a subfolder, or non-standard WAV chunk layout from your editor.

The SampleStack workflow

The whole reason we built SampleStack’s Morphagene profile is that the manual workflow is tedious and easy to get wrong. With the app:

  1. Drag any audio file in. WAV, AIFF, FLAC, MP3, doesn’t matter.
  2. Apply the Morphagene profile. SampleStack flags anything that needs converting and shows you exactly what it’ll change (sample rate, bit depth, channel count, length).
  3. Place splice markers visually on the waveform, or import existing cue points from your DAW. SampleStack writes them in the exact cue chunk format the Morphagene expects.
  4. Export to the SD card. SampleStack handles file naming, FAT32 layout, and the gap reels (so if you have reels 1, 3, and 5 it’ll generate silent placeholders for 2 and 4, or skip them, your call).

Same end result as the manual workflow, but a 30-second job instead of 30 minutes. And the file is guaranteed to load on the first try. Try it free for 7 days.

Common pitfalls

A few things that catch even experienced Morphagene users:

“My reel plays at the wrong speed.” Almost always a sample rate mismatch. The file is 44.1 kHz but the module is expecting 48 kHz. Convert it.

“The reel loads but there are no splices.” The cue points are in a non-standard chunk in your WAV file. Re-export from a different editor, or use SampleStack to rewrite the cue chunk.

“The reel knob doesn’t see my reel.” File is in a subfolder, not named correctly (mg1.wav etc.), or the card isn’t FAT32.

“The audio sounds wrong on one side.” Mono file forced into a stereo container with the wrong channel layout. Duplicate the mono signal to both channels rather than putting it on one side only.

“My reels are showing up in the wrong order.” The naming sequence is mg1, mg2, …, mg9, mga, mgb, …, mgw. After 9 it’s hexadecimal-style letters, not mg10.

“Long reels stutter.” Slow SD card. The Morphagene needs sustained read speeds for full-length reels. A modern Class 10 / U1 microSD will handle it. Older cards sometimes won’t.

Going further

Once you have the basics down, the Morphagene rewards experimentation:

  • Field recordings make incredible Morphagene reels. Long, evolving textures with splice markers placed at moments of change.
  • Multi-tracked source material (drum loops, melodic phrases) on a single reel lets you splice between layers performatively.
  • Noise, drones, and feedback at the start of a reel give the granular engine fascinating textures to chew on.
  • Repurposed splices. Set splices in unconventional places (mid-syllable, mid-transient) and the splice navigation becomes a glitch instrument.

The Make Noise Morphagene manual covers the front-panel side of the module in depth. Once you have a steady supply of well-prepped reels, the rest is just playing.