🗂️ How Presets Are Organized

Turning a collection of scattered Caustic presets into an organized, searchable archive.

The Caustic 3 Archive contains community presets recovered from surviving collections and historical sources. Over time, these files can accumulate with inconsistent filenames, duplicate copies, different versions of the same preset, and little information about what each sound is intended to be.

To make the collection easier to explore and use, the Archive processes presets through a custom classification and organization system.

Presets are organized by machine type and practical sound category, while the original files themselves remain unchanged.

🎛️ Why Organize Presets?

🛡️ Preserving File Integrity

🔍 How Classification Works

📁 Built for Caustic's File Browser

📋 Transparent & Auditable

🔄 An Evolving System

📦 From Individual Presets to Archive Collections

🔬 Part of the Technical Archive

🎛️ Why Organize Presets?

A large collection of presets is difficult to use when everything is stored in one folder or depends entirely on filenames.
A preset called Bass 01 gives some information. A preset called 001 gives almost none.

The Archive therefore adds structure around the original files by identifying likely sound categories such as:

Bass · Lead · Pad · Pluck · Keys· Drums/Percussion · FX · Arp · Vox · Strings · Brass · Guitar · Chord/ Stab

The purpose isn't to replace the original naming or reinterpret the preset. It is to make a large historical collection easier to browse and discover.

🛡️ Preserving File Integrity

Non-Destructive Deduplication

The sorter does not modify the contents of the original preset files. Every preset is checked using a content hash before it is organized.

Exact Duplicates: Files with identical content are identified so the same preset does not need to be archived repeatedly.
Different Presets, Same Filename: If two files have the same filename but different binary content, both are retained. The additional version receives a variant name rather than replacing the original.
Original Files Preserved: The preset binary itself is not rewritten as part of the classification process.

This means organization can be changed or improved later without having to reconstruct the original preset data.

🔍 How Classification Works

A Multi-Stage Classification System

The sorter uses several levels of evidence rather than depending on a single rule.

Stage 1 — Filename Analysis

The first step examines the preset filename for useful category information.

Names containing terms such as Bass, Lead, Pad, Pluck, Keys, Drums, FX, Arp, Vocal, Strings, Brass, Guitar, Woodwind, or Chord can provide a strong indication of the intended sound.

The system also uses safeguards against accidental matches and misleading word fragments.

For example, a filename containing a category word as part of an unrelated word should not automatically be treated as evidence of that category.

Stage 2 — Preset Parameter Analysis

When the filename does not provide enough information, the sorter can examine the actual parameters stored inside supported Caustic preset binaries.

Instead of listening to the preset as audio, the system reads its stored synthesizer parameters and creates a parameter signature.

These values can then be compared against curated reference presets whose sound categories are already known.

This provides another source of evidence when a preset has an uninformative name such as: MyPreset_01

The purpose is not to claim that a preset can always be identified perfectly from its parameters, but to provide additional evidence when the filename alone is insufficient.

Stage 3 — Machine-Specific Processing

Different Caustic machines store and represent sounds differently, so the same classification method cannot always be applied to every machine.

The sorter therefore uses machine-specific handling where necessary.

Synthesizer Machines: Supported machines can be compared using their stored parameter values and known reference profiles.
PCMSynth: Sample-based information is considered where appropriate rather than treating it like a conventional synthesizer.
BeatBox: Embedded sample information, including sample names where available, can provide useful classification evidence.
Modular Synth: Modular presets do not provide the same fixed parameter structure as the other machines, so classification uses a more conservative fallback rather than pretending that a standard parameter fingerprint is available.

The result is a system that adapts to the way each Caustic machine actually stores its presets.

📁 Built for Caustic's File Browser

Practical Collection Limits

A well-organized archive also needs to remain practical to use inside Caustic itself.

The sorter therefore structures its output around known Caustic browser and filename constraints.

Folder Splitting: Large collections are automatically divided into smaller folders rather than placing an excessive number of files into a single directory.
File Limits: Output is structured around Caustic's known 1024-file folder limit.
Filename Limits: Filenames are kept within the known 64-character limit used by the organization process.

This makes the resulting collections easier to browse and reduces the risk of creating folders that are inconvenient or unsuitable for use within Caustic.

📋 Transparent & Auditable

Every Sorting Decision Has a Reason

The sorter records information about how each preset was classified.

A decision can be based on evidence such as:

Filename Match
Preset Parameter Analysis
Embedded Sample Information
Machine-Specific Rules
Fallback Classification

This means the resulting collection is not simply a set of folders produced by an unexplained automated process.

The sorting information can be inspected to understand why a preset ended up where it did.

As more known reference presets are researched, the classification rules and reference data can be refined.

🔄 An Evolving System

Preset classification is not treated as a finished, infallible process.

Some presets are easy to identify from their names. Others require examination of their stored parameters, sample information, or machine type. Some remain ambiguous.

As the Archive's technical research develops, classification methods can be improved and additional reference material can be added.

The important principle is that uncertainty is preferable to inventing information.

Where there isn't enough evidence for a confident classification, the sorter can leave the preset in a more conservative category rather than forcing it into an arbitrary sound type.

📦 From Individual Presets to Archive Collections

Once presets have been identified, deduplicated and classified, the resulting collection can be packaged into a much more useful form.

The Archive can then provide:

Original preset files → Organized sound categories → Machine-specific folders → Metadata and indexes → Public archive collections

This turns a large collection of recovered files into something that can actually be explored and used.

Explore Preset Collections

🔬 Part of the Technical Archive

The classification system depends on research into how Caustic stores preset information internally.

That research includes identifying machine signatures, locating preset names, determining parameter layouts, comparing binary structures and testing those findings against real Caustic presets.

The public-facing sorting system is therefore only the visible result of a much larger preservation and reverse-engineering process.

Explore Technical Research

Page updated

Google Sites

Report abuse