System Requirements

SonikSearch scales from small sample collections to libraries with millions of sounds. Here's what you need.

Back to SonikSearch
Minimum
Up to 100,000 samples
Requirements
  • OS: macOS 12+ or Windows 10+
  • CPU: 4-core — Apple M1, Intel i5, or AMD Ryzen 5
  • RAM: 8 GB
  • Storage: SSD with space for your audio library

Recommended
Up to 1,000,000 samples

Power User
1,000,000+ samples
Requirements
  • CPU: Apple M2 Pro/Max/Ultra or 12+ core Intel/AMD
  • RAM: 32 GB
  • Storage: Fast NVMe SSD with ample capacity

Database Size

Approximate database size based on number of samples imported.

Library Size Database
10,000~41 MB
50,000~206 MB
100,000~411 MB
500,000~2 GB
1,000,000~4.1 GB
5,000,000~20.5 GB

Estimated Import & Analysis Times

Time to import, analyze BPM/key, and fully process your samples.

Library Size M1 M2 Pro M3 Pro M4 Max i7-12700 i5-10400
10K 15–25 min 10–15 min 8–14 min 7–12 min 12–20 min 25–40 min
50K 1.5–2.5 hrs 50–80 min 40–70 min 35–60 min 1–1.5 hrs 2–3.5 hrs
100K 3–5 hrs 1.5–2.5 hrs 1.5–2 hrs 1–2 hrs 2–3 hrs 4–7 hrs
500K 15–25 hrs 8–14 hrs 7–12 hrs 5–10 hrs 10–16 hrs 20–35 hrs
1M 30–50 hrs 16–28 hrs 14–24 hrs 10–20 hrs 20–32 hrs 40–70 hrs
5M 6–10 days 3–6 days 3–5 days 2–4 days 4–7 days 8–15 days

Audio files are not duplicated — SonikSearch references your original files in place. The database size above is for metadata and analysis data.

GPU acceleration is used during import for advanced audio analysis. Apple Silicon uses the Neural Engine via CoreML; Windows uses DirectML with any compatible GPU. This is not required but significantly speeds up analysis.

SSD storage is strongly recommended for importing. SonikSearch has been optimized to reduce I/O where possible. Once the library is imported, performance is fast even on slower drives.

Import can be paused and resumed at any time. If you close SonikSearch during an import, it will pick up where it left off on the next launch.