L, M, Q, H
ISO/IEC 18004 defines four Reed–Solomon error-correction levels. Creators usually meet them as single-letter controls in generators like QRburst Design Studio.
| Level | Approx. recovery | Typical use |
|---|---|---|
| L | ~7% | Clean digital display, short payloads, no logo |
| M | ~15% | General purpose when damage risk is modest |
| Q | ~25% | More wear, outdoor print, or light branding |
| H | ~30% | Logos, rough handling, QRburst default |
Density tradeoff
Higher ECC stores more recovery data. For a fixed payload, that can enlarge the required capacity and push the encoder to a higher QR version. Version growth adds modules:
- Same physical print size + higher version → smaller modules
- Smaller modules are harder to resolve after blur, compression, or distance
- Sometimes lowering ECC and removing a logo is less helpful than shortening the URL and keeping H
matrix size (modules per side) = 21 + 4 × (version − 1)
Logos and damage
Center logos intentionally destroy modules. ECC is what makes branded codes possible at all—but only within limits. Larger logos, dense payloads, and low ECC stack badly.
QRburst’s logo size control allows roughly 10–35% of width. That upper bound is a UI range, not a safe operating claim. Larger logos need more careful testing, higher ECC, clearer logo backdrops, and often a larger physical print.
- Prefer H when a logo is present
- Keep finders (corner squares) visually clear
- Use simulated obstruction tests as evidence, then proof on devices and print
Why QRburst defaults to H
QRburst defaults to error correction H because many real jobs include logos, table wear, pocket creases, or less-than-ideal lighting. The default biases toward recovery headroom for those cases.
You can still choose L, M, or Q when you understand the tradeoff—for example a short URL on a large, high-contrast digital slide with no logo. Change deliberately, then re-run configuration analysis and decode stress tests in QR Lab before you publish.