Interactive HTML/SVG viewer for comparing common camera Log and HDR transfer curves.
GitHub Pages: https://xizhonghua.github.io/camera-log-curve/
The chart plots each curve as:
- X axis: exposure stops relative to 18% middle gray
- Y axis: encoded Log signal in IRE
- Dashed vertical guide: diffuse white at 90% reflectance, about +2.32 stops over 18% gray
Move the pointer over the chart to read the IRE value for every visible curve at the same scene exposure. Use the curve list to show or hide individual curves.
- Apple Log
- ARRI LogC3
- ARRI LogC4
- Blackmagic Film Gen 5
- Canon C-Log2
- Canon C-Log3
- DJI D-Log
- Fujifilm F-Log
- Fujifilm F-Log2
- HLG BT.2100, normalized so 18% gray is 38% signal
- Nikon N-Log
- Panasonic V-Log
- RED Log3G10
- Sony S-Log3
Each curve includes a reference link in the UI. Some vendors publish formulas in camera-white-paper form, while others publish LUTs or technical notes; the implementation follows the cited public references as closely as practical for an interactive comparison tool.
Camera Log curves allocate code values differently across shadows, midtones, and highlights. Looking at them on a common "stops over/under middle gray" axis makes it easier to compare:
- where middle gray lands
- how much signal space each curve gives to shadows
- how aggressively highlights are compressed
- how close different camera Log curves are before color management or creative grading
This viewer is meant for exposure, LUT, color-management, and codec-workflow experiments. It is not a replacement for a full color-managed transform pipeline.
Transfer curves can affect codec behavior because they change code-value distribution, local contrast, visible noise, and transform residuals before compression. This is especially relevant for HEVC/H.265 in constant-quality or rate-distortion optimized modes.
The strongest published evidence is around HDR transfer functions such as PQ, HLG, adaptive PQ, and HDR/WCG reshaping rather than direct brand-by-brand camera Log comparisons such as N-Log vs S-Log3 vs C-Log3. In practice, curve choice can matter, but the size of the effect depends heavily on scene content, exposure, noise reduction, bit depth, chroma format, codec settings, and the quality metric used.
Useful references:
- MPEG HDR/WCG HEVC verification work reported about 27% effective bitrate reduction from improved HDR coding practices: https://mpeg.chiariglione.org/standards/mpeg-h/high-efficiency-video-coding/n16506-revised-verification-test-report-hdrwcg-video.html
- HLG vs PQ coding performance has been studied for HEVC Main10 HDR workflows: https://cir.nii.ac.jp/crid/1390282679506233984
- Adaptive perceptual quantization papers show that content-adaptive HDR transfer mapping can improve HEVC rate-distortion behavior compared with fixed PQ: https://doi.org/10.1109/ICIP.2017.8296437
- Apple ProRes is a mezzanine codec with target data rates that are primarily determined by ProRes flavor, frame size, and frame rate; image complexity still affects actual VBR size: https://support.apple.com/en-us/102207
For a rigorous camera Log test, start from the same linear or RAW source, encode multiple transfer curves with identical gamut and bit depth, then compare H.265 or ProRes outputs using bitrate, visually inspected banding/noise, and HDR-aware metrics such as DeltaE-based error, HDR-VQM, or a validated display-referred pipeline.
This is a single static page:
camera-log-curve/
index.html
README.md
Open index.html directly in a browser, or serve the folder with any static web server. No build step is required.