
Before supplying both a bias and dark to Siril, determine whether the dark master still contains the bias component. The ordinary manual tutorial subtracts an uncalibrated master dark, which already contains bias. The documented dark-optimization route instead uses a bias-calibrated dark and accounts for bias separately. Mixing those recipes can account for the same component incorrectly.
Follow the history of the dark master
A filename such as master-dark does not reveal how it was made. Find the calibration record or rebuild from preserved source frames if that history cannot be established.
For an actual night-sky photo session, record the light sequence, dark sequence, exposure conditions and whether bias was subtracted before the dark frames were stacked. Keep flats in the record too, but do not assume their preparation is identical to the dark preparation.
This concerns removal of a repeatable capture signal. The RAW denoising-stage guide addresses a separate operation and cannot determine which calibration component is already present.
Keep the ordinary tutorial internally consistent
Siril's manual pre-processing tutorial explains its ordinary route by retaining bias in the dark. It separately calibrates the flats with bias correction.
Use that explanation to audit the complete recipe, not as permission to disable fields indiscriminately in an unrelated script. A script may have prepared its masters differently. Inspect its documented assumptions and actual inputs before changing the light-calibration stage.
If the processing record contradicts the intended recipe, stop and resolve that discrepancy before calibrating a whole session.
Treat dark optimization as a different dependency
The Siril 1.4.4 calibration manual describes optimization using a dark that has had bias removed, together with separate bias handling for the lights.
It also distinguishes an automatically evaluated coefficient from one based on recorded exposure times. Those approaches do not remove the need to know the master's preparation. Do not turn optimization on merely to compensate for an unknown dark.
On a working copy, choose the documented route matching the master's history. Preserve the original inputs and record the selected optimization method when used. This guide gives no universal coefficient or camera-independent exposure recipe.
Inspect the calibrated photographs before stacking
Review representative calibrated lights, including dark background areas and any strong sensor-pattern residue. Record clipping or unexpectedly depressed backgrounds as observations requiring investigation, not automatic proof of double subtraction.
A few isolated hot pixels are a different repair decision; the localized hot-pixel guide does not replace calibration bookkeeping.
Only proceed when the master history and selected recipe agree. No calibration sequence or subtraction experiment was performed here, and a pleasing stretched preview cannot by itself certify correct calibration.
Related tasks
- Are the Bright Colored Dots in My Night Photo Stars or Hot Pixels?
- When Should I Use raw denoise Instead of denoise (profiled) on a High-ISO RAW in darktable 5.6?
Sources and scope
- Siril: Manual pre-processing — Pre-processing principle and Darks: The ordinary tutorial leaves bias inside the stacked dark so dark subtraction also removes bias; flats receive bias correction separately.
- Siril 1.4.4: Calibration — Dark optimization: Dark optimization uses a bias-calibrated master dark and incorporates separate bias subtraction for the lights; auto-evaluation and exposure-based methods have different coefficient sources.
Official RawPedia and Siril documentation was opened and read on October 23, 2026 (Asia/Shanghai). The Siril calibration reference identifies version 1.4.4. All acquisition, processing and acceptance examples are proposed; no camera, RAW file, calibration sequence or command was tested. The cover is decorative.
Prepared by Imagild Editorial, the company editorial team responsible for these guides. AI assisted research and drafting. This is documentation-based guidance with proposed examples, not a claim of hands-on testing or an independent product review. Report a correction.
Cover: 85 LENS (Ngô Hoàng Auth) / Pexels, Pexels License. This is an illustrative source photograph, not an application screenshot or a generated test result.