Look for transparency drops.
Thin clouds and haze often reduce contrast and star quality before they become obvious visually. A timeline view makes these windows easier to isolate.
Seestar bad-frame workflow
How to filter bad Seestar frames before stacking.
A stack is only as honest as the frames that enter it. Seestar sessions can include good frames, weak frames and frames that quietly damage the final result. The job is to identify them before stacking, not after hours of processing.
Stack decision
Bad-frame filtering is a decision workflow: identify damage, confirm pattern, keep useful data and reject frames that reduce the stack.
Separate one bad moment from a bad session.
Wind, vibration and short tracking events can ruin a subset of frames. The key is whether the damage is isolated, repeated or persistent.
Do not keep every frame just because it exists.
Soft frames dilute the stack. If the quality score drops for a long period, the best stack may use fewer frames but produce a cleaner result.
1. Load the FITS session
Open a Seestar folder or any valid Universal FITS dataset.
2. Sort by quality
Find frames marked as weak, borderline or suspicious.
3. Check the timeline
See if bad frames are clustered around clouds, wind, focus changes or tracking drift.
4. Build a stack subset
Use the recommendation as a starting point, then continue in your stacking workflow.
Next step
Turn bad-frame analysis into a repeatable habit.
Seestar Companion is built for this exact workflow: session verdicts, diagnostics, recommendations and exports without touching original files.