Astrowarden reads every sub, culls the junk, and tells you exactly what went wrong, before you waste an hour stacking. Read-only. Evidence-first. It never invents a pixel of your data.
You know them by heart. Astrowarden answers each one with a number you can check, not a guess at the eyepiece.
Three things, every session, so you never pour an evening into data that was never going to work.
Point it at three hundred subs. It grades every frame on FWHM, eccentricity, trails and SNR, and tosses the junk. Free and unlimited, every night.
287 kept · 13 rejectedTilt, coma, guiding, focus drift, dew, gradients, trails. It sees the cause and shows the measurement behind every call.
BR FWHM 4.6 vs 2.1 · tiltStack it, salvage it, or reshoot. No more eight hours of processing only to learn the subs were never going to work.
verdict: process w/ caveatsThe community is right to be wary of AI in their data. So Astrowarden cites a number you can check or a region you can see for every call. A pro acts on "bottom-right FWHM 4.6 versus 2.1 top-left, monotonic, so tilt not curvature." Never on "you have tilt."
It reads your files and never writes a pixel. No deconvolution, no denoise, no "enhancement." Your data is your data.
Full-resolution frames stay on your machine. Only tiny previews and the measured numbers are ever analyzed in the cloud.
It learns from thousands of nights across every rig and sky. Your app gets smarter with every update, at no extra cost, and runs more of the work right on your machine.
Remote observatory, backyard pier, or a dark site an hour from home: the companion mirrors your desktop session and drives it from anywhere, in real time.
Reads FITS · XISF · TIFF · PHD2 logs. Ha, OIII, SII, or one-shot color. Dedicated astro cam or smart scope.
One purchase, yours on every platform. Unlimited culling never costs a thing. Deep diagnosis runs on simple prepaid credits, so you only ever pay for what you use, and never a cent more.