Pick a deep-sky target, your Bortle class and a smart telescope. We estimate the total stacking time for a clean, recognizable image, with and without a dual-band filter, and compare every scope side by side.
Filter
How a smart telescope builds an imageEvery smart scope here follows the same loop; the differences are optics, sensor and software polish.
1. Plate-solve
The camera takes a short frame, matches the star pattern to a catalog and works out exactly where it's pointing. No star alignment, no finder scope.
2. GoTo & track
Motors slew to your target and follow it as the Earth turns. Alt-az mounts also rotate the image slowly, which the software corrects.
3. Short exposures
Instead of one long exposure, the scope takes many 10–30 s frames. Frames spoiled by clouds, wind or bumps are rejected.
4. Live stacking
Each new frame is aligned and averaged into the running image, so noise drops and faint detail appears as you watch: roughly √N better signal-to-noise after N frames.
5. Stretch & share
The app stretches the faint data, removes gradients and denoises, then saves a JPEG (and often FITS/TIFF) to your phone.
We start from a baseline time for each target's editorial difficulty tier at a Bortle 4 sky with f/5 optics and 2.9 µm pixels: about 5 minutes for showpieces, 20 for easy targets, an hour for moderate ones, 2.5 hours for hard ones and 6 hours for very faint ones. Then:
Sky brightness. Background light grows by about 2.5× per magnitude of skyglow. For faint targets the time scales with it; for bright targets (tiers 1–3) we soften the penalty, because they're less background-limited. Bortle class sky brightness comes from the commonly cited Bortle-scale table (22 mag/arcsec² at class 1, 17.8 at class 9).
Dual-band filter. On emission nebulae, the sky penalty is cut sharply (we raise it to the power 0.35), reflecting how much skyglow H-alpha/OIII filters reject.
Optics. Per-pixel signal scales with (pixel size ÷ focal ratio)². Faster optics such as the Origin's f/2.2 RASA need much less time; we count pixel size at half weight because small-pixel images are usually viewed downscaled.
Small, faint targets get an aperture adjustment, since resolving them needs more aperture as well as time.
The result is a planning estimate, not a prediction. Moonlight, haze, a low target, dew and the app's frame rejection all add time, and "clean" is subjective. Use it to decide whether a target is a one-evening job or a project.
A worked example
The Rosette Nebula from a Bortle 7 suburb: on a Seestar S50 Pro with its built-in dual-band filter, the estimate stays within a single long evening. Untick the filter and it climbs into multi-night territory, because the faint red nebulosity is buried in skyglow. Switch the target to the Pinwheel Galaxy (M101): the filter makes no difference (galaxies aren't emission targets), and from Bortle 7 it runs past ten hours on a 50 mm Seestar. On the Celestron Origin's f/2.2 optics the same galaxy drops to about a quarter of that, still a multi-night project but a realistic one, which is the real reason that 41.6 lb telescope exists.
The total exposure time stacked into a final image: 120 ten-second frames is 20 minutes of integration. Signal grows in proportion to time while random noise grows with its square root, so doubling the time improves signal-to-noise by about 1.4×. That's why faint targets need hours, not minutes.
How much does light pollution increase stacking time?
For faint targets, roughly in proportion to sky brightness. Each magnitude per square arcsecond of extra skyglow is about 2.5× more background light, so moving from a Bortle 4 sky (about 21.1 mag/arcsec²) to a Bortle 8 sky (about 18.2) means roughly 14× more background and, in the background-limited case, roughly 14× the time for the same result. Bright targets are less affected.
Does a dual-band filter really save time?
On emission nebulae, yes, a lot: it passes the H-alpha and OIII light those nebulae emit and blocks most of the skyglow around them. On galaxies, clusters and reflection nebulae it removes signal along with the skyglow, so leave it out. The estimator applies the filter only to targets where it helps.
Can I stack over several nights?
Yes. Vaonis' Vespera 3 and Pro 2 resume a stack automatically across nights; with other scopes you can save the raw frames (FITS) and stack them together on a computer in free software such as Siril. Multi-night projects are how owners reach the multi-hour totals faint galaxies need from suburbs.