# Stacking Starlight — Full Text Corpus

> The complete agent-readable content of stackingstarlight.com, an interactive guide to deep-space astrophotography by Michael Kalika.

## LLM resources

- [LLM index (llms.txt)](https://stackingstarlight.com/llms.txt)
- [Complete LLM text (llms-full.txt)](https://stackingstarlight.com/llms-full.txt)
- [Markdown homepage (index.md)](https://stackingstarlight.com/index.md)

## Astrophotography — The Science of Starlight
Source: [https://stackingstarlight.com/](https://stackingstarlight.com/)

Stacking Starlight is an interactive guide to deep-space astrophotography: how faint photons become finished images through capture, calibration, stacking, and stretching.

Astrophotography turns a few photons per minute from objects thousands of light-years away into detailed color images. It works by collecting light over many long exposures, removing the camera's noise and optical defects with calibration frames, averaging dozens or hundreds of frames together to raise signal above noise, and finally stretching the faint data so the human eye can see it.

This site teaches the full pipeline interactively. Suggested path: **Start Here** (the physics foundations) → **Equipment** (telescope, mount, camera, guiding, filters) → **Sensor** and **Noise** (how a sensor records light and what corrupts it) → **Signal Journey** (the whole pipeline end to end) → **Calibration**, **Stacking**, **Processing** (the core techniques) → **Practical** and **Advanced** (sub-exposure planning, mono vs color, narrowband, the Hubble palette).

---

## Start Here — Astrophotography Foundations
Source: [https://stackingstarlight.com/start-here](https://stackingstarlight.com/start-here)

The foundations of astrophotography: deep-space objects, light and photons, color and wavelength, signal vs noise, Earth's rotation, and how sensors capture light.

Before equipment and processing, six ideas make everything else click: **what** you photograph (nebulae, galaxies, clusters, supernova remnants), **what light is** (photons — discrete packets you collect over time), **how color works** (wavelength, and the narrow emission lines nebulae glow at), **the core challenge** (signal vs noise), **why tracking is needed** (Earth's rotation), and **how a sensor records light** (pixels counting electrons). Each sub-page below covers one.

---

## Astrophotography Equipment Guide
Source: [https://stackingstarlight.com/equipment](https://stackingstarlight.com/equipment)

The astrophotography gear chain: telescope, equatorial mount, camera, focuser, plate solving, guiding, filters, and software — and why the mount matters most.

A deep-sky imaging rig is a system: a **telescope** (the lens that gathers light), an **equatorial mount** (tracks the sky — the single most important component), a **camera** (records photons, ideally cooled), a **focuser** (holds critical focus as temperature drifts), **plate solving** (tells the mount exactly where it points), **autoguiding** (corrects tracking errors in real time), **filters** (select wavelengths / fight light pollution), and **software** (orchestrates the session). The sub-pages cover each in depth.

---

## How Camera Sensors Work — Astrophotography
Source: [https://stackingstarlight.com/sensor](https://stackingstarlight.com/sensor)

How camera sensors capture starlight: pixels and photon counting, full-well and read noise, and the analog-to-digital pipeline to ADU.

A sensor is a grid of light-sensitive pixels. Photons free electrons that accumulate in each pixel's well during the exposure (charge proportional to light). At readout, an amplifier and **ADC** convert each well's charge into a digital number (ADU). Limits and noise enter here: **full-well capacity** caps the brightest recordable signal, **read noise** is added every readout, **quantum efficiency** sets how many photons become electrons, and **bit depth** sets quantization fineness. This pipeline — photon → electron → voltage → ADU — is where signal and several noise sources originate.

---

## Noise in Astrophotography
Source: [https://stackingstarlight.com/noise](https://stackingstarlight.com/noise)

The noise sources in astrophotography — photon shot, read, thermal/dark, fixed-pattern, dust, amp glow, banding, quantization — and how each is reduced.

Every astrophotograph is **signal** (real light from the sky) plus **noise** (unwanted variation). The entire acquisition-and-processing workflow exists to separate the two. Noise comes in two families: *random* noise that differs every frame (beaten by stacking more exposures) and *fixed-pattern* noise that repeats every frame (removed by calibration frames).

The noise sources covered on this page:

- **Photon (Shot)** — Fundamental quantum noise. Photons arrive randomly following a Poisson distribution — like raindrops on a window. Brighter areas receive more photons, so they have more absolute noise but BETTER signal-to-noise ratio. This is the only noise that cannot be calibrated away — you beat it by collecting more photons (longer total integration). _Formula:_ `SNR = √(signal electrons)`. _Reduced by:_ More integration time (stacking).
- **Read Noise** — Every time the sensor is read, the amplifier and analog-to-digital converter add a small random error. It's like a toll booth — you pay it every single frame regardless of exposure length. Modern cooled CMOS sensors achieve 1–3 e⁻. This is why longer subs are preferred: you pay the toll once for more signal. _Formula:_ `σ_read ≈ constant (e⁻/read)`. _Reduced by:_ Bias frame subtraction.
- **Thermal (Dark)** — Heat causes electrons to spontaneously appear in pixel wells, even in total darkness. The rate doubles every ~6°C. Some pixels have manufacturing defects causing extremely high dark current — these are 'hot pixels.' This is why astro-cameras are actively cooled to −20°C or colder, reducing thermal noise by 64× vs room temperature. _Formula:_ `∝ e^(−Eg/2kT) × time`. _Reduced by:_ Dark frame subtraction + cooling.
- **Fixed Pattern** — Each pixel has a slightly different quantum efficiency — how well it converts photons to electrons. Some pixels are 95% efficient, others 107%. This creates a fixed multiplicative pattern that's the same in every frame. Because it's perfectly repeatable, it's perfectly correctable with flat frames. _Formula:_ `output = signal × QE(x,y)`. _Reduced by:_ Flat frame division.
- **Dust Motes** — Tiny particles on the sensor cover glass or filters cast soft, out-of-focus circular shadows called 'dust donuts.' Particles farther from the sensor cast larger, more diffuse shadows. They don't move between frames (unless you touch the camera), so flat frames remove them perfectly. _Formula:_ `shadow ∝ 1 − opacity`. _Reduced by:_ Flat frame division.
- **Amp Glow** — The readout amplifier chip emits infrared photons that leak into nearby pixel wells. This creates a warm gradient, typically in one corner of the image. It scales linearly with exposure time — longer subs have more glow. Visible in uncooled CMOS and long narrowband exposures. _Formula:_ `glow ∝ exposure time`. _Reduced by:_ Dark frame subtraction.
- **Banding** — Each column of pixels has its own tiny amplifier with a slightly different electronic offset. This creates vertical (or horizontal) stripes. Some banding is fixed (same every frame), some is random. Dithering between exposures helps average out residual banding after bias subtraction. _Formula:_ `offset varies per column`. _Reduced by:_ Bias subtraction + dithering.
- **Quantization** — The analog-to-digital converter (ADC) rounds continuous voltages to discrete integer values. With a 16-bit ADC (65,536 levels), each step is tiny and quantization noise is negligible. With 8-bit (256 levels), smooth gradients become visible staircase steps — 'posterization.' Most astro cameras use 12–16 bit ADCs. _Formula:_ `σ_q = LSB / √12`. _Reduced by:_ Higher bit-depth ADC.

---

## The Signal Journey — From Photon to Pixel
Source: [https://stackingstarlight.com/signal-journey](https://stackingstarlight.com/signal-journey)

Follow a photon end to end: from deep space through the telescope and sensor, then calibration, stacking, and stretching to a finished image.

This page traces one photon's journey and, with it, the whole pipeline. **1) Capture:** photons cross light-years, enter the telescope, and are recorded as a noisy raw frame. **2) Calibration:** bias/dark/flat frames remove the camera's electronic baseline, thermal signal, and optical defects. **3) Stacking:** dozens to hundreds of calibrated frames are aligned and averaged, raising SNR by √N and rejecting outliers (satellites, cosmic rays). **4) Stretching:** the faint linear data is non-linearly stretched so dim detail becomes visible without blowing out stars. The result: a clean, detailed image from individually unusable frames.

---

## Calibration Frames — Bias, Darks, Flats
Source: [https://stackingstarlight.com/calibration](https://stackingstarlight.com/calibration)

Calibration frames — bias, darks, and flats — remove fixed, repeatable defects so stacking only averages real signal and random noise.

Calibration removes everything in your frames that *isn't* sky signal. **Bias frames** (zero-length, capped) capture the electronic baseline — read-noise pattern, ADC pedestal, column banding. **Dark frames** (same exposure and temperature as your lights, lens capped) capture thermal current, hot pixels, and amp glow. **Flat frames** (evenly illuminated, ~50% histogram) map vignetting, dust shadows, and per-pixel sensitivity. The calibration math: `calibrated = (light − master_dark) / normalized_flat`. Bias/darks are *subtracted*; flats are *divided*. Stack many of each so calibration adds no noise of its own.

---

## Image Stacking — Why More Frames = Less Noise
Source: [https://stackingstarlight.com/stacking](https://stackingstarlight.com/stacking)

Stacking aligns and combines many calibrated frames to raise SNR by √N, with dithering and sigma clipping to reject artifacts.

Stacking is how random noise is beaten. Averaging N frames improves signal-to-noise by **√N** (4 frames → 2×, 100 frames → 10×). Frames are first **registered** (aligned on stars), then combined. **Sigma clipping** rejects pixels that deviate too far from the median — removing satellite trails, cosmic rays, and plane lights without harming real signal. **Dithering** (shifting the mount a few pixels between frames) decorrelates fixed patterns and walking noise so they average away during integration. Total integration time is the real lever: more, longer, well-dithered subs make cleaner images.

---

## Image Processing & Stretching
Source: [https://stackingstarlight.com/processing](https://stackingstarlight.com/processing)

Processing and stretching: histogram stretching turns faint linear data into a visible image while preserving stars and controlling background.

Stacked data is **linear** — faint nebulosity sits just above the background and is invisible until stretched. **Histogram stretching** applies a non-linear curve that lifts the dim midtones (revealing faint detail) while holding the black point near the sky background and keeping bright stars from saturating. Done well, stretching unveils structure that was always in the data; done poorly, it crushes blacks or bloats stars. Further processing (color calibration, noise reduction, sharpening, star reduction) refines the stretched image, but stretching is the pivotal step from data to picture.

---

## Practical Astrophotography Tips
Source: [https://stackingstarlight.com/practical](https://stackingstarlight.com/practical)

Practical workflow: choosing sub-exposure length (read-noise-limited vs sky-limited), session planning, and total integration.

Practical choices that decide image quality. **Sub-exposure length:** long enough that sky-background noise dominates read noise (so read noise becomes negligible), but short enough to avoid saturating stars, tolerate guiding errors, and discard fewer frames. Light pollution and f/ratio shorten the optimal sub; narrowband lengthens it. **Total integration** matters more than any single sub — hours, not minutes. **Session planning:** image targets near transit (highest, least atmosphere), plan around the meridian flip, and collect calibration frames. Many short well-guided subs beat a few long risky ones.

---

## Advanced Topics — Cameras, Color & Filters
Source: [https://stackingstarlight.com/advanced](https://stackingstarlight.com/advanced)

Advanced topics: the Bayer color mosaic, mono vs color cameras, narrowband filters, and RGB/LRGB/Hubble-palette color.

Deeper material once the fundamentals are in place: how **color cameras** reconstruct color from the **Bayer mosaic**, the trade-offs of **mono vs one-shot-color** sensors, how **narrowband filters** isolate emission lines to image through light pollution, and how color is assembled in **RGB, LRGB, and the SHO Hubble palette**. Each sub-page goes deep on one.

---

## Telescopes for Astrophotography
Source: [https://stackingstarlight.com/equipment/telescope](https://stackingstarlight.com/equipment/telescope)

Telescopes for imaging: refractors, reflectors, and compound scopes; aperture, focal length, and f/ratio explained.

The telescope gathers and focuses light. **Aperture** (diameter) sets how much light you collect; **focal length** sets magnification and field of view; **f/ratio** (focal length ÷ aperture) sets how fast you collect — faster (e.g. f/5) needs shorter exposures than slow (f/10). **Refractors (APO)** are sharp, low-maintenance, great for widefield. **Reflectors (Newtonian)** give the most aperture per dollar but need **collimation** — periodically aligning the primary and secondary mirrors (with a laser or Cheshire tool) so stars focus to tight points instead of comet-shaped flares. **Compound (SCT/Mak)** pack long focal length into a compact tube for galaxies. Beginners: a 60–80mm APO refractor is forgiving and produces beautiful widefield images.

---

## Equatorial Mounts for Astrophotography
Source: [https://stackingstarlight.com/equipment/eq-mount](https://stackingstarlight.com/equipment/eq-mount)

The equatorial mount is the most important piece of gear: it cancels Earth's rotation. Star trackers, German equatorial mounts, and harmonic drives.

Because Earth rotates, you need a mount whose motor turns at the sidereal rate about an axis aligned with the celestial pole. Types: **star trackers** (lightweight, camera + lens, 5–8 kg), **German equatorial mounts (GEM)** (the standard for telescopes — HEQ5, EQ6, CEM26), and **harmonic-drive** mounts (newer, smooth, high capacity for their weight — ZWO AM5, iOptron HEM27). A great scope on a poor mount yields blurry stars; a modest scope on a great mount yields sharp images. GEMs perform a **meridian flip** when the target crosses due south. Before imaging you must **polar align** — aim the mount's RA axis at the celestial pole (Polaris in the north) so its single motor follows the sky; poor alignment causes field rotation and drift. This is why astrophotography uses **equatorial** mounts rather than **alt-azimuth (alt-az)** mounts: an alt-az mount tracks in two axes and slowly rotates the field over long exposures, while an EQ mount cancels Earth's rotation about one polar axis. Every geared mount also has **backlash** — a small dead zone where the worm gear's teeth sit between contact, so when a motor reverses direction the scope doesn't move until that gap is taken up. Backlash hurts DEC guiding most; software like PHD2 applies **backlash compensation**, or you guide in one direction only. Harmonic-drive mounts have far less of it.

---

## Imaging Night Simulation
Source: [https://stackingstarlight.com/equipment/night-sim](https://stackingstarlight.com/equipment/night-sim)

A full imaging session: select target, slew, track, capture hundreds of subs, meridian-flip at transit, repeat — typically 6–10 hours.

A real session spans the hours of darkness. You pick a target, slew and plate-solve onto it, engage tracking and guiding, and capture a long sequence of sub-exposures. As the target crosses the meridian (its highest point), a GEM must perform a **meridian flip** to avoid hitting the tripod. Plan so the target transits near the middle of the night for the best altitude (least atmosphere). Sequencer software (NINA, Voyager, SGPro) runs the whole thing autonomously: GoTo → plate solve → autofocus → capture → flip → repeat. More subs = better SNR.

---

## Astrophotography Cameras — Cooled CMOS & DSLR
Source: [https://stackingstarlight.com/equipment/camera](https://stackingstarlight.com/equipment/camera)

Astrophotography cameras: cooled CMOS (mono or color) vs DSLR/mirrorless. Cooling, quantum efficiency, and read noise.

Dedicated astro cameras use **active cooling** (Peltier to −20°C or colder) to cut thermal noise (~64× vs room temperature) and make dark frames repeatable, plus high quantum efficiency (90%+) and low read noise (1–3 e⁻). **Mono cooled CMOS** (ASI294MM, QHY268M) give the best quality but need filters. **Color/OSC cooled CMOS** (ASI533MC, QHY268C) simplify the workflow. **DSLR/mirrorless** cameras work well to start but can't regulate temperature. Start with whatever you have; upgrade to a cooled camera when ready.

---

## Focusing & Autofocus for Astrophotography
Source: [https://stackingstarlight.com/equipment/focuser](https://stackingstarlight.com/equipment/focuser)

Critical focus and autofocus: electronic focusers, V-curve/HFR autofocus, temperature drift, the critical focus zone, and Bahtinov masks.

Even a perfect telescope makes blurry stars if the sensor isn't exactly at the focal plane, and temperature changes shift focus all night as metal expands and contracts. An **electronic focuser** (ZWO EAF, Pegasus FocusCube) replaces the manual knob with a stepper motor. **Autofocus software** measures star sharpness via **HFR** (half-flux radius), steps through positions to build a **V-curve**, and solves for the minimum — micron-precision focus, repeated every 30–60 minutes or per degree of temperature change. A **Bahtinov mask** is a cheap manual aid: its diffraction spikes align exactly at focus. Focus tolerance is the **critical focus zone (CFZ)** — the tiny range of focuser travel (often just tens of microns, and narrower at fast f/ratios) within which stars stay acceptably sharp, so focus must land and stay inside it. Focusers also have **backlash** (mechanical play on direction reversal), which is why autofocus routines always approach the final position from the same direction. An EAF is one of the best upgrades for unattended imaging.

---

## Plate Solving — GPS for Your Telescope
Source: [https://stackingstarlight.com/equipment/plate-solve](https://stackingstarlight.com/equipment/plate-solve)

Plate solving is GPS for your telescope: it matches star patterns in a test frame against a catalog to find exact pointing in seconds.

After a GoTo, how does the mount know precisely where it points? **Plate solving** takes a short exposure, extracts star positions, forms geometric triangles between them, and matches those patterns against a catalog of billions of stars. In 1–3 seconds it returns your exact coordinates to sub-arcsecond accuracy, and the mount corrects any residual error. Solvers: **Astrometry.net** (open source, offline), **ASTAP** (fast local, great with NINA), **All-Sky Plate Solver**. It removes star-hopping entirely: GoTo → solve → correct → center, even if the initial alignment is off by degrees.

---

## Autoguiding — Precision Star Tracking
Source: [https://stackingstarlight.com/equipment/guiding](https://stackingstarlight.com/equipment/guiding)

Autoguiding corrects residual tracking errors in real time using a second camera watching a guide star. Guidescopes vs off-axis guiders; PHD2.

Even good mounts have small errors — periodic gear error, bearing play, wind, refraction — that streak stars over a multi-minute exposure. **Autoguiding** uses a small second camera to watch a **guide star** and, every 1–2 seconds, sends correction pulses to the mount's RA/DEC motors to keep it locked. A **guidescope** (small refractor + guide cam) is simple and cheap with a wide field; an **off-axis guider (OAG)** picks light off the main optical path with a prism, eliminating differential flexure at long focal lengths. DEC corrections must also account for **backlash** (gear play on direction reversal), so many imagers guide DEC in one direction only or enable backlash compensation. **PHD2** (free) is the standard software. Essential above ~200mm focal length.

---

## Filters for Astrophotography
Source: [https://stackingstarlight.com/equipment/filters](https://stackingstarlight.com/equipment/filters)

Filters select wavelengths: LRGB broadband for natural color, narrowband (Hα/OIII/SII) to isolate nebula lines and block light pollution.

Filters choose which light reaches the sensor. **LRGB** (luminance + red/green/blue) capture natural color with mono cameras. **Narrowband** filters (Hα 656nm, OIII 501nm, SII 672nm; 3–7nm wide) isolate single nebula emission lines and reject nearly all artificial light. **Dual-narrowband** filters (L-eXtreme, L-Ultimate) pass Hα+OIII together for one-shot-color cameras — the simplest way to image emission nebulae from a city. For light-polluted skies a dual-narrowband filter is a game-changer.

---

## Astrophotography Software — NINA, PHD2, ASIAIR
Source: [https://stackingstarlight.com/equipment/software](https://stackingstarlight.com/equipment/software)

Software runs the session: capture sequencing, mount control + plate solving, autoguiding, and autofocus — often from a Raspberry Pi controller.

A controller at the telescope orchestrates everything: **capture/sequencing** (NINA — free, Windows; APT; SGPro; ASIAIR), **mount control + plate solving**, **autoguiding** (PHD2 — free, cross-platform), and **autofocus**. Integrated controllers like **ASIAIR** or **Stellarmate** run on a Raspberry Pi controlled from your phone. **Processing** happens later in PixInsight, Siril (free), or DeepSkyStacker (free). NINA + PHD2 is the free gold standard on Windows; ASIAIR suits a phone-controlled, self-contained rig.

---

## My Astrophotography Setup
Source: [https://stackingstarlight.com/equipment/my-setup](https://stackingstarlight.com/equipment/my-setup)

Michael Kalika's personal astrophotography rig — telescope, mount, camera, and accessories, with field photos.

A personal-gallery page showing the author's real imaging setup (telescope, equatorial mount, camera, and accessories) in the field — balcony, desert, and travel sessions. It illustrates how the equipment described across the site comes together into a working rig.

---

## Deep Space Objects — What's Out There
Source: [https://stackingstarlight.com/start-here/objects](https://stackingstarlight.com/start-here/objects)

The deep-space objects you can photograph: emission/reflection/planetary nebulae, galaxies, star clusters, and supernova remnants.

Deep-sky targets fall into a few families. **Emission nebulae** are clouds of hydrogen/oxygen/sulfur gas glowing at specific wavelengths (great for narrowband). **Reflection nebulae** scatter blue starlight. **Galaxies** are distant island universes — small and faint, rewarding long focal lengths. **Star clusters** (open and globular) are bright and beginner-friendly. **Supernova remnants** and **planetary nebulae** are the colorful debris of dying stars. Brightness (magnitude) and apparent size determine the exposure and focal length you need.

---

## Light & Photons — The Basics of Astrophotography
Source: [https://stackingstarlight.com/start-here/light-and-photons](https://stackingstarlight.com/start-here/light-and-photons)

Light is made of photons — discrete packets of energy. Astrophotography is fundamentally the business of collecting enough of them.

A photon is the smallest unit of light. Faint deep-sky objects deliver only a handful of photons per pixel per minute, which is why exposures are long and total integration (the sum of all exposures) is measured in hours. Because photons arrive randomly, doubling your photons does not double your noise — signal grows faster than noise, so more integration always helps. Everything downstream (sensor, calibration, stacking) is in service of collecting and preserving photons.

---

## Color & Wavelength in Astrophotography
Source: [https://stackingstarlight.com/start-here/color-and-wavelength](https://stackingstarlight.com/start-here/color-and-wavelength)

Wavelength determines color, from ~400nm violet to ~700nm red. Nebulae emit at specific lines like Hα (656nm), OIII (501nm), and SII (672nm).

Visible light spans roughly 400nm (violet) to 700nm (red). Stars emit a continuous spectrum, but nebulae glow at discrete **emission lines** set by their chemistry: hydrogen-alpha (Hα, 656nm, deep red), doubly-ionized oxygen (OIII, 501nm, teal), and singly-ionized sulfur (SII, 672nm, red). Because these lines are narrow and specific, you can isolate them with narrowband filters — capturing nebula light while rejecting almost all light pollution. This is also the basis of false-color palettes like SHO (the Hubble palette).

---

## Signal vs Noise — The Core Challenge
Source: [https://stackingstarlight.com/start-here/signal-vs-noise](https://stackingstarlight.com/start-here/signal-vs-noise)

Every image is signal plus noise. The whole workflow exists to raise signal-to-noise ratio (SNR) so faint detail emerges.

Signal is the real light you want; noise is random variation that hides it. The metric that matters is **signal-to-noise ratio (SNR)**. Random noise grows with the square root of the number of frames, while signal grows linearly — so stacking N frames improves SNR by √N. Fixed-pattern noise (the same every frame) is removed by calibration instead. Understanding which noise is random and which is fixed tells you whether to *collect more* (stacking) or *calibrate it out* (bias/dark/flat frames).

---

## Earth's Rotation & Star Tracking
Source: [https://stackingstarlight.com/start-here/earths-rotation](https://stackingstarlight.com/start-here/earths-rotation)

Earth rotates once per sidereal day, so stars drift. Equatorial mounts cancel that motion for sharp long exposures.

Earth spins once every 23h 56m (a sidereal day), making stars appear to drift east-to-west. Any exposure longer than a few seconds shows star trails unless the camera tracks. An **equatorial mount** aligns one axis (the RA axis) with Earth's rotation axis and turns it at exactly the sidereal rate, so the sky appears frozen. This is what makes minutes-long exposures of faint objects possible.

---

## How Sensors & Pixels Capture Light
Source: [https://stackingstarlight.com/start-here/sensor-and-pixels](https://stackingstarlight.com/start-here/sensor-and-pixels)

A camera sensor is a grid of pixels that count photons by accumulating electrons, then convert them to numbers via an ADC.

Each pixel is a well that fills with electrons as photons strike it (the photoelectric effect). At the end of an exposure the charge in every well is read out and digitized by an analog-to-digital converter (ADC) into a number (ADU). Key properties: **quantum efficiency** (fraction of photons converted), **full-well capacity** (before saturation), **read noise** (added each readout), and **bit depth** (ADC levels). These set how faint a signal you can record and how cleanly.

---

## The Bayer Filter Mosaic
Source: [https://stackingstarlight.com/advanced/bayer-mask](https://stackingstarlight.com/advanced/bayer-mask)

The Bayer filter mosaic: an RGGB grid puts one color filter over each pixel; software (debayering) reconstructs full color.

A one-shot-color sensor has a **Bayer mosaic** — a repeating 2×2 pattern of red, green, green, blue filters over the pixels (RGGB; green is doubled because the eye is most sensitive to it). Each pixel therefore records only one color. **Debayering** (demosaicing) interpolates the two missing colors at every pixel from its neighbors to build a full-color image. The trade-off: simplicity (color in one shot) at the cost of resolution and efficiency versus a mono sensor with filters.

---

## Mono vs Color Cameras for Astrophotography
Source: [https://stackingstarlight.com/advanced/mono-vs-color](https://stackingstarlight.com/advanced/mono-vs-color)

Mono vs one-shot-color cameras: resolution, sensitivity, and workflow trade-offs, and which suits you.

**Monochrome** sensors have no Bayer mask, so every pixel records every photon at full resolution and efficiency; you capture through external R/G/B or narrowband filters and combine channels in processing — maximum quality and flexibility (essential for serious narrowband), but more gear (filter wheel) and more time. **One-shot-color (OSC)** captures all colors at once via the Bayer mask — far simpler and faster per target, ideal for limited clear nights and broadband or dual-narrowband work, at some cost in resolution/sensitivity. Choose OSC for simplicity, mono for ultimate control.

---

## Narrowband Filters — Imaging Through Light Pollution
Source: [https://stackingstarlight.com/advanced/narrowband-filters](https://stackingstarlight.com/advanced/narrowband-filters)

Narrowband filters (3–7nm) isolate single emission lines (Hα, OIII, SII), blocking 99%+ of light pollution to image from cities.

Narrowband filters pass only a sliver of spectrum (typically 3–7nm) around a single emission line — **Hα** (656nm), **OIII** (501nm), or **SII** (672nm). Because nebulae emit strongly at these lines while light pollution and moonlight are broadband, narrowband filters reject 99%+ of unwanted light, enabling deep emission-nebula imaging from bright cities and even under a full moon. The cost: only emission targets benefit (not galaxies/broadband), and per-channel exposures are long. Narrower filters reject more pollution but cost more and demand precise focus.

---

## RGB, LRGB & The Hubble Palette
Source: [https://stackingstarlight.com/advanced/hubble-palette](https://stackingstarlight.com/advanced/hubble-palette)

Assembling color: true-color RGB, LRGB (luminance for detail), and the SHO Hubble palette (SII→R, Hα→G, OIII→B) false color.

Color is built from channels. **RGB** maps red/green/blue filters to their natural colors for true color. **LRGB** adds a high-SNR **luminance** layer for detail while RGB supplies color — efficient and sharp. The **Hubble (SHO) palette** is *false color*: it maps the **SII**, **Hα**, and **OIII** narrowband channels to **red, green, blue** respectively, revealing the chemical structure of nebulae as the gold-and-teal look the Hubble Space Telescope made famous. Palettes are creative and scientific choices, not fixed truths.

---