Changelog
What changed, and when
Every change that reached the site, newest first. Each date is the day the change went into the site. The site went live on 25 September 2026, so the earlier entries are the work before launch. Errors in published figures have their own record in the corrections log.
46 entries, 15 August 2026 to 13 October 2026
October 202619 entries
Field of View & Framing
Any of the 13 targets drawn at true angular scale against a telescope and camera from the catalogue or typed in, at any camera angle, with the field, the image scale, how much of the frame it fills and the mosaic that covers it when it does not fit. Two rigs can be compared on one target, and the five galaxies with an OpenNGC position angle are drawn north up. Rig pages and Rig Reality Check builds open it with their own telescope and camera.
Rig Reality Check: a shared link opens in a fraction of the time
The builds for a link are worked out on the server once and kept, and the page carries the builds it shows instead of the whole gear catalogue, under a third of its old size. The catalogue loads when you first change an answer. Every result is the same as before, build for build.
Session Checklist: your own steps, your order, several lists
Press Customize to add your own items and phases, reword or hide the built-in items, and put everything in your order. Keep a list for the pier and another for a dark-site trip; they stay in your browser, and a file or a link moves one elsewhere. Built-in items keep their sources, and yours are marked as yours.
Rigs: every telescope with every camera
A page for each of 2,185 telescope and camera pairings, computed from the catalogue: image scale, field and sampling, the 13 framing targets, the corrector's spacing, the mounts that carry it and the shortest sub under three skies. Pairings whose sensor is larger than the stated image circle are listed on each telescope's page with the numbers. Rig Reality Check builds and Backfocus Solver results link to them.
Diagnose: the failure atlas, with a symptom picker
Tick what you see in a bad frame and the 29 failure modes that could produce it are ranked, with the look-alikes to rule out. Each mode has a page with its causes in order of likelihood, the fix for each, and what one frame cannot tell you. Session Checklist items now link the failure they prevent.
A changelog, a corrections log and a colophon
This page, a log of every data error found and fixed since the first commit, and a colophon listing what the site is built with and the sources it credits. Pages whose published figures changed now carry a dated update note where the change happened.
Method page: the exposure example at 1.5 e⁻
The worked example still ran on ZWO's headline 1.0 e⁻ after the physics reference moved to the 1.5 e⁻ on ZWO's chart. It now matches the reference and the planner: an 8.1 s shortest sub without a filter, where it showed 3.6 s.
Guides, and an essay on what astrophotography costs
The first year at $1,500, $3,000 and $6,000, itemized, including what the catalogue cannot price yet. Every figure in it is computed by the Rig Reality Check's engine when the site is built. Guides is a new section in the header.
Fixes for phones, and a working solver on the home page
An audit at nine screen widths found 85 broken route and width pairs, and now finds none. The solver scrolled sideways at 320 px, form fields made iOS zoom in, and footer links were too small to tap. The home page's train is now a small solver running three cameras with the rings from their own boxes.
Sky from your location, for the lower 48 states
The Exposure Planner and the Rig Reality Check can estimate your sky brightness from your position, with the model on the sky brightness method page and 72 map tiles served by this site. Your coordinates stay in your browser. Outside the lower 48 it says the area is not covered yet.
Rig Reality Check: sensor type from the makers, built-in guiding, reducers
Mono or color now comes from each maker's own statement, which moved ten mono cameras off color pricing. ZWO's DUO and Air cameras no longer get a separate guide camera, and the Air cameras no control computer. A reducing corrector now shortens the focal length a build is computed at. The scoring weights are unchanged; scores moved only where a build's inputs did.
Tube rings, Celestron reducer threads and fifteen more cameras
Ring sets priced for the two Takahashi scopes that ship without them, and Askar's FRA400, FRA400C and FRA600 recorded as shipping rings and a plate. Celestron's f/6.3 and EdgeHD 8 reducers now exit on the SCT thread, so the solver reports trains through them as blocked until Celestron publishes its T-adapter lengths. Prices or weights for fifteen more cameras, among them the SkyEye line and the QHY600M.
The home page stopped promising a resale estimate
Its commitments said every build carries a resale estimate. None does yet, because there is no data source for one, and every build says so.
A redesign, with night mode on every page
The tools share one instrument layout: inputs, then the result in a frame, then the detail and how it was computed. The method pages are set as a cited paper with numbered notes, and every page shows the date it last changed, taken from its own history. Red night mode moved from the checklist to a button in the header, and covers the charts, photos and 3D model.
Bortle 2 preset corrected to 21.75 mag/arcsec²
Picking Bortle 2 in the Exposure Planner or the Rig Reality Check used 21.9, a value the site's own table counts as class 1. It is now the middle of the class 2 range. The table's bounds did not change.
Copy pass across the site
Every visible string was reworked for plainer wording and punctuation. The observatory guide is now in the builder's first person, with its facts and numbers unchanged.
Observatory 3D model
The deck, pier and dome generated from the build spec, as designed and as built, with layer toggles, four preset views and both models to download. Parts the spec does not fix are drawn hatched and listed as approximate.
A links page for the Instagram bio
The tools, the observatory guide, the method page and the YouTube channel on one page. No link on it carries a tracking parameter.
Observatory guide: design and as built, side by side
The builder measured the finished observatory. The chapters keep the design and add what was built: the pier top 28⅞″ above the deck, ¾″ under the design; the adapter gap 2½″ and level; 21 deck boards with the middle one ripped; and a 2×8 rim where the plan had a 2×10. Two questions stay open.
September 202624 entries
Rig Reality Check relaunched as a beta
Back in the navigation after matching the prototype it was built from: a framing view and a preview of the data for each build, a backfocus check, and the carry weight with a 10 kg travel line that never changes the ranking. The budget is checked on hardware, and a build whose first year runs past it is flagged.
Essentials priced, and the twelve-month true cost
Guiding, dew control, power, filters, the control computer and storage are priced from 30 sourced rows over 13 slots. Each build shows hardware plus essentials. The total is displayed only and never enters the score.
Rig Reality Check launched as a beta, then unlisted the same day
The scoring weights were signed off as version 2026-09-27 and published with the tool. Later that day it came out of the navigation until it matched its prototype. The weights stayed signed off.
Catalogue gaps closed where a maker source allowed
Eleven tripods pair 14 of the 18 head-only mounts, each pairing a maker's own statement. Correctors were linked and priced for ten scopes, the RST-135E priced from Rainbow's listed US dealers, three ZWO mono cameras added, and the EdgeHD 9.25/11/14 reducer row split into the three products it described. At $1,500 the tool still says the catalogue is incomplete.
Camera and corrector prices
28 camera and 17 corrector rows with dated prices and the makers' weights. When no build ranks only because prices are missing, the tool now says the catalogue is incomplete. Before, it advised against buying.
Telescope catalogue
45 optical tubes and 8 smart telescopes keyed from manufacturer pages and manuals, every row marked unchecked until it is checked by hand. Most of the classic beginner line-up turned out to be discontinued, so the current successors were keyed instead.
Mount catalogue
29 mounts. Payload is the maker's figure, never derated in the data. Periodic error is stored with the maker's own convention, since no two publish it the same way, and the lever arm or moment a strain-wave rating depends on is recorded with it.
Natural sky set to 21.83 mag/arcsec²
The sky-brightness model's natural zenith is now the median the National Park Service measured on 295 nights, in place of the textbook 22.0. The bias against measured total-sky brightness fell from +0.15 to +0.02 mag, and the model can no longer return Bortle 1.
Physics reference worked examples corrected
The exposure examples did not reproduce from their own inputs. They now use 1.5 e⁻ read noise throughout, and three pending tests built on them now run.
Sky brightness from satellite data
Our own model of zenith sky brightness from VIIRS night-time radiance, published with its formula, its data licenses, its accuracy against measurements it was never fitted to, and what it cannot tell you.
Sub-Exposure & Calibration Planner
21 sensor profiles. Read noise comes from each maker's own table or chart, and every curve says how it was obtained. Only published gains are offered, never interpolated, and the calibration plan says when a maker is silent about amp glow.
Session Checklist
77 start-of-night and end-of-night items in 11 phases, filtered to the rig you own, with a red night mode and a print layout for sites with no signal. Two items are marked unverified because no manufacturer source backs them.
Observatory build guide
Ten chapters and four reference pages on the founder's backyard observatory, a NexDome on a floating deck with an isolated pier. Every table renders from the build spec, and the photos are stripped of metadata before they are published.
A phone menu
The header stopped wrapping at phone width. The sections sit behind a menu that works without JavaScript.
Correctors that publish a focus range
ZWO's eight FF-APO refractors and the Askar petzvals publish the longest train they can focus, and sometimes the shortest. The solver treats a maximum as a maximum and says how far the focuser racks out.
ZWO's manual index swept
Cameras added from ZWO's own manuals, among them the ASI2600MM DUO, ASI585MM Pro, ASI4400MC Pro, ASI461MM Pro and the Air cameras. The ASI294MC Pro's with-ring figure, an open question until then, is printed on its June 2026 manual's chart.
ToupTek cameras, filter wheel and off-axis guider
24 cameras and two spacer kits from ToupTek's own manuals, plus the AFW filter wheel and the OAG with its three telescope-side adapters.
Filter wheels, off-axis guiders and bolt-on parts
ZWO's EFW wheels, the ZWO OAG and OAG-L, and the tilt plates a camera can swap for. Spacing stays behind an off-axis guider, where ZWO's own charts put it, so the guide camera can still reach focus.
Thread gender at every junction
The solver checks which way each thread faces, so it cannot offer a ring that would not screw on.
The solver asks about the filter drawer
The question is now required, because a 21 mm drawer left out is a bigger error than any tolerance. ZWO's two current drawers are catalogued with the rings in their boxes, and the drawer's filter glass is counted.
Home page, method page, share cards and a 404
Every number on the home page is computed from the same rows and solver the tool uses, so it cannot go stale. The method page works the backfocus model through the same train. Links pasted into a forum unfurl into generated cards.
One design vocabulary
Links moved from a blue that read like a chart series to the brand violet. Geist became the typeface, with tabular figures wherever numbers line up in a column.
firstrig.com went live
The Backfocus Solver, the home page, the method page and the about page, served from the founder's own server with Cloudflare in front. Cookieless analytics followed on 26 September.
Rings are threaded parts, solved with their lengths
Spacer kits had been lists of lengths, which sent an ASI2600 Pro owner with an M48 corrector to buy an adapter they did not need. Every spacer, extender and adapter is now a threaded part with an optical length, and a kit holds one of each piece. The solver never picks a thread adapter for you; it asks which one you own.
August 20263 entries
Backfocus Solver built
Cooled cameras, camera bodies with their T-rings, and correctors, each figure re-keyed from a manufacturer page or drawing and linked to it. Where a camera ships in more than one configuration the solver asks which you have, because ZWO's 17.5 mm means two different things.
Optics and backfocus formulas, tests first
Image scale, field of view, the Dawes and Rayleigh limits, the Airy disk, the critical focus zone and the backfocus model, each with a unit test built from a worked example in the physics reference. Writing those tests found two errors in the reference itself.
Project started
firstrig.com registered. The specs, the decisions log and the physics reference written, and the site scaffolded.
Last updated