Skip to content

EDEXO 1.4.0 — portable

Choose a tag to compare

@bahuckel bahuckel released this 16 Nov 15:47

A journal-linked exobiology and colonisation companion for Elite Dangerous. Compare reads your local journal, merges scans and organic signals, and predicts the candidate species on a body from FSS/DSS telemetry alone — with payout ranges, first-footfall tracking, a system map, a customisable HUD and a species encyclopedia. Colonize follows your construction sites and plans the hauling, trip by trip.

This release is the single-file portable build. For the unpacked program folder plus the two console builds, see v1.4.0-zip — same version, same code.

Download EDExoCompare.exe and run it. Nothing is installed.

The species model is rebuilt on all the data: every profile, the body-type table and the calibration now come from every confirmed sighting in a year of EDDN — 2,065,004 bodies, codex entries and composition scans counted, plus the old corpus' sightings the year lacked and Spansh's bodies of the one-species genera — instead of 145,829 bodies from the old corpus alone. "-rich" atmospheres are their own atmosphere, as the game treats them: a Carbon dioxide-rich world is no longer read as a Carbon dioxide one, and the species lists name the "-rich" types each species is actually found on. On very light worlds (under 0.047 g) thin carbon dioxide life grows past the 195 K ceiling it keeps everywhere else, and EDEXO now shows it there. Ranges follow the data with one rule: a lone find never sets a range.

New

  1. A header picture of where you are (Options › Look › Header picture: Turning, Still, Calm or Off, kept per PC): the ground of the body on show under a sky that turns once in 30 minutes, in your colour scheme. The skyline is drawn from the system's name, so a system looks the same on every visit; the land follows the kind of body (sharp ranges on an icy one, craters on a rocky one, mesas on a metal-rich one), and the plants are the genera EDEXO shows for that body, the likeliest in the middle; a body without biology is bare ground. The sky turns only while Motion is Full and the window is on show, and it is painted off the main thread.
  2. Interface styles (Options › Look › Interface style, kept per PC, switch at once): Classic, the app as it was; Holo, a cockpit hologram with corner brackets, glowing titles and scan lines; Nebula, glass panels over a sky made from your colour scheme. Text in both carries a dark edge so it reads on a light ground.
  3. Faster to open: Compare's window draws in about 0.1 s instead of 0.5 s, and the system map and the notable-body pop-up open in 30 ms instead of 320 the first time. Species photos fade in as they load, and body tabs on a phone are no longer cut off.

Changed

  1. Every species profile from every sighting. 117 profiles rebuilt from every confirmed sighting of the EDDN year (codex entries and composition scans count as sightings), the 2,083 sightings of the old corpus the year lacked, and, for Bark Mounds, Amphora Plant and Crystalline Shards (each the only species of its genus), the 7,449 bodies Spansh's galaxy dump lists with that genus: 4,215,072 species-on-body sightings, none sampled away; only finds without a body record and 332 lone outliers are left out, each listed in the outlier register. Aurasus alone is drawn from 330,785 bodies, tectonicas from 275,049. Before, the profiles held 145,829 bodies in all and the ranking prior 27,652. Thargoid species and Radicoida Unica are kept as they were.
  2. What grows on a body of this kind, the table behind the first guess, is counted on 1.79 million species finds over bodies where every genus was sampled (was 51,272 finds over 13,983 bodies). Where the old table had 39 bodies to go on, the new one has thousands: on icy neon worlds with nitrogen magma Bacterium acies leads (37.8 % of 8,458 bodies), with water magma verrata (36.5 % of 8,423). Against each species' own profile, now counted on every sighting too, the table weighs 0.3 (was 0.48): the most that keeps acies first on those nitrogen-magma moons.
  3. "-rich" atmospheres are their own type. The game names an atmosphere "X-rich" when X is the second gas behind one the label does not name; it is a different atmosphere to the plants, not more of X. Bacterium aurasus grows on 0 of 2,998 Carbon dioxide-rich bodies and Stratum tectonicas on 2,430 of them; Bacterium acies on none of 12,214 Neon-rich ones. EDEXO kept them together, so aurasus was offered on Carbon dioxide-rich worlds at up to 60 %. Now each species' atmosphere list names the "-rich" types it is found on — added for 14 species (tectonicas, paleas, cucumisis, frigus and limaxus on Carbon dioxide-rich; Electricae pluma and radialem on Argon- and Neon-rich; Osseus pumice; Tussock capillum; Brain Trees, Sinuous Tubers and Crystalline Shards) and taken off 20 whose codex rows named one they have never been found on. A species off its list is listed as unlikely, not hidden.
  4. Thin carbon dioxide under 0.047 g: the 195 K ceiling every CO₂ species stops at does not hold on the lightest worlds. Every body above it carrying the whole community (eight species on Blu Thua RN-S c17-23 7 b at 196.7 K) is at 0.042-0.044 g; at 0.1 g and above, 471 bodies between 195.5 and 200 K average 1.1 signals and none has four. On those worlds the ceiling is now 200 K (the warmest such body in the galaxy is 199.2 K).
  5. Observed temperature ranges are back for every species: a species is demoted outside the temperatures it has been found at across all its bodies (outliers cleaned), as it was before profiles came from the EDDN year. Bacterium aurasus at 402 K, past every one of its 330,791 carbon dioxide bodies, is shown at ×0.02 again.
  6. "Chance here" recalibrated on the rebuilt model: before a DSS every band of the shown chance now comes true as often as it says, within about a point (372,812 fully identified bodies of the EDDN year: 10-20 % said 14.4, came true 14.1; 90-100 % said 99.2, came true 99.1). The "why this chance" popover counts on the same bodies. Two species with the same chance keep the model's own order.
  7. Never-found-here rules checked against every sighting: 22 of 198 that the full data contradicts are dropped; the colour-star rules and the firm temperature edges all hold. Tubus sororibus reaches 195 K (was 190) and Brain Tree viride 255 K (was 220), as 13 and 9 systems show.
  8. Ranges re-measured on the full data, and the rules that rested on the old corpus checked one by one: Stratum tectonicas' gravity cap holds on 275,312 bodies, Bacterium tela's rule (volcanism, or 300 K and up) on 65,180, the argon and neon bands that split Fonticulua campestris / upupam and Bacterium acies / Fonticulua segmentatus on thousands each; Stratum paleas starts at 165.0 K exactly (139,096 bodies); Fonticulua digitos grows round K to Y stars alike, so its host star no longer rules it out anywhere.
  9. Crystalline Shards' chance: before a DSS it read 48.6 % on every body that meets their rules, and 93 % of those bodies have them; it now reads 92.7 %, corrected like every other genus (2,003 such bodies of the EDDN year: 93.2 % had them). Once a DSS names their genus they read 100 %, not 48.6 % (6 % with more than one signal) as before.

Measured

On 9,512 bodies of a hold-out set no part of the model was built from: species missed before a DSS 18 (unchanged), the found species first in its genus 94.8 % (as before), wrong rows on bodies where every genus was sampled 0.82 → 0.81. On the whole EDDN year (1.37 million species rows a mode, each system's other planets known as they are after an FSS): shown before a DSS 99.98 % (was 99.97 %; 80 rows gained, 12 lost), after a DSS 100.00 % (was 99.99 %; 54 gained, 6 lost), the found species first in its genus 94.59 % before a DSS and 94.60 % after (were 94.57 / 94.58 %), and the chance shown for the species really there 87.2 % on average before a DSS (was 86.6 %) and 94.4 % after (was 94.3 %). The rows lost are rare finds in the thin tail of a species' range, still listed as unlikely. On the developer's journals: 792 species found, none missed.

Going back to 1.3.9 is safe; predictions are made again from the older profiles.

Where your data lives

%LOCALAPPDATA%\ED Exo Compare\ — settings, your LAN key, your EDSM credentials if you add them, the species you have scanned on foot, your Fix corrections, the journal folder you chose, your HUD layout, where your windows were, the systems you looked up on Spansh (kept 30 days), the shared exomastery folder, and the prediction caches. Nothing of yours is kept in the program folder, so replacing it on an update loses nothing. On Linux the same lives in ~/.config/edexo-compare. Backups go only where you point the Backups tile.

Your journals are read, never copied out. Uploads to EDSM, Canonn and EDDN happen only if you turn them on yourself. Looking a system up sends only its name or number to Spansh.

Windows will say the publisher is unrecognised. The executable is not code-signed, so SmartScreen asks first: "More info" → "Run anyway".

Licence

Code: PolyForm Noncommercial 1.0.0 with two additional permissions (showing the app in streams and videos is free; releases up to and including 1.2.20 stay MIT). Data: CC BY-NC-SA 3.0 (credit, no commercial use, share alike; part of it is derived from EDAstro) — see data/LICENSE-DATA.txt. The species photographs are Elite Dangerous screenshots: Frontier's game art, under Frontier's terms — NOTICE.md names every contributor and says which images are borrowed from the ED-DSN community. The list of Frontier-populated systems and looked-up systems come from Spansh.