Everything FLARE does, in one place.

Section 2 · Desktop application

A small WPF / .NET 10 app for MSFS 2024.

A standalone Windows window, not a sim add-on. Talks to MSFS 2024 over Microsoft's SimConnect SDK and runs in user space, with no kernel hooks. Two jobs: capture and grade landings, and (unless you turn it off) sync them to your account.

The window is designed to live on a second monitor. It never competes with MSFS for cockpit real estate, so your EFB, panels, and views inside the sim stay untouched. It shows live data while you fly, and as you roll out after touchdown a score card pops up over the sim with your grade and the top coaching point. It closes itself after about 30 seconds, and the full breakdown is on the Landing page.

Landing capture & scoring

Every touchdown is detected automatically. No button to press. For each landing, FLARE captures:

Each landing is graded A+ → F. Sub-scores cover descent rate, G-force, centerline accuracy, touchdown-zone position, approach stability and sideslip, with bounce and tail-strike penalties and a crosswind bonus. Thresholds adjust by aircraft type, a 737 isn't expected to land like a Cessna. Full breakdown in landing scoring and the aircraft profiles catalog below.

Runway detection (four-tier)

FLARE figures out which runway you landed on through a four-tier fallback so missing MSFS data doesn't silently turn into a free score:

1. The simulator's own runway data: FLARE asks MSFS for the runway geometry at the airport you're lined up with on final, and again at touchdown, and matches your touchdown to it. Because this is the runway exactly as the sim renders it, centerline and touchdown distances stay accurate even at airports where public map data disagrees with MSFS scenery.
2. MSFS ATC: when the sim's runway data doesn't cover your touchdown, FLARE uses the runway ATC has active, but only if you actually flew along it. ATC names its active runway wherever you are, so a runway you crossed or set down beside isn't used. MSFS measures ATC's touchdown distance from its aiming point rather than the threshold, so that distance comes from the runway data instead, or isn't scored.
3. GPS lookup against bundled navdata: failing both, FLARE looks up the runway by your touchdown coordinates against a bundled OurAirports SQLite database, over 72,000 airports and around 31,000 runways, CC0 public-domain data shipped inside the installer.
4. Not detected: when everything misses (fictional MSFS sceneries, genuine off-airport landings), centerline and touchdown zone are excluded from the weighted total rather than awarded a free 100. The same goes for a touchdown that clearly wasn't on the runway, like a helicopter on a pad beside it or a floatplane on the water. Honest grades over inflated ones.

The score card adds (via GPS) after the runway when it came from the airport database, and the Runway data line on the Landing page names the source (MSFS facility data, MSFS ATC, GPS lookup or Not detected), so you always know how the runway was identified. Parallel-runway suffixes (15L vs 15R) come from MSFS when it shares them and from the navdata otherwise, so a touchdown that MSFS reports as just "15" gets disambiguated to the correct side.

FLARE checks for a fresh airport database once a day, when you launch the app. FLARE → Settings → NavData shows the current version, the airport count, the magnetic model in use and when it last refreshed, and has a button to check for a newer one right away. No internet connection is needed in flight; the lookup always runs against the local copy.

Non-runway surfaces

When you land at a heliport, seaplane base, or grass strip, FLARE records the surface type and the landing page in your portal shows a context chip ("Heliport landing", "Water landing", "Grass strip"). Useful when a fixed-wing touching down at the wrong surface type would otherwise look like a normal landing in the logbook.

Local logbook & export

Every landing is saved on your PC in a SQLite database under %LOCALAPPDATA%\FLARE, including the ones you fly offline. Landing History lists them newest first, 50 to a page, and filters by aircraft, airport and date range; double-click one to open it on the Landing page. Settings → Data Export saves every landing on the PC to CSV or JSON, and Download raw data on the Landing page saves one landing's second-by-second trace.

FLARE never deletes the database by itself, but the uninstaller asks what to remove and has every box ticked by default. Untick Landing history to keep your landings on the PC. Landings already synced stay in your account either way.

Sync to the customer portal

Once FLARE is activated, each landing uploads to your private account by itself shortly after the flight. If it can't (you're offline, say), FLARE retries at startup and every few minutes. Settings → Licensing → My landings → Sync now sends them straight away.

Sync also runs the other way. Settings → Licensing → Download my landings pulls everything in your account down onto this PC, which is how you populate a second computer or recover after a Windows reinstall. Landings already on the PC are matched and skipped rather than duplicated. The second-by-second approach traces stay on the PC that recorded them.

Deletions travel in both directions too. Deleting a landing in the desktop app removes it from your account (so the PC has to be online), and deleting one in the portal removes it from every PC the next time each syncs.

Sync is on by default. To turn it off: Portal → Settings → Account → Sync landings to FLARE-Web → toggle off. Takes effect immediately, and your landings keep being saved on your PC. Existing synced landings stay on the server until you delete them. With sync off, uploads and downloads are both refused, so "Download my landings" will not work until you turn it back on.

Section 3 · Landing scoring & grading

How FLARE grades a landing.

FLARE's grading isn't a vibes-based system. Every threshold maps to an industry standard from the FAA, EASA, the Flight Safety Foundation, or aircraft manufacturers' Flight Crew Operating Manuals. Centerline, sideslip and bank are scored the same way for every aircraft. Descent rate, G-force and the touchdown zone depend on the aircraft's category: the tables below show the curves FLARE uses for an aircraft it doesn't recognize, and every category's own curves are in aircraft scoring profiles below.

Composite formula

Every landing produces a 0–100 score computed in three stages:

raw      = Σ (component × weight)  −  bounce_penalty  −  tail_strike_penalty
adjusted = raw × wind_difficulty_factor
final    = clamp(adjusted, 0, 100)

Six component scores feed the weighted sum, with two penalties and one multiplier shaping the final number:

The final score maps to a letter grade:

1. Descent rate at touchdown (30%)

What we measure: the aircraft's vertical speed in feet per minute at the moment the wheels touch the runway, taken from the sim's dedicated touchdown reading (PLANE TOUCHDOWN NORMAL VELOCITY). If FLARE's own recording shows the sim missed the contact (it has reported almost nothing for a crash), FLARE uses the rate from its recording instead.

Why this can differ from your VSI: the sim exposes more than one plausible source for vertical speed and they do not agree. The vertical-speed instrument is damped and lags, so at the instant of contact it can read noticeably softer or harsher than the touchdown actually was. FLARE scores the value captured at contact, which is the one a real flight-data recorder would use. If your VSI showed 150 and FLARE says 210, neither is broken; they are measuring different things.

Why it's the heaviest weight: sink rate at touchdown is the single most reliable indicator of a soft landing. Transport aircraft are certified (14 CFR 25.473) for touchdowns of 600 fpm at maximum landing weight and only 360 fpm at maximum takeoff weight. For light aircraft, even a "firm" Cessna touchdown is rarely more than 200 fpm.

Each category has its own curve, listed in the catalog below. This is the one FLARE uses for an aircraft it doesn't recognize. Scores fall in a straight line between the points.

2. G-force at touchdown (20%)

What we measure: the peak vertical load factor around touchdown (SimConnect G FORCE), expressed as a multiple of normal gravity (1.0 G = level flight). A reading below 1.0 G can't happen at a real touchdown. It means the sim didn't report G for that flight, so FLARE leaves G-force out and grades the criteria it did measure.

Why it's separate from descent rate: two landings with the same touchdown fpm can produce very different impact loads depending on aircraft mass, gear stroke, and how progressively the pilot let the weight settle on the wheels. G-force captures what the airframe actually felt. A smooth airliner landing usually registers 1.1–1.3 G; structural inspection thresholds sit around 2.0–2.6 G for transport aircraft.

As with descent rate, each category has its own curve, and some aircraft carry their own on top (the Airbus airliners, the CRJs and the Dash 8s among them). This is the one for an aircraft FLARE doesn't recognize:

3. Centerline deviation (15%)

What we measure: lateral offset (in meters) between the aircraft and the runway centerline at touchdown, computed from the aircraft's position and the runway's geometry. The same curve applies to every aircraft.

Why it matters: the FAA's airman certification standards explicitly require landings "with the airplane's longitudinal axis aligned with and over the runway centerline." A 5-meter offset on a typical 30 m wide runway puts you visibly off-center; 10 m starts threatening edge lights on narrower airstrips.

FLARE finds the runway from the sim's own runway data first, then MSFS ATC, then its bundled airport database. If all of them miss (fictional scenery or an off-airport landing), centerline and touchdown zone are excluded from the weighted total rather than awarded a free 100. Your grade reflects only what was actually measurable. See Runway detection for the full chain.

4. Touchdown zone accuracy (15%)

What we measure: longitudinal distance (in meters) from the runway threshold to the touchdown point.

Why it matters: the FAA defines the official Touchdown Zone as the first 3,000 ft (914 m) of paved runway for runways ≥ 4,200 ft long, or the first one-third for shorter runways. Landing past the TDZ progressively shortens the available stopping distance, and runway excursions are one of the most common landing accidents.

The window depends on the category. For an aircraft FLARE doesn't recognize it's scored like this:

Each category sets its own ideal window: a narrow-body airliner's is 240–600 m, a light single's 50–200 m. Per-category values live in the aircraft profiles section.

5. Approach stability (10%)

What we measure: two pass/fail gates, taken once as the aircraft descends through 500 ft AGL and again at 300 ft AGL. At each gate FLARE checks three criteria in parallel:

• Speed holding steady: how far the speed at the gate has drifted from the speed you flew over the previous 300 ft of descent. The tolerance is a share of the aircraft's Vref that depends on its category, from 3% for fighters to 13% for vintage aircraft (7% for most airliners, 8% for light singles), never less than 10 kt, or ±15 kt where no Vref is known. This asks whether your speed was settled, not whether it was the right number. Being fast or slow in absolute terms is reported separately as approach speed deviation, against a Vref adjusted for your landing weight, and does not fail this gate.
• Bank under control: bank angle under 5°.
• On the correct descent path: this is the authoritative descent-rate check, and what counts as correct depends on whether you are following a published vertical path.
  • On an ILS glideslope, or an RNAV (GPS) approach flown in its final approach mode, there is a published path to fly, so FLARE checks you were within ±1.5° of the standard 3°.
  • On any other approach (visual, or a VOR or localizer approach with no glideslope) there's no path to follow, and 3° is a convention rather than your target. FLARE instead checks that the path you flew was steady, within ±2° of the descent path you had established, plus a 1,500 ft/min limit so a uniformly steep dive is still caught.
  Either way it is skipped below 50 kt groundspeed, where the angle stops meaning anything.

There used to be a fourth check comparing your descent rate at the gate against the average since 1,000 ft. It was removed in v1.2.31 because it was measuring the wrong thing: on any real approach your descent rate must change as you slow and configure, so converging onto a steady rate scored as unstable. A real PMDG 777 CAT III autoland failed it at both gates by 4 fpm and 2 fpm while passing everything else comfortably, and 42% of all landings ever synced were sitting on the resulting floor. The glideslope check above already tests descent rate, geometrically and correctly.

Both gates need all three criteria true to count as "stable":

When a gate could not be measured at all (you started the flight on short final, or FLARE connected late) it is no longer treated as a failure. If neither gate was seen, approach stability drops out of the score entirely and the remaining criteria are re-weighted. If only one gate was seen, you are scored from the gate we did measure rather than punished for the one we missed: 85 or 40 from the 500 ft gate alone, 75 or 50 from the 300 ft gate alone.

Why this matters more than its weight suggests: Flight Safety Foundation analysis of approach-and-landing accidents found unstabilized approaches as a contributing factor in roughly two-thirds of cases. Most carriers' SOPs mandate a go-around if the aircraft isn't stabilized by 1,000 ft AGL in IMC or 500 ft AGL in VMC. We've calibrated FLARE's gates against the latter.

6. Sideslip and bank at touchdown (10%)

What we measure: the absolute sideslip angle (β) and absolute bank angle (φ) at the touchdown event. Each component is scored independently then combined as 0.6 × sideslip + 0.4 × bank.

Why it matters: a sideslip or bank at touchdown puts asymmetric load on the gear, one main wheel takes more than its share. The FAA's preferred crosswind landing technique is to fly the approach in a crab and transition to a wing-down sideslip in the flare, with the longitudinal axis aligned with the runway by touchdown.

7. Bounce penalty (−10 per bounce)

What we measure: a "bounce" is a ground → air → ground transition within 5 seconds of the previous ground contact. Each bounce subtracts 10 points from the composite score (the score is then floored at 0).

How we detect them: after the initial touchdown, FLARE watches the sim's on-ground flag and each main gear separately, so a wing-low bounce where one wheel stays on the runway still counts. The re-contact has to carry a real load (at least 1.3 G) to count; lighter flickers from detailed gear models on a normal rollout are ignored. The window resets on each new contact, so three bounces within seconds of each other all count.

Why a bounce is more than aesthetic: bounces indicate excess energy at touchdown, usually too much approach speed, an early flare that bled altitude before settling, or a momentary porpoise. The FAA's recommended response after a single bounce is to maintain landing attitude and let the aircraft settle; after the second bounce the standard recommendation is a go-around.

8. Wind difficulty multiplier

Crosswind landings ask more of the pilot than calm-air landings, so a clean touchdown in a 25 kt crosswind earns a small bonus. The bonus is a multiplier applied to the raw composite score, and the final score still clamps to 100, so a strong crosswind can lift a B → A but never turns an F into an A.

Why a multiplier and not a raw bonus? A bonus that adds points scales weirdly when other components score well, a 99 base can't go above 100, so the bonus disappears. A multiplier respects the existing balance of the score: a sloppy 50 in a strong crosswind still doesn't become an A.

9. Tail-strike penalty

Every aircraft has a maximum safe pitch attitude at touchdown, beyond which the tail meets the runway. For the Boeing 737-800 it's 9°; for the Airbus A320 it's 10°; for the long Boeing 777-300ER it's just 8°. When FLARE detects a touchdown pitch at or above the limit, the landing earns a −25 point penalty on the composite score plus a "Tail strike risk" warning badge on the landing detail page.

An aircraft uses its own published limit where the catalog has one, and its category's otherwise. Helicopters, eVTOLs and balloons have no tail-strike check. Every limit is listed in the catalog below.

Worked examples

Four landings, same engine, very different stories. All numbers traced through the formula above.

Honest caveats

FLARE's grading is grounded in real standards but a sim is not a cockpit. A few things to know:

• Per-aircraft thresholds are the default. The descent-rate, G-force and touchdown-zone tables in this section are the ones for an aircraft FLARE doesn't recognize. For everything in the catalog (default MSFS and the major payware) FLARE applies the curves for its category. See aircraft scoring profiles.
• The sim is the source. FLARE samples MSFS 50 times a second on approach and through the landing, but every reading is only as good as the aircraft's flight model. Centerline and touchdown-zone data depend on accurate runway geometry, and third-party scenery sometimes places a runway slightly differently.
• Calibrated against VFR / training standards, not airline SOPs. If you're flying an airliner intentionally to commercial-cockpit standards, expect to score in the A range routinely.
• The grading rewards consistency. A single A+ landing is great. Ten landings in a row averaging A is the real signal of skill, and that's what your portal stats track.
• Honest scoring beats inflated scoring. When runway data is missing (rare: fictional scenery or genuine off-airport landings), FLARE excludes centerline and touchdown zone from the composite rather than handing out a free 100 on each.
• "Not measured" is never treated as zero. Any criterion FLARE could not measure is dropped from the weighted total and the remaining weights are rescaled, so your score reflects only what was actually observed. A criterion that was measured can never score 0, because every curve has a floor of at least 10.
• Some landings get no letter grade at all. If the runway could not be identified and the approach was never seen, the only things left are how hard you touched down and whether you were crabbed. That says nothing about where you landed or how you got there, so from v1.2.35 those landings show a dash instead of a letter and do not count toward grade streaks or score-based achievements. Every criterion FLARE did measure is still shown. Before this, one such landing scored a flat 100 and was graded A+ on two criteria.
• Impossible captures count toward nothing. A slew, a reposition, a crash, or the scenery reloading under the aircraft can produce a "landing" that never happened: readings no aircraft could survive, readings too gentle to be a landing at all, an aircraft inverted at touchdown, or one that was not moving in any direction. FLARE checks the whole touchdown for this, and when it does not believe the capture the landing gets no grade, no coaching and no place in your statistics or achievements. It stays in your logbook with an explanation, because it is your data.

Section 4 · Aircraft scoring profiles

Per-aircraft thresholds for 498 airframes.

A 200 fpm touchdown is well below a Boeing 737's certification threshold but at the upper end of acceptable in a Cessna 152. FLARE's per-aircraft profile system applies the right curves for the airframe you're flying, across 28 categories covering 498 named aircraft, default and add-on.

Profiles are bundled with the desktop app as profiles.json. The same source spreadsheet generates both the desktop profile bundle and this catalog, so the numbers below match what the latest version of FLARE applies in flight.

How FLARE matches your aircraft

When MSFS reports the aircraft title (e.g. "PMDG Boeing 737-800 BCF"), FLARE walks the tiers below in order, first hit wins. Every landing page shows exactly which tier and profile it was scored against.

If your add-on isn't recognized or lands on a profile you think is wrong, email support@flaresim.io with the exact aircraft title shown in Landing History. We add aircraft to the library in regular updates.

Per-aircraft scoring examples

Same telemetry, different aircraft, very different scores. This is the per-aircraft system's whole point.

The touchdown-zone window and the approach-speed tolerance adapt by category too, and the tail-strike limit by aircraft where one is published. See the catalog cards below for each category's exact thresholds.

Wind multiplier & tail-strike penalty

Two universal modifiers fold into the composite formula on top of the per-aircraft component scores: a crosswind multiplier (×1.00 – ×1.15 by tier) and a tail-strike penalty triggered when touchdown pitch reaches the per-aircraft safe limit (−25 pts). Both are documented in the grading section above; the per-aircraft pitch limits are listed inside each category card below.

The catalog, 28 categories

Every MSFS 2024 default and major payware aircraft maps to one of the categories below. Click a card to expand the full threshold table, the per-aircraft list, and the cited sources.