mxbmrp3
MXBMRP3 - Usage Survey Report
Data window: 2026-06-27 → 2026-09-13 (79 days, 62 observed)
Generated by tools/analytics_report.py from Aptabase exports - do not edit. Re-run the tool after each monthly export; the raw exports stay out of the repo, and the charts land in charts/.
| Installs | Launches | Countries |
|---|---|---|
| 14,355 | 478,558 | 100 |
Highlights
- Reach: 14,355 installs across 3 games in 100 countries; peak 3,191 active on a single day.
- Main game: MX Bikes - 99% of launches.
- Platform: among installs reporting each field, 78% run Windows 11 and >99% use Steam.
- Repeat use: 70% of installs launched more than once.
- Most-used HUD: Standings - 76% of MX Bikes installs.
Activity over time
Daily charts omit partial export days (2026-06-27, 2026-09-13); totals include them.
-
Installs seen to date: 14,355
-
avg launches/day: 7,330
Games
| Game | Installs | Launches | Share of launches |
|---|---|---|---|
| MX Bikes | 14,188 | 472,926 | 98.8% |
| GP Bikes | 127 | 4,754 | 1.0% |
| Kart Racing Pro | 40 | 878 | 0.2% |
Plugin version adoption
Each install counts once, at its most recent version (an upgrade moves it, never double-counts). Versions under 10 installs are grouped.
By release line: 1.28 4,591 (32%) · 1.27 3,170 (22%) · 1.30 2,467 (17%) · 1.29 2,302 (16%) · 1.26 1,825 (13%)
Update channel: stable 14,347 (>99%) · prerelease 8 (<1%)
Geography
Operating system
Repeat usage
- Returning installs (launched more than once): 10,068 (70%) of 14,355 reporting · median launches per install: 12
-
Median session: 16 min · 276,220 sessions · from 75% of launches on versions that report one; 110,435 launches on earlier versions report no session end
-
At least 6,995 days of session time across those sessions
Achievements
- Achievement tiers earned, median per install: 16% · installs holding at least one achievement: 2,282 (94%) of 2,438 reporting
- Achievements unlocked, median per install: 30
-
Earned so far: 64 of the 82 listed achievements, which is what the chart shows. 32 hidden ones are never charted or named - they still count towards the totals above, which is why that percentage can pass 100%.
-
Rarest, of those earned by anyone: Photo Finish 2 (<1%) · Made It Yours 3 (<1%) · Tinkerer 5 (<1%) · Split Decision 5 (<1%) · Metronome 6 (<1%) · On Air 8 (<1%) · Brick Breaker 9 (<1%) · Factory Fresh 17 (<1%) · Mudder 21 (<1%) · Clean Sheet 40 (2%)
Feature & HUD adoption - MX Bikes
HUDs and widgets differ by game, so adoption is shown for the primary game, MX Bikes, over its 14,188 installs. (GP Bikes 127 and Kart Racing Pro 40 have too few installs for their own breakdown.)
- Median HUDs enabled per install: 5 · median widgets: 7
Crashes (upstream / stability)
Of 46,246 crash reports, 43,841 come from plugin 1.27.5+ builds, which record full diagnostics (a backtrace and the access-violation type). The rest are excluded from the analysis below: 2,394 from earlier builds and 11 from dev tooling.
- Crash-report rate: 12.7% (43,841 of 345,363 launches)
- Affected installs: 6,165 of 11,967 running 1.27.5+ (52%)
The plugin’s crash handler detects a fault, saves a report, and submits it on the next launch. Where a fault location was recorded, execution failed outside the plugin binary (in the game or another module), though that location is not necessarily the root cause. Reports are grouped by faulting module below and linked to the known-crash list, where a fix or workaround may exist.
Crash rate by game
| Game | Crash reports | 1.27.5+ launches | Launches with a crash report |
|---|---|---|---|
| GP Bikes | 357 | 3,548 | 10.1% |
| Kart Racing Pro | 130 | 661 | 19.7% |
| MX Bikes | 43,354 | 341,154 | 12.7% |
The Plugin (MXBMRP3) slice (44 crashes) is where the fault offset landed, not proof the plugin caused it - inspect each via its backtrace.
Access-violation type: read 37,434, write 4,873, execute 1,304 · exception codes: 0xC0000005 43,612, 0xE06D7363 149, 0xC0000094 72, 0x80000001 2
Most common crashes
Ranked by number of crash reports. 75% (32,768) of reports match signatures already catalogued in known_game_crashes.json; each links to its full write-up.
| Crash | Share | Trigger | Fix / workaround |
|---|---|---|---|
| Offline track-load crash (bad string in per-profile data) | 39% (17,231) | At track load the game passes a garbage string pointer to an sprintf-family call and the CRT walk… | ✔ |
| Physics contact blow-up (bad table index) | 18% (8,046) | During a physics contact event – a hard object impact, but also ordinary terrain/ground contact… | - |
| Session teardown null dereference (leaving a session / connecting) | 8% (3,661) | While the player is idle between sessions (not riding), the game dereferences a null object point… | - |
| OpenGL render buffer write (vertex-fill overrun) | 7% (3,044) | The game’s OpenGL render code (a +0x240xxx vertex/command-buffer fill loop handling ~4001 element… | ✔ |
| Session-teardown crash via input/message pump (stale pointer) | <1% (365) | While leaving/ending a session (teardown, idle after the race), the game dereferences a stale/gar… | - |
| Null dereference in network / session-end code | <1% (256) | At the end of an online race the game dereferences a null object pointer (reads field at null - 0… | - |
| Out-of-range array index read at session end | <1% (164) | As an online race session ends, the game indexes a 32-bit array with a roughly 985-million-elemen… | - |
| 64-bit pointer truncated to 32-bit (stack above 4 GB) | <1% (1) | Game code narrows a 64-bit pointer to 32 bits (e.g. a ‘mov ecx,edx’ where a full 64-bit move was… | ✔ |
✔ = a documented workaround (follow the crash link). Matched on game build + fault offset.
Not yet catalogued
The remaining 25% (11,073) of reports do not yet match the catalogue. The most frequent unmatched signatures are listed below by module and per-build offset:
| Fault (module + offset) | Category | Share |
|---|---|---|
unknown+0x0 |
Unknown | 3% (1,277) |
mxbikes.exe+0x1902be |
Game | 1% (599) |
MSVCR90.dll+0x3ec27 |
System / runtime | 1% (477) |
mxbikes.exe+0x1e403c |
Game | 1% (471) |
mxbikes.exe+0x1d742c |
Game | 1% (448) |
mxbikes.exe+0x1a29ae |
Game | <1% (375) |
mxbikes.exe+0x255f1f |
Game | <1% (374) |
mxbikes.exe+0x2889e0 |
Game | <1% (305) |
mxbikes.exe+0x255ef0 |
Game | <1% (304) |
mxbikes.exe+0x1eea9c |
Game | <1% (249) |
About this data
The plugin only sends anonymous, aggregate telemetry (no personal data - see the privacy note in the main README). Reporting began in 1.26 (a few stray older builds send only a basic launch event) and expanded through 1.27, so older versions report fewer fields. Percentages are always taken over the installs that actually report a given field, and each chart notes how many installs that is, so incomplete rollout is never shown as a real trend.
Definitions. Install - a unique install_id (regenerated if the analytics file is deleted, so a wipe-and-reinstall reads as a new install). Launch - one plugin start (app_started). Active install - an install that launched on a given day. Crash report - one fault caught by the crash handler, saved on crash and sent on the next launch (≈ one per crashed launch). Each install belongs to one game (the plugin installs separately per game); its game, country, and version are its most recently seen values.
Developer/test machines are excluded report-wide (3 installs, 1,294 events): they launch every dev build and deliberately trigger crashes to validate the telemetry, which would otherwise read as phantom plugin crashes.
What each release line reports (share of its launches):
| Release line | Launches | Features | HUDs / widgets | OS version | Update channel | Panel theme | Crash detail |
|---|---|---|---|---|---|---|---|
1.25 |
1 | 0% | 0% | 0% | 0% | 0% | 0% |
1.26 |
110,579 | 100% | 100% | 0% | 100% | 0% | 0% |
1.27 |
147,389 | 100% | 100% | 100% | 100% | 0% | 85% |
1.28 |
123,622 | 100% | 100% | 100% | 100% | 0% | 100% |
1.29 |
68,930 | 100% | 100% | 100% | 100% | 100% | 100% |
1.30 |
28,037 | 100% | 100% | 100% | 100% | 100% | 100% |