New Features

• Intelligent stress test report - When a stress test ends, the recap now opens with a clear verdict (PASS / WARN / FAIL), a 0-100 score, and an answer to the question that actually matters - how many players can this server handle - instead of only raw min/max TPS and memory numbers. The console report, the in-game ops summary, and a saved file all share the same analysis.
• Honest player-capacity estimate - Reports an equivalent real-player range (e.g. ~9-17 real players) derived from the measured per-bot tick cost and the 50 ms tick budget, with an explicit caveat: stress bots only exercise chunk loading, so real players - who also add networking and entity-tracking load this test does not measure - may lower that figure. No more reading "5 bots were fine" as "5 players is my limit".
• Bottleneck diagnosis - Identifies the real limiting resource - the single-threaded main loop, overall CPU, or memory - so the advice fits the actual wall. Because Minecraft's tick loop is mostly single-threaded, it tells apart "one core maxed out" from "the whole machine is maxed out", and won't tell you to add RAM when the main thread is the bottleneck.
• Trend & memory-leak detection - Samples performance every ~10 s during the run and reports whether TPS held stable, degraded, or collapsed over time, rather than only the worst instant. Memory is judged by the post-GC live-set, so a healthy sawtooth heap (which briefly spikes near 100% before each GC) is no longer mistaken for trouble, while a steadily climbing live-set is flagged as a suspected leak.
• Prioritized, hardware-aware recommendations - Advice is now ranked by impact, de-duplicated, and quantified against your actual hardware (cores, max heap, GC type) - for example view-distance 10→8 ≈ -18 ms/tick or increase -Xmx to ~6G, rather than fixed, generic threshold messages.
• Saved Markdown report - Every run also writes a timestamped report to <world>/stressmark-reports/ with TPS/MSPT sparklines, the full per-sample time-series table, and a playbook for each recommendation (probable cause, remedy, exact flag/command) - so results survive the console scrollback and can be kept or shared.

Bug Fixes

• Fixed misleading Young-Gen GC time in reports - The "Young Gen total ms" figure was a fabricated proportional estimate (total pause × young / (young + old)) because young and old GC pause times were never tracked separately. Reports now sum the real per-collection pause time for each generation.

Improvements

• Verified analysis engine - The new report logic (trend, memory, bottleneck, capacity, grading, recommendations) lives in a self-contained, Minecraft-independent module covered by unit tests, so its verdicts are checked against synthetic scenarios rather than only observed in passing.

Bug Fixes

• Fixed bot spawning stalling at 2,000 on servers that stop ticking - The 1.3.x incremental spawn created bots across server ticks, so if the server stopped ticking right after stress start (for example an empty server paused by pause-when-empty-seconds, or a server stalled by a long GC / chunk backlog), only the first batch was ever created and the test reported e.g. 40000 requested, 2000 spawned. Bots up to 50,000 are now created synchronously inside the stress start command, so they exist immediately and no longer depend on the server continuing to tick. Counts above 50,000 still spawn the remainder across ticks (to avoid a long single-tick freeze) and now log a clear warning that the remainder needs the server to keep ticking.

Bug Fixes

• Fixed "Server handled 0 bots comfortably" on large tests - With the 1.3.0 incremental spawning, stress start no longer created bots in the command call, so a quick stress stop (or an overloaded server that could not tick) found an empty bot list and the final report claimed 0 bots were tested - even when thousands were requested. The report now uses the requested count and never reports a misleading "handled 0 bots".

Improvements

• Immediate first spawn batch - stress start now creates the first batch (up to 2,000 bots) synchronously, so a stress test always has bots right away (small tests are instant), while larger counts still spread the remainder across ticks to avoid freezing. The reported bot count is therefore never 0 for a started test.
• Accurate requested-vs-spawned reporting - When the server cannot create every requested bot in time (e.g. chunk-generation overload), the final report and stress status now show both numbers - e.g. Bots: 30000 requested, 6000 spawned - plus a clear recommendation (Only 6000 of 30000 requested bots spawned ...). The "handled comfortably" note now only appears when the full requested count actually spawned.

New Features

• Up to 1,000,000 stress test bots - Raised the stress start <count> cap from 10,000 to 1,000,000 for extreme-scale chunk-generation / disk load testing
• Speed range 0 - 10,000 blocks/sec - Widened stress start <count> [speed] from 0.5 - 20.0 to 0.0 - 10000.0. 0 keeps bots stationary (holds chunks loaded without moving); high speeds drive bots into fresh terrain fast to push chunk generation harder

Improvements

• Smooth, non-freezing bot spawning - Bots are now created incrementally in batches of 2,000 per tick instead of all at once. Large counts (e.g. 1,000,000) spread their creation over many ticks (~25s) so a single tick never freezes and the server watchdog is never tripped
• Bounded console and status output at scale - Bot creation failures are logged at most 10 times (plus a final summary) instead of one line per bot, and stress status now shows the first 20 bot positions followed by ... and N more bots. stress status also reports live spawning progress (Spawning: X / Y bots created)
• Removed O(n²) bot-list cost - Switched the internal bot list from CopyOnWriteArrayList to a plain ArrayList (all access is on the server main thread), eliminating the full-array copy on every bot added - which previously made large counts unusable

New Features

• Up to 10,000 stress test bots - Raised the stress start <count> cap from 100 to 10,000 for large-scale load testing
• Multi-version spawn support (MC 1.21 - 1.21.11) - Version-specific source sets (spawn-legacy / spawn-modern) resolve the world spawn correctly across the spawn-chunk API change in 1.21.9, selected at build time via the spawn_api property

Bug Fixes

• Fixed stress bots always spawning at world origin (0, 0) - The spawn lookup used reflection (getSharedSpawnPos by string name), which Loom does not remap; on Fabric it always threw and silently fell back to (0, 80, 0). Replaced with a direct, version-correct call - bots now spread out in a star pattern from the real spawn
• Fixed spawn resolution on MC 1.21.9+ - reads the spawn from LevelData.RespawnData since getSharedSpawnPos() was removed

New Features

• Rolling TPS/MSPT windows - Independent tick timing with 10s, 1m, and 5m rolling averages, replacing reliance on vanilla getTickTimesNanos() buffer
• Tick monitor with baseline - Auto-calibrates over 120 ticks during stress tests, reports tick ratio vs baseline for relative performance tracking
• Real-time GC monitoring - Listens to JVM garbage collection events via JMX, tracks Young/Old Gen separately with duration, frequency, and memory freed
• Colored metrics in chat - TPS, MSPT, CPU, memory, and scores are color-coded by severity thresholds (green/yellow/red)
• Unicode memory bar - Visual [████████░░░░░░░░] 67% bar in benchmark and stress test output
• GC section in reports - Both /benchmark detailed and stress test final reports now include GC activity (collection counts, pause times, longest pause)
• Sub-scores exposed - /benchmark detailed shows individual TPS, Memory, CPU, and Stability scores alongside the overall grade
• Bottleneck identification - Player capacity calculator now identifies the limiting factor (TPS, Memory, or CPU)
• Disk metrics - /benchmark detailed includes disk space and I/O stats
• Contextual recommendations - Smart suggestions based on TPS, MSPT, memory, CPU, GC, and disk metrics

Bug Fixes

• Fixed TPS always showing 20.0 during lag - Replaced custom calculateTps() with TpsMetricCollector that correctly reads vanilla tick times, and added MSPT spike detection

Improvements

• Stress test periodic updates now log to console only (reduced ops chat spam)
• Ops are only alerted for significant events: TPS drops >20% or memory usage >80%
• GC events during stress tests are logged to console only
• Stress test final report includes GC activity section and GC-based recommendations

New Features

• Rolling TPS/MSPT windows - Independent tick timing with 10s, 1m, and 5m rolling averages, replacing reliance on vanilla getTickTimesNanos() buffer
• Tick monitor with baseline - Auto-calibrates over 120 ticks during stress tests, reports tick ratio vs baseline for relative performance tracking
• Real-time GC monitoring - Listens to JVM garbage collection events via JMX, tracks Young/Old Gen separately with duration, frequency, and memory freed
• Colored metrics in chat - TPS, MSPT, CPU, memory, and scores are color-coded by severity thresholds (green/yellow/red)
• Unicode memory bar - Visual [████████░░░░░░░░] 67% bar in benchmark and stress test output
• GC section in reports - Both /benchmark detailed and stress test final reports now include GC activity (collection counts, pause times, longest pause)
• Sub-scores exposed - /benchmark detailed shows individual TPS, Memory, CPU, and Stability scores alongside the overall grade
• Bottleneck identification - Player capacity calculator now identifies the limiting factor (TPS, Memory, or CPU)
• Disk metrics - /benchmark detailed includes disk space and I/O stats
• Contextual recommendations - Smart suggestions based on TPS, MSPT, memory, CPU, GC, and disk metrics

Bug Fixes

• Fixed TPS always showing 20.0 during lag - Replaced custom calculateTps() with TpsMetricCollector that correctly reads vanilla tick times, and added MSPT spike detection

Improvements

• Stress test periodic updates now log to console only (reduced ops chat spam)
• Ops are only alerted for significant events: TPS drops >20% or memory usage >80%
• GC events during stress tests are logged to console only
• Stress test final report includes GC activity section and GC-based recommendations

New Features

• Rolling TPS/MSPT windows - Independent tick timing with 10s, 1m, and 5m rolling averages, replacing reliance on vanilla getTickTimesNanos() buffer
• Tick monitor with baseline - Auto-calibrates over 120 ticks during stress tests, reports tick ratio vs baseline for relative performance tracking
• Real-time GC monitoring - Listens to JVM garbage collection events via JMX, tracks Young/Old Gen separately with duration, frequency, and memory freed
• Colored metrics in chat - TPS, MSPT, CPU, memory, and scores are color-coded by severity thresholds (green/yellow/red)
• Unicode memory bar - Visual [████████░░░░░░░░] 67% bar in benchmark and stress test output
• GC section in reports - Both /benchmark detailed and stress test final reports now include GC activity (collection counts, pause times, longest pause)
• Sub-scores exposed - /benchmark detailed shows individual TPS, Memory, CPU, and Stability scores alongside the overall grade
• Bottleneck identification - Player capacity calculator now identifies the limiting factor (TPS, Memory, or CPU)
• Disk metrics - /benchmark detailed includes disk space and I/O stats
• Contextual recommendations - Smart suggestions based on TPS, MSPT, memory, CPU, GC, and disk metrics

Bug Fixes

• Fixed TPS always showing 20.0 during lag - Replaced custom calculateTps() with TpsMetricCollector that correctly reads vanilla tick times, and added MSPT spike detection

Improvements

• Stress test periodic updates now log to console only (reduced ops chat spam)
• Ops are only alerted for significant events: TPS drops >20% or memory usage >80%
• GC events during stress tests are logged to console only
• Stress test final report includes GC activity section and GC-based recommendations