Sightglass Benchmark Results

Results use 95.00% confidence intervals (α = 0.05). CV is Coefficient of Variation (measurement consistency).
Benchmark CV% 📊 ./main-65abe37.so …branch-a84af50.so
cm-online-stats/cm-online-stats 2.26% 882.1M cycles (p25) No significant difference
dart-deltablue/dart-deltablue 1.26% 75.6M cycles (p25) No significant difference
dotnet-raytracer/dotnet-raytracer 1.84% 45.3M cycles (p25) No significant difference
gcc-loops 0.15% 6.2M cycles (p25) 2.11% faster
hashset 1.08% 10.3M cycles (p25) 0.38% slower
hex-simd 1.01% 7.9M cycles (p25) No significant difference
hoot-earley-boyer/hoot-earley-boyer 6.35% 205.4M cycles (p25) No significant difference
intgemm-simd 1.84% 7.3M cycles (p25) 0.78% faster
java-navier-stokes/java-navier-stokes 0.11% 16.9M cycles (p25) No significant difference
kotlin-richards/kotlin-richards 1.41% 38.0M cycles (p25) 0.73% slower
libsodium/libsodium-core5 0.08% 26.0M cycles (p25) No significant difference
libsodium/libsodium-hash3 0.57% 5.7M cycles (p25) 0.82% faster
libsodium/libsodium-keygen 0.19% 22.6M cycles (p25) No significant difference
libsodium/libsodium-siphashx24 3.67% 5.8M cycles (p25) 1.14% faster
ocaml-raytrace/ocaml-raytrace 1.74% 52.0M cycles (p25) No significant difference
richards 1.47% 19.7M cycles (p25) No significant difference
rust-json 10.57% 12.0M cycles (p25) No significant difference
rust-protobuf 2.42% 25.6M cycles (p25) 0.69% faster
rust-wasmprinter/rust-wasmprinter 2.05% 22.1M cycles (p25) No significant difference
shootout/shootout-heapsort 0.74% 25.1M cycles (p25) No significant difference
shootout/shootout-keccak 9.18% 18.7M cycles (p25) No significant difference
shootout/shootout-memmove 1.55% 338.2M cycles (p25) No significant difference
shootout/shootout-sieve 4.93% 7.3M cycles (p25) No significant difference
shootout/shootout-switch 1.18% 1.0M cycles (p25) No significant difference
shootout/shootout-xchacha20 0.37% 5.9M cycles (p25) 0.19% slower
splay/splay 0.71% 4.7B cycles (p25) No significant difference
starlingmonkey/starlingmonkey-markdown 1.72% 7.1M cycles (p25) 1.08% slower
tinygo/tinygo-regex 1.01% 13.4M cycles (p25) 0.62% faster
wat-bytecode-interp/wat-bytecode-interp 0.64% 6.6M cycles (p25) No significant difference

cm-online-stats/cm-online-stats

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 882.1M cycles (p25) — — 2.26%
…branch-a84af50.so 879.7M cycles No significant difference ±5.9M cycles 2.41%

dart-deltablue/dart-deltablue

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 75.6M cycles (p25) — — 1.26%
…branch-a84af50.so 75.8M cycles No significant difference ±282.3K cycles 1.37%

dotnet-raytracer/dotnet-raytracer

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 45.3M cycles (p25) — — 1.84%
…branch-a84af50.so 45.6M cycles No significant difference ±202.1K cycles 1.25%

gcc-loops

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 6.2M cycles (p25) — — 0.15%
…branch-a84af50.so 6.1M cycles 2.11% faster ±2.6K cycles 0.15%

hashset

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 10.3M cycles (p25) — — 1.08%
…branch-a84af50.so 10.4M cycles 0.38% slower ±27.4K cycles 0.77%

hex-simd

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 7.9M cycles (p25) — — 1.01%
…branch-a84af50.so 7.9M cycles No significant difference ±20.0K cycles 0.77%

hoot-earley-boyer/hoot-earley-boyer

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 205.4M cycles (p25) — — 6.35%
…branch-a84af50.so 205.7M cycles No significant difference ±3.8M cycles 6.39%

intgemm-simd

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 7.3M cycles (p25) — — 1.84%
…branch-a84af50.so 7.3M cycles 0.78% faster ±36.6K cycles 1.66%

java-navier-stokes/java-navier-stokes

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 16.9M cycles (p25) — — 0.11%
…branch-a84af50.so 16.9M cycles No significant difference ±5.4K cycles 0.12%

kotlin-richards/kotlin-richards

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 38.0M cycles (p25) — — 1.41%
…branch-a84af50.so 38.1M cycles 0.73% slower ±193.2K cycles 2.10%

libsodium/libsodium-core5

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 26.0M cycles (p25) — — 0.08%
…branch-a84af50.so 26.0M cycles No significant difference ±5.2K cycles 0.06%

libsodium/libsodium-hash3

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 5.7M cycles (p25) — — 0.57%
…branch-a84af50.so 5.7M cycles 0.82% faster ±9.3K cycles 0.59%

libsodium/libsodium-keygen

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 22.6M cycles (p25) — — 0.19%
…branch-a84af50.so 22.6M cycles No significant difference ±10.0K cycles 0.11%

libsodium/libsodium-siphashx24

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 5.8M cycles (p25) — — 3.67%
…branch-a84af50.so 5.8M cycles 1.14% faster ±43.2K cycles 0.55%

ocaml-raytrace/ocaml-raytrace

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 52.0M cycles (p25) — — 1.74%
…branch-a84af50.so 52.2M cycles No significant difference ±533.8K cycles 4.75%

richards

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 19.7M cycles (p25) — — 1.47%
…branch-a84af50.so 19.7M cycles No significant difference ±97.9K cycles 2.00%

rust-json

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 12.0M cycles (p25) — — 10.57%
…branch-a84af50.so 12.1M cycles No significant difference ±274.6K cycles 2.77%

rust-protobuf

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 25.6M cycles (p25) — — 2.42%
…branch-a84af50.so 25.4M cycles 0.69% faster ±155.7K cycles 1.86%

rust-wasmprinter/rust-wasmprinter

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 22.1M cycles (p25) — — 2.05%
…branch-a84af50.so 22.1M cycles No significant difference ±119.9K cycles 1.76%

shootout/shootout-heapsort

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 25.1M cycles (p25) — — 0.74%
…branch-a84af50.so 25.1M cycles No significant difference ±45.8K cycles 0.55%

shootout/shootout-keccak

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 18.7M cycles (p25) — — 9.18%
…branch-a84af50.so 18.7M cycles No significant difference ±584.7K cycles 12.20%

shootout/shootout-memmove

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 338.2M cycles (p25) — — 1.55%
…branch-a84af50.so 337.9M cycles No significant difference ±1.4M cycles 1.29%

shootout/shootout-sieve

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 7.3M cycles (p25) — — 4.93%
…branch-a84af50.so 7.3M cycles No significant difference ±108.7K cycles 5.32%

shootout/shootout-switch

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 1.0M cycles (p25) — — 1.18%
…branch-a84af50.so 1.0M cycles No significant difference ±3.5K cycles 1.23%

shootout/shootout-xchacha20

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 5.9M cycles (p25) — — 0.37%
…branch-a84af50.so 5.9M cycles 0.19% slower ±7.9K cycles 0.56%

splay/splay

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 4.7B cycles (p25) — — 0.71%
…branch-a84af50.so 4.7B cycles No significant difference ±11.8M cycles 1.05%

starlingmonkey/starlingmonkey-markdown

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 7.1M cycles (p25) — — 1.72%
…branch-a84af50.so 7.1M cycles 1.08% slower ±47.5K cycles 2.84%

tinygo/tinygo-regex

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 13.4M cycles (p25) — — 1.01%
…branch-a84af50.so 13.3M cycles 0.62% faster ±70.5K cycles 2.43%

wat-bytecode-interp/wat-bytecode-interp

Engine Cycles (p25) Performance Confidence Interval CV%
📊 ./main-65abe37.so 6.6M cycles (p25) — — 0.64%
…branch-a84af50.so 6.6M cycles No significant difference ±12.1K cycles 0.65%

Statistical Methods

Uses effect size analysis with confidence intervals based on Student's t-test. Statistical significance indicates the difference is likely real (not measurement noise). Tests with CV > 5% may have unreliable measurements.

Baseline Engine (📊): The baseline engine is the reference point for all comparisons. Performance differences are calculated relative to the baseline. The baseline engine appears with a chart icon (📊) and is highlighted with a light blue background in detailed tables.

P25 Metric: Cycle counts show the 25th percentile (p25) rather than mean or median. This metric is more robust to outliers, as noise in benchmarking typically appears as spikes on the higher end (system interrupts, background processes, etc.). P25 represents a "good run" that is consistently achievable—25% of runs are this fast or faster.

Color Coding: Green indicates significantly faster performance, red indicates significantly slower, and gray indicates high measurement noise (CV > 5%). No highlighting means no statistically significant difference from baseline.

References: Benchmark Suite RFC | Sightglass Documentation