Skip to content

internal/asm/f32: add Ger asm kernels with tests

Sebastien Binet requested to merge f32/ger into master

Created by: Kunde21

Full Sger calculation done in assembly. Traversal is done in the body of the function, while the calculation is extracted into macros.

Benchmarks against old design
Old Asm New Kernel
time/op delta
BlasGer/Sger_3x3_(1_1)-12 39.2ns ± 0% 11.5ns ± 0% −70.66%
BlasGer/Sger_3x3_(1_2)-12 41.5ns ± 0% 15.5ns ± 0% −62.65%
BlasGer/Sger_3x3_(1_3)-12 40.7ns ± 0% 14.2ns ± 0% −65.11%
BlasGer/Sger_3x3_(1_4)-12 33.6ns ± 0% 15.1ns ± 0% −55.06%
BlasGer/Sger_3x3_(1_10)-12 37.5ns ± 0% 15.2ns ± 0% −59.47%
BlasGer/Sger_3x3_(2_1)-12 39.1ns ± 0% 15.5ns ± 0% −60.36%
BlasGer/Sger_3x3_(2_2)-12 41.0ns ± 0% 14.0ns ± 0% −65.85%
BlasGer/Sger_3x3_(2_3)-12 41.1ns ± 0% 15.5ns ± 0% −62.29%
BlasGer/Sger_3x3_(2_4)-12 40.7ns ± 0% 15.2ns ± 0% −62.65%
BlasGer/Sger_3x3_(2_10)-12 38.3ns ± 0% 13.4ns ± 0% −65.01%
BlasGer/Sger_3x3_(3_1)-12 33.8ns ± 0% 16.1ns ± 0% −52.37%
BlasGer/Sger_3x3_(3_2)-12 36.3ns ± 0% 15.5ns ± 0% −57.30%
BlasGer/Sger_3x3_(3_3)-12 41.1ns ± 0% 15.1ns ± 0% −63.26%
BlasGer/Sger_3x3_(3_4)-12 43.3ns ± 0% 14.1ns ± 0% −67.44%
BlasGer/Sger_3x3_(3_10)-12 41.3ns ± 0% 15.6ns ± 0% −62.23%
BlasGer/Sger_3x3_(4_1)-12 40.2ns ± 0% 15.6ns ± 0% −61.19%
BlasGer/Sger_3x3_(4_2)-12 37.3ns ± 0% 15.6ns ± 0% −58.18%
BlasGer/Sger_3x3_(4_3)-12 39.1ns ± 0% 15.6ns ± 0% −60.10%
BlasGer/Sger_3x3_(4_4)-12 37.7ns ± 0% 13.3ns ± 0% −64.72%
BlasGer/Sger_3x3_(4_10)-12 33.6ns ± 0% 15.3ns ± 0% −54.46%
BlasGer/Sger_3x3_(10_1)-12 36.1ns ± 0% 14.0ns ± 0% −61.22%
BlasGer/Sger_3x3_(10_2)-12 37.8ns ± 0% 13.2ns ± 0% −65.08%
BlasGer/Sger_3x3_(10_3)-12 38.3ns ± 0% 15.2ns ± 0% −60.31%
BlasGer/Sger_3x3_(10_4)-12 38.6ns ± 0% 14.2ns ± 0% −63.21%
BlasGer/Sger_3x3_(10_10)-12 37.0ns ± 0% 13.7ns ± 0% −62.97%
BlasGer/Sger_3x10_(1_1)-12 51.4ns ± 0% 15.2ns ± 0% −70.43%
BlasGer/Sger_3x10_(1_2)-12 57.1ns ± 0% 18.3ns ± 0% −67.95%
BlasGer/Sger_3x10_(1_3)-12 51.7ns ± 0% 18.1ns ± 0% −64.99%
BlasGer/Sger_3x10_(1_4)-12 57.1ns ± 0% 21.6ns ± 0% −62.17%
BlasGer/Sger_3x10_(1_10)-12 50.9ns ± 0% 15.6ns ± 0% −69.35%
BlasGer/Sger_3x10_(2_1)-12 55.2ns ± 0% 16.1ns ± 0% −70.83%
BlasGer/Sger_3x10_(2_2)-12 55.2ns ± 0% 16.7ns ± 0% −69.75%
BlasGer/Sger_3x10_(2_3)-12 51.0ns ± 0% 18.5ns ± 0% −63.73%
BlasGer/Sger_3x10_(2_4)-12 56.6ns ± 0% 17.3ns ± 0% −69.43%
BlasGer/Sger_3x10_(2_10)-12 50.8ns ± 0% 15.5ns ± 0% −69.49%
BlasGer/Sger_3x10_(3_1)-12 52.9ns ± 0% 20.2ns ± 0% −61.81%
BlasGer/Sger_3x10_(3_2)-12 45.5ns ± 0% 16.6ns ± 0% −63.52%
BlasGer/Sger_3x10_(3_3)-12 57.3ns ± 0% 18.1ns ± 0% −68.41%
BlasGer/Sger_3x10_(3_4)-12 50.3ns ± 0% 17.4ns ± 0% −65.41%
BlasGer/Sger_3x10_(3_10)-12 57.6ns ± 0% 16.0ns ± 0% −72.22%
BlasGer/Sger_3x10_(4_1)-12 55.2ns ± 0% 17.9ns ± 0% −67.57%
BlasGer/Sger_3x10_(4_2)-12 50.7ns ± 0% 17.9ns ± 0% −64.69%
BlasGer/Sger_3x10_(4_3)-12 44.4ns ± 0% 17.7ns ± 0% −60.14%
BlasGer/Sger_3x10_(4_4)-12 57.0ns ± 0% 17.2ns ± 0% −69.82%
BlasGer/Sger_3x10_(4_10)-12 60.3ns ± 0% 18.0ns ± 0% −70.15%
BlasGer/Sger_3x10_(10_1)-12 56.4ns ± 0% 16.5ns ± 0% −70.74%
BlasGer/Sger_3x10_(10_2)-12 52.5ns ± 0% 16.9ns ± 0% −67.81%
BlasGer/Sger_3x10_(10_3)-12 53.3ns ± 0% 18.0ns ± 0% −66.23%
BlasGer/Sger_3x10_(10_4)-12 50.5ns ± 0% 17.4ns ± 0% −65.54%
BlasGer/Sger_3x10_(10_10)-12 57.6ns ± 0% 15.5ns ± 0% −73.09%
BlasGer/Sger_3x30_(1_1)-12 60.6ns ± 0% 21.8ns ± 0% −64.03%
BlasGer/Sger_3x30_(1_2)-12 112ns ± 0% 27ns ± 0% −76.25%
BlasGer/Sger_3x30_(1_3)-12 110ns ± 0% 31ns ± 0% −71.45%
BlasGer/Sger_3x30_(1_4)-12 112ns ± 0% 33ns ± 0% −70.36%
BlasGer/Sger_3x30_(1_10)-12 96.4ns ± 0% 31.9ns ± 0% −66.91%
BlasGer/Sger_3x30_(2_1)-12 110ns ± 0% 25ns ± 0% −77.18%
BlasGer/Sger_3x30_(2_2)-12 103ns ± 0% 28ns ± 0% −72.33%
BlasGer/Sger_3x30_(2_3)-12 112ns ± 0% 32ns ± 0% −71.79%
BlasGer/Sger_3x30_(2_4)-12 86.0ns ± 0% 33.9ns ± 0% −60.58%
BlasGer/Sger_3x30_(2_10)-12 86.1ns ± 0% 30.9ns ± 0% −64.11%
BlasGer/Sger_3x30_(3_1)-12 96.4ns ± 0% 25.7ns ± 0% −73.34%
BlasGer/Sger_3x30_(3_2)-12 93.9ns ± 0% 26.1ns ± 0% −72.20%
BlasGer/Sger_3x30_(3_3)-12 98.4ns ± 0% 27.7ns ± 0% −71.85%
BlasGer/Sger_3x30_(3_4)-12 112ns ± 0% 31ns ± 0% −71.88%
BlasGer/Sger_3x30_(3_10)-12 98.9ns ± 0% 34.0ns ± 0% −65.62%
BlasGer/Sger_3x30_(4_1)-12 98.4ns ± 0% 30.6ns ± 0% −68.90%
BlasGer/Sger_3x30_(4_2)-12 100ns ± 0% 31ns ± 0% −69.30%
BlasGer/Sger_3x30_(4_3)-12 112ns ± 0% 30ns ± 0% −73.30%
BlasGer/Sger_3x30_(4_4)-12 96.5ns ± 0% 33.0ns ± 0% −65.80%
BlasGer/Sger_3x30_(4_10)-12 97.3ns ± 0% 30.2ns ± 0% −68.96%
BlasGer/Sger_3x30_(10_1)-12 104ns ± 0% 29ns ± 0% −71.73%
BlasGer/Sger_3x30_(10_2)-12 88.4ns ± 0% 29.5ns ± 0% −66.63%
BlasGer/Sger_3x30_(10_3)-12 98.0ns ± 0% 34.3ns ± 0% −65.00%
BlasGer/Sger_3x30_(10_4)-12 105ns ± 0% 33ns ± 0% −68.19%
BlasGer/Sger_3x30_(10_10)-12 97.2ns ± 0% 29.3ns ± 0% −69.86%
BlasGer/Sger_3x100_(1_1)-12 75.6ns ± 0% 52.6ns ± 0% −30.42%
BlasGer/Sger_3x100_(1_2)-12 270ns ± 0% 63ns ± 0% −76.81%
BlasGer/Sger_3x100_(1_3)-12 256ns ± 0% 60ns ± 0% −76.68%
BlasGer/Sger_3x100_(1_4)-12 238ns ± 0% 65ns ± 0% −72.61%
BlasGer/Sger_3x100_(1_10)-12 292ns ± 0% 68ns ± 0% −76.88%
BlasGer/Sger_3x100_(2_1)-12 253ns ± 0% 64ns ± 0% −74.70%
BlasGer/Sger_3x100_(2_2)-12 271ns ± 0% 68ns ± 0% −74.72%
BlasGer/Sger_3x100_(2_3)-12 237ns ± 0% 66ns ± 0% −71.94%
BlasGer/Sger_3x100_(2_4)-12 297ns ± 0% 68ns ± 0% −77.07%
BlasGer/Sger_3x100_(2_10)-12 258ns ± 0% 72ns ± 0% −72.17%
BlasGer/Sger_3x100_(3_1)-12 245ns ± 0% 69ns ± 0% −71.96%
BlasGer/Sger_3x100_(3_2)-12 258ns ± 0% 67ns ± 0% −73.88%
BlasGer/Sger_3x100_(3_3)-12 249ns ± 0% 69ns ± 0% −72.17%
BlasGer/Sger_3x100_(3_4)-12 276ns ± 0% 66ns ± 0% −75.91%
BlasGer/Sger_3x100_(3_10)-12 233ns ± 0% 73ns ± 0% −68.50%
BlasGer/Sger_3x100_(4_1)-12 271ns ± 0% 64ns ± 0% −76.20%
BlasGer/Sger_3x100_(4_2)-12 263ns ± 0% 70ns ± 0% −73.35%
BlasGer/Sger_3x100_(4_3)-12 276ns ± 0% 72ns ± 0% −73.95%
BlasGer/Sger_3x100_(4_4)-12 242ns ± 0% 76ns ± 0% −68.64%
BlasGer/Sger_3x100_(4_10)-12 234ns ± 0% 75ns ± 0% −67.95%
BlasGer/Sger_3x100_(10_1)-12 272ns ± 0% 75ns ± 0% −72.57%
BlasGer/Sger_3x100_(10_2)-12 233ns ± 0% 73ns ± 0% −68.71%
BlasGer/Sger_3x100_(10_3)-12 268ns ± 0% 73ns ± 0% −72.91%
BlasGer/Sger_3x100_(10_4)-12 279ns ± 0% 73ns ± 0% −73.91%
BlasGer/Sger_3x100_(10_10)-12 242ns ± 0% 70ns ± 0% −71.07%
BlasGer/Sger_10x3_(1_1)-12 128ns ± 0% 20ns ± 0% −84.06%
BlasGer/Sger_10x3_(1_2)-12 112ns ± 0% 23ns ± 0% −79.82%
BlasGer/Sger_10x3_(1_3)-12 97.2ns ± 0% 20.9ns ± 0% −78.50%
BlasGer/Sger_10x3_(1_4)-12 120ns ± 0% 31ns ± 0% −74.50%
BlasGer/Sger_10x3_(1_10)-12 105ns ± 0% 26ns ± 0% −75.24%
BlasGer/Sger_10x3_(2_1)-12 105ns ± 0% 22ns ± 0% −79.43%
BlasGer/Sger_10x3_(2_2)-12 101ns ± 0% 21ns ± 0% −79.50%
BlasGer/Sger_10x3_(2_3)-12 126ns ± 0% 21ns ± 0% −83.65%
BlasGer/Sger_10x3_(2_4)-12 111ns ± 0% 23ns ± 0% −79.46%
BlasGer/Sger_10x3_(2_10)-12 104ns ± 0% 26ns ± 0% −74.81%
BlasGer/Sger_10x3_(3_1)-12 131ns ± 0% 26ns ± 0% −80.08%
BlasGer/Sger_10x3_(3_2)-12 115ns ± 0% 26ns ± 0% −77.65%
BlasGer/Sger_10x3_(3_3)-12 119ns ± 0% 25ns ± 0% −78.66%
BlasGer/Sger_10x3_(3_4)-12 123ns ± 0% 26ns ± 0% −79.02%
BlasGer/Sger_10x3_(3_10)-12 112ns ± 0% 24ns ± 0% −78.66%
BlasGer/Sger_10x3_(4_1)-12 123ns ± 0% 25ns ± 0% −79.51%
BlasGer/Sger_10x3_(4_2)-12 110ns ± 0% 26ns ± 0% −76.36%
BlasGer/Sger_10x3_(4_3)-12 107ns ± 0% 26ns ± 0% −75.89%
BlasGer/Sger_10x3_(4_4)-12 109ns ± 0% 21ns ± 0% −80.46%
BlasGer/Sger_10x3_(4_10)-12 123ns ± 0% 23ns ± 0% −81.30%
BlasGer/Sger_10x3_(10_1)-12 124ns ± 0% 23ns ± 0% −81.13%
BlasGer/Sger_10x3_(10_2)-12 113ns ± 0% 23ns ± 0% −79.65%
BlasGer/Sger_10x3_(10_3)-12 122ns ± 0% 26ns ± 0% −79.02%
BlasGer/Sger_10x3_(10_4)-12 107ns ± 0% 26ns ± 0% −75.51%
BlasGer/Sger_10x3_(10_10)-12 124ns ± 0% 26ns ± 0% −79.35%
BlasGer/Sger_10x10_(1_1)-12 145ns ± 0% 25ns ± 0% −82.62%
BlasGer/Sger_10x10_(1_2)-12 140ns ± 0% 34ns ± 0% −75.50%
BlasGer/Sger_10x10_(1_3)-12 162ns ± 0% 33ns ± 0% −79.63%
BlasGer/Sger_10x10_(1_4)-12 137ns ± 0% 33ns ± 0% −75.77%
BlasGer/Sger_10x10_(1_10)-12 161ns ± 0% 31ns ± 0% −80.56%
BlasGer/Sger_10x10_(2_1)-12 138ns ± 0% 38ns ± 0% −72.61%
BlasGer/Sger_10x10_(2_2)-12 163ns ± 0% 30ns ± 0% −81.90%
BlasGer/Sger_10x10_(2_3)-12 182ns ± 0% 34ns ± 0% −81.59%
BlasGer/Sger_10x10_(2_4)-12 178ns ± 0% 33ns ± 0% −81.24%
BlasGer/Sger_10x10_(2_10)-12 163ns ± 0% 30ns ± 0% −81.78%
BlasGer/Sger_10x10_(3_1)-12 149ns ± 0% 34ns ± 0% −77.11%
BlasGer/Sger_10x10_(3_2)-12 153ns ± 0% 28ns ± 0% −81.44%
BlasGer/Sger_10x10_(3_3)-12 162ns ± 0% 31ns ± 0% −81.17%
BlasGer/Sger_10x10_(3_4)-12 163ns ± 0% 29ns ± 0% −82.39%
BlasGer/Sger_10x10_(3_10)-12 161ns ± 0% 35ns ± 0% −78.01%
BlasGer/Sger_10x10_(4_1)-12 162ns ± 0% 30ns ± 0% −81.79%
BlasGer/Sger_10x10_(4_2)-12 142ns ± 0% 32ns ± 0% −77.75%
BlasGer/Sger_10x10_(4_3)-12 166ns ± 0% 32ns ± 0% −80.54%
BlasGer/Sger_10x10_(4_4)-12 164ns ± 0% 34ns ± 0% −79.45%
BlasGer/Sger_10x10_(4_10)-12 162ns ± 0% 30ns ± 0% −81.30%
BlasGer/Sger_10x10_(10_1)-12 147ns ± 0% 30ns ± 0% −79.93%
BlasGer/Sger_10x10_(10_2)-12 141ns ± 0% 34ns ± 0% −76.10%
BlasGer/Sger_10x10_(10_3)-12 157ns ± 0% 28ns ± 0% −81.97%
BlasGer/Sger_10x10_(10_4)-12 142ns ± 0% 31ns ± 0% −78.03%
BlasGer/Sger_10x10_(10_10)-12 155ns ± 0% 33ns ± 0% −78.45%
BlasGer/Sger_10x30_(1_1)-12 200ns ± 0% 50ns ± 0% −74.75%
BlasGer/Sger_10x30_(1_2)-12 329ns ± 0% 54ns ± 0% −83.68%
BlasGer/Sger_10x30_(1_3)-12 325ns ± 0% 53ns ± 0% −83.72%
BlasGer/Sger_10x30_(1_4)-12 347ns ± 0% 65ns ± 0% −81.15%
BlasGer/Sger_10x30_(1_10)-12 321ns ± 0% 67ns ± 0% −79.13%
BlasGer/Sger_10x30_(2_1)-12 322ns ± 0% 63ns ± 0% −80.37%
BlasGer/Sger_10x30_(2_2)-12 282ns ± 0% 52ns ± 0% −81.49%
BlasGer/Sger_10x30_(2_3)-12 340ns ± 0% 66ns ± 0% −80.74%
BlasGer/Sger_10x30_(2_4)-12 315ns ± 0% 61ns ± 0% −80.73%
BlasGer/Sger_10x30_(2_10)-12 325ns ± 0% 54ns ± 0% −83.42%
BlasGer/Sger_10x30_(3_1)-12 364ns ± 0% 60ns ± 0% −83.46%
BlasGer/Sger_10x30_(3_2)-12 323ns ± 0% 64ns ± 0% −80.12%
BlasGer/Sger_10x30_(3_3)-12 313ns ± 0% 60ns ± 0% −80.93%
BlasGer/Sger_10x30_(3_4)-12 342ns ± 0% 66ns ± 0% −80.58%
BlasGer/Sger_10x30_(3_10)-12 362ns ± 0% 67ns ± 0% −81.38%
BlasGer/Sger_10x30_(4_1)-12 335ns ± 0% 61ns ± 0% −81.88%
BlasGer/Sger_10x30_(4_2)-12 293ns ± 0% 57ns ± 0% −80.68%
BlasGer/Sger_10x30_(4_3)-12 318ns ± 0% 66ns ± 0% −79.12%
BlasGer/Sger_10x30_(4_4)-12 359ns ± 0% 67ns ± 0% −81.34%
BlasGer/Sger_10x30_(4_10)-12 285ns ± 0% 68ns ± 0% −76.14%
BlasGer/Sger_10x30_(10_1)-12 295ns ± 0% 67ns ± 0% −77.42%
BlasGer/Sger_10x30_(10_2)-12 305ns ± 0% 54ns ± 0% −82.33%
BlasGer/Sger_10x30_(10_3)-12 274ns ± 0% 58ns ± 0% −79.01%
BlasGer/Sger_10x30_(10_4)-12 346ns ± 0% 60ns ± 0% −82.57%
BlasGer/Sger_10x30_(10_10)-12 275ns ± 0% 61ns ± 0% −77.82%
BlasGer/Sger_10x100_(1_1)-12 280ns ± 0% 127ns ± 0% −54.64%
BlasGer/Sger_10x100_(1_2)-12 873ns ± 0% 166ns ± 0% −80.99%
BlasGer/Sger_10x100_(1_3)-12 749ns ± 0% 177ns ± 0% −76.37%
BlasGer/Sger_10x100_(1_4)-12 863ns ± 0% 168ns ± 0% −80.53%
BlasGer/Sger_10x100_(1_10)-12 747ns ± 0% 144ns ± 0% −80.72%
BlasGer/Sger_10x100_(2_1)-12 827ns ± 0% 168ns ± 0% −79.69%
BlasGer/Sger_10x100_(2_2)-12 1.02µs ± 0% 0.16µs ± 0% −83.91%
BlasGer/Sger_10x100_(2_3)-12 867ns ± 0% 163ns ± 0% −81.20%
BlasGer/Sger_10x100_(2_4)-12 1.04µs ± 0% 0.17µs ± 0% −83.70%
BlasGer/Sger_10x100_(2_10)-12 1.00µs ± 0% 0.16µs ± 0% −84.40%
BlasGer/Sger_10x100_(3_1)-12 882ns ± 0% 160ns ± 0% −81.86%
BlasGer/Sger_10x100_(3_2)-12 811ns ± 0% 167ns ± 0% −79.41%
BlasGer/Sger_10x100_(3_3)-12 1.04µs ± 0% 0.17µs ± 0% −83.96%
BlasGer/Sger_10x100_(3_4)-12 831ns ± 0% 166ns ± 0% −80.02%
BlasGer/Sger_10x100_(3_10)-12 1.06µs ± 0% 0.15µs ± 0% −85.51%
BlasGer/Sger_10x100_(4_1)-12 851ns ± 0% 148ns ± 0% −82.61%
BlasGer/Sger_10x100_(4_2)-12 994ns ± 0% 164ns ± 0% −83.50%
BlasGer/Sger_10x100_(4_3)-12 892ns ± 0% 160ns ± 0% −82.06%
BlasGer/Sger_10x100_(4_4)-12 761ns ± 0% 146ns ± 0% −80.81%
BlasGer/Sger_10x100_(4_10)-12 874ns ± 0% 144ns ± 0% −83.52%
BlasGer/Sger_10x100_(10_1)-12 842ns ± 0% 168ns ± 0% −80.05%
BlasGer/Sger_10x100_(10_2)-12 871ns ± 0% 149ns ± 0% −82.89%
BlasGer/Sger_10x100_(10_3)-12 862ns ± 0% 149ns ± 0% −82.71%
BlasGer/Sger_10x100_(10_4)-12 900ns ± 0% 166ns ± 0% −81.56%
BlasGer/Sger_10x100_(10_10)-12 879ns ± 0% 172ns ± 0% −80.43%
BlasGer/Sger_10x300_(1_1)-12 635ns ± 0% 392ns ± 0% −38.27%
BlasGer/Sger_10x300_(1_2)-12 2.70µs ± 0% 0.43µs ± 0% −83.91%
BlasGer/Sger_10x300_(1_3)-12 2.52µs ± 0% 0.45µs ± 0% −82.04%
BlasGer/Sger_10x300_(1_4)-12 2.40µs ± 0% 0.55µs ± 0% −77.19%
BlasGer/Sger_10x300_(1_10)-12 2.28µs ± 0% 0.43µs ± 0% −81.22%
BlasGer/Sger_10x300_(2_1)-12 2.96µs ± 0% 0.49µs ± 0% −83.42%
BlasGer/Sger_10x300_(2_2)-12 2.39µs ± 0% 0.49µs ± 0% −79.67%
BlasGer/Sger_10x300_(2_3)-12 2.70µs ± 0% 0.43µs ± 0% −84.17%
BlasGer/Sger_10x300_(2_4)-12 2.40µs ± 0% 0.44µs ± 0% −81.73%
BlasGer/Sger_10x300_(2_10)-12 2.34µs ± 0% 0.50µs ± 0% −78.52%
BlasGer/Sger_10x300_(3_1)-12 2.46µs ± 0% 0.45µs ± 0% −81.79%
BlasGer/Sger_10x300_(3_2)-12 2.32µs ± 0% 0.47µs ± 0% −79.62%
BlasGer/Sger_10x300_(3_3)-12 2.71µs ± 0% 0.44µs ± 0% −83.78%
BlasGer/Sger_10x300_(3_4)-12 2.32µs ± 0% 0.49µs ± 0% −78.72%
BlasGer/Sger_10x300_(3_10)-12 2.24µs ± 0% 0.41µs ± 0% −81.48%
BlasGer/Sger_10x300_(4_1)-12 3.03µs ± 0% 0.42µs ± 0% −86.23%
BlasGer/Sger_10x300_(4_2)-12 2.82µs ± 0% 0.42µs ± 0% −85.13%
BlasGer/Sger_10x300_(4_3)-12 2.17µs ± 0% 0.42µs ± 0% −80.55%
BlasGer/Sger_10x300_(4_4)-12 3.01µs ± 0% 0.43µs ± 0% −85.61%
BlasGer/Sger_10x300_(4_10)-12 2.47µs ± 0% 0.53µs ± 0% −78.66%
BlasGer/Sger_10x300_(10_1)-12 2.37µs ± 0% 0.41µs ± 0% −82.76%
BlasGer/Sger_10x300_(10_2)-12 2.70µs ± 0% 0.49µs ± 0% −81.81%
BlasGer/Sger_10x300_(10_3)-12 2.70µs ± 0% 0.50µs ± 0% −81.38%
BlasGer/Sger_10x300_(10_4)-12 2.44µs ± 0% 0.44µs ± 0% −81.75%
BlasGer/Sger_10x300_(10_10)-12 2.24µs ± 0% 0.48µs ± 0% −78.46%
BlasGer/Sger_30x3_(1_1)-12 376ns ± 0% 54ns ± 0% −85.69%
BlasGer/Sger_30x3_(1_2)-12 376ns ± 0% 54ns ± 0% −85.74%
BlasGer/Sger_30x3_(1_3)-12 357ns ± 0% 53ns ± 0% −85.27%
BlasGer/Sger_30x3_(1_4)-12 378ns ± 0% 45ns ± 0% −88.15%
BlasGer/Sger_30x3_(1_10)-12 344ns ± 0% 52ns ± 0% −84.80%
BlasGer/Sger_30x3_(2_1)-12 359ns ± 0% 53ns ± 0% −85.21%
BlasGer/Sger_30x3_(2_2)-12 383ns ± 0% 52ns ± 0% −86.45%
BlasGer/Sger_30x3_(2_3)-12 379ns ± 0% 51ns ± 0% −86.65%
BlasGer/Sger_30x3_(2_4)-12 386ns ± 0% 44ns ± 0% −88.52%
BlasGer/Sger_30x3_(2_10)-12 346ns ± 0% 53ns ± 0% −84.60%
BlasGer/Sger_30x3_(3_1)-12 345ns ± 0% 50ns ± 0% −85.51%
BlasGer/Sger_30x3_(3_2)-12 322ns ± 0% 49ns ± 0% −84.91%
BlasGer/Sger_30x3_(3_3)-12 347ns ± 0% 60ns ± 0% −82.71%
BlasGer/Sger_30x3_(3_4)-12 376ns ± 0% 52ns ± 0% −86.28%
BlasGer/Sger_30x3_(3_10)-12 346ns ± 0% 52ns ± 0% −85.06%
BlasGer/Sger_30x3_(4_1)-12 344ns ± 0% 55ns ± 0% −83.90%
BlasGer/Sger_30x3_(4_2)-12 336ns ± 0% 55ns ± 0% −83.57%
BlasGer/Sger_30x3_(4_3)-12 380ns ± 0% 50ns ± 0% −86.97%
BlasGer/Sger_30x3_(4_4)-12 348ns ± 0% 63ns ± 0% −81.81%
BlasGer/Sger_30x3_(4_10)-12 337ns ± 0% 48ns ± 0% −85.64%
BlasGer/Sger_30x3_(10_1)-12 343ns ± 0% 56ns ± 0% −83.73%
BlasGer/Sger_30x3_(10_2)-12 345ns ± 0% 54ns ± 0% −84.46%
BlasGer/Sger_30x3_(10_3)-12 369ns ± 0% 53ns ± 0% −85.58%
BlasGer/Sger_30x3_(10_4)-12 379ns ± 0% 56ns ± 0% −85.36%
BlasGer/Sger_30x3_(10_10)-12 373ns ± 0% 50ns ± 0% −86.70%
BlasGer/Sger_30x10_(1_1)-12 441ns ± 0% 61ns ± 0% −86.26%
BlasGer/Sger_30x10_(1_2)-12 499ns ± 0% 77ns ± 0% −84.49%
BlasGer/Sger_30x10_(1_3)-12 500ns ± 0% 75ns ± 0% −85.00%
BlasGer/Sger_30x10_(1_4)-12 497ns ± 0% 63ns ± 0% −87.36%
BlasGer/Sger_30x10_(1_10)-12 499ns ± 0% 75ns ± 0% −84.89%
BlasGer/Sger_30x10_(2_1)-12 498ns ± 0% 76ns ± 0% −84.78%
BlasGer/Sger_30x10_(2_2)-12 498ns ± 0% 64ns ± 0% −87.07%
BlasGer/Sger_30x10_(2_3)-12 507ns ± 0% 76ns ± 0% −84.93%
BlasGer/Sger_30x10_(2_4)-12 489ns ± 0% 75ns ± 0% −84.74%
BlasGer/Sger_30x10_(2_10)-12 510ns ± 0% 67ns ± 0% −86.86%
BlasGer/Sger_30x10_(3_1)-12 509ns ± 0% 83ns ± 0% −83.65%
BlasGer/Sger_30x10_(3_2)-12 497ns ± 0% 73ns ± 0% −85.29%
BlasGer/Sger_30x10_(3_3)-12 497ns ± 0% 77ns ± 0% −84.47%
BlasGer/Sger_30x10_(3_4)-12 496ns ± 0% 71ns ± 0% −85.71%
BlasGer/Sger_30x10_(3_10)-12 426ns ± 0% 75ns ± 0% −82.39%
BlasGer/Sger_30x10_(4_1)-12 487ns ± 0% 64ns ± 0% −86.76%
BlasGer/Sger_30x10_(4_2)-12 481ns ± 0% 64ns ± 0% −86.63%
BlasGer/Sger_30x10_(4_3)-12 450ns ± 0% 76ns ± 0% −83.09%
BlasGer/Sger_30x10_(4_4)-12 486ns ± 0% 71ns ± 0% −85.33%
BlasGer/Sger_30x10_(4_10)-12 429ns ± 0% 70ns ± 0% −83.68%
BlasGer/Sger_30x10_(10_1)-12 487ns ± 0% 74ns ± 0% −84.83%
BlasGer/Sger_30x10_(10_2)-12 417ns ± 0% 75ns ± 0% −81.99%
BlasGer/Sger_30x10_(10_3)-12 488ns ± 0% 70ns ± 0% −85.74%
BlasGer/Sger_30x10_(10_4)-12 442ns ± 0% 66ns ± 0% −85.02%
BlasGer/Sger_30x10_(10_10)-12 497ns ± 0% 64ns ± 0% −87.14%
BlasGer/Sger_30x30_(1_1)-12 610ns ± 0% 111ns ± 0% −81.80%
BlasGer/Sger_30x30_(1_2)-12 953ns ± 0% 147ns ± 0% −84.58%
BlasGer/Sger_30x30_(1_3)-12 927ns ± 0% 163ns ± 0% −82.42%
BlasGer/Sger_30x30_(1_4)-12 891ns ± 0% 165ns ± 0% −81.48%
BlasGer/Sger_30x30_(1_10)-12 1.05µs ± 0% 0.15µs ± 0% −85.32%
BlasGer/Sger_30x30_(2_1)-12 948ns ± 0% 153ns ± 0% −83.86%
BlasGer/Sger_30x30_(2_2)-12 958ns ± 0% 181ns ± 0% −81.11%
BlasGer/Sger_30x30_(2_3)-12 944ns ± 0% 162ns ± 0% −82.84%
BlasGer/Sger_30x30_(2_4)-12 955ns ± 0% 138ns ± 0% −85.55%
BlasGer/Sger_30x30_(2_10)-12 1.15µs ± 0% 0.17µs ± 0% −85.49%
BlasGer/Sger_30x30_(3_1)-12 1.00µs ± 0% 0.16µs ± 0% −83.95%
BlasGer/Sger_30x30_(3_2)-12 1.13µs ± 0% 0.16µs ± 0% −86.05%
BlasGer/Sger_30x30_(3_3)-12 964ns ± 0% 169ns ± 0% −82.47%
BlasGer/Sger_30x30_(3_4)-12 862ns ± 0% 157ns ± 0% −81.79%
BlasGer/Sger_30x30_(3_10)-12 822ns ± 0% 161ns ± 0% −80.41%
BlasGer/Sger_30x30_(4_1)-12 1.04µs ± 0% 0.16µs ± 0% −84.78%
BlasGer/Sger_30x30_(4_2)-12 1.08µs ± 0% 0.15µs ± 0% −86.52%
BlasGer/Sger_30x30_(4_3)-12 1.08µs ± 0% 0.16µs ± 0% −85.41%
BlasGer/Sger_30x30_(4_4)-12 912ns ± 0% 165ns ± 0% −81.91%
BlasGer/Sger_30x30_(4_10)-12 1.08µs ± 0% 0.17µs ± 0% −84.66%
BlasGer/Sger_30x30_(10_1)-12 1.10µs ± 0% 0.16µs ± 0% −85.88%
BlasGer/Sger_30x30_(10_2)-12 828ns ± 0% 153ns ± 0% −81.52%
BlasGer/Sger_30x30_(10_3)-12 1.09µs ± 0% 0.15µs ± 0% −86.28%
BlasGer/Sger_30x30_(10_4)-12 957ns ± 0% 165ns ± 0% −82.76%
BlasGer/Sger_30x30_(10_10)-12 950ns ± 0% 164ns ± 0% −82.74%
BlasGer/Sger_30x100_(1_1)-12 886ns ± 0% 387ns ± 0% −56.32%
BlasGer/Sger_30x100_(1_2)-12 2.72µs ± 0% 0.43µs ± 0% −84.06%
BlasGer/Sger_30x100_(1_3)-12 2.49µs ± 0% 0.43µs ± 0% −82.79%
BlasGer/Sger_30x100_(1_4)-12 2.71µs ± 0% 0.46µs ± 0% −83.07%
BlasGer/Sger_30x100_(1_10)-12 2.73µs ± 0% 0.41µs ± 0% −84.92%
BlasGer/Sger_30x100_(2_1)-12 2.69µs ± 0% 0.42µs ± 0% −84.22%
BlasGer/Sger_30x100_(2_2)-12 2.48µs ± 0% 0.43µs ± 0% −82.47%
BlasGer/Sger_30x100_(2_3)-12 2.70µs ± 0% 0.49µs ± 0% −81.84%
BlasGer/Sger_30x100_(2_4)-12 2.48µs ± 0% 0.48µs ± 0% −80.48%
BlasGer/Sger_30x100_(2_10)-12 2.72µs ± 0% 0.43µs ± 0% −84.02%
BlasGer/Sger_30x100_(3_1)-12 2.29µs ± 0% 0.49µs ± 0% −78.61%
BlasGer/Sger_30x100_(3_2)-12 2.50µs ± 0% 0.51µs ± 0% −79.44%
BlasGer/Sger_30x100_(3_3)-12 2.65µs ± 0% 0.48µs ± 0% −81.70%
BlasGer/Sger_30x100_(3_4)-12 2.49µs ± 0% 0.41µs ± 0% −83.51%
BlasGer/Sger_30x100_(3_10)-12 2.53µs ± 0% 0.49µs ± 0% −80.60%
BlasGer/Sger_30x100_(4_1)-12 2.27µs ± 0% 0.49µs ± 0% −78.46%
BlasGer/Sger_30x100_(4_2)-12 2.42µs ± 0% 0.49µs ± 0% −79.90%
BlasGer/Sger_30x100_(4_3)-12 2.69µs ± 0% 0.49µs ± 0% −81.87%
BlasGer/Sger_30x100_(4_4)-12 2.69µs ± 0% 0.45µs ± 0% −83.14%
BlasGer/Sger_30x100_(4_10)-12 2.22µs ± 0% 0.49µs ± 0% −77.93%
BlasGer/Sger_30x100_(10_1)-12 2.85µs ± 0% 0.49µs ± 0% −82.82%
BlasGer/Sger_30x100_(10_2)-12 2.62µs ± 0% 0.49µs ± 0% −81.40%
BlasGer/Sger_30x100_(10_3)-12 2.59µs ± 0% 0.47µs ± 0% −81.84%
BlasGer/Sger_30x100_(10_4)-12 2.86µs ± 0% 0.48µs ± 0% −83.35%
BlasGer/Sger_30x100_(10_10)-12 2.77µs ± 0% 0.43µs ± 0% −84.42%
BlasGer/Sger_30x300_(1_1)-12 2.28µs ± 0% 1.32µs ± 0% −41.90%
BlasGer/Sger_30x300_(1_2)-12 6.91µs ± 0% 1.40µs ± 0% −79.72%
BlasGer/Sger_30x300_(1_3)-12 7.74µs ± 0% 1.19µs ± 0% −84.58%
BlasGer/Sger_30x300_(1_4)-12 7.44µs ± 0% 1.44µs ± 0% −80.58%
BlasGer/Sger_30x300_(1_10)-12 7.71µs ± 0% 1.44µs ± 0% −81.31%
BlasGer/Sger_30x300_(2_1)-12 7.69µs ± 0% 1.41µs ± 0% −81.68%
BlasGer/Sger_30x300_(2_2)-12 7.76µs ± 0% 1.43µs ± 0% −81.55%
BlasGer/Sger_30x300_(2_3)-12 7.65µs ± 0% 1.19µs ± 0% −84.49%
BlasGer/Sger_30x300_(2_4)-12 7.81µs ± 0% 1.35µs ± 0% −82.74%
BlasGer/Sger_30x300_(2_10)-12 7.78µs ± 0% 1.33µs ± 0% −82.90%
BlasGer/Sger_30x300_(3_1)-12 7.77µs ± 0% 1.40µs ± 0% −81.99%
BlasGer/Sger_30x300_(3_2)-12 7.69µs ± 0% 1.44µs ± 0% −81.23%
BlasGer/Sger_30x300_(3_3)-12 7.74µs ± 0% 1.22µs ± 0% −84.21%
BlasGer/Sger_30x300_(3_4)-12 7.65µs ± 0% 1.44µs ± 0% −81.17%
BlasGer/Sger_30x300_(3_10)-12 7.77µs ± 0% 1.18µs ± 0% −84.79%
BlasGer/Sger_30x300_(4_1)-12 7.74µs ± 0% 1.27µs ± 0% −83.54%
BlasGer/Sger_30x300_(4_2)-12 7.88µs ± 0% 1.54µs ± 0% −80.40%
BlasGer/Sger_30x300_(4_3)-12 7.10µs ± 0% 1.43µs ± 0% −79.92%
BlasGer/Sger_30x300_(4_4)-12 7.77µs ± 0% 1.19µs ± 0% −84.73%
BlasGer/Sger_30x300_(4_10)-12 7.80µs ± 0% 1.22µs ± 0% −84.34%
BlasGer/Sger_30x300_(10_1)-12 8.36µs ± 0% 1.24µs ± 0% −85.17%
BlasGer/Sger_30x300_(10_2)-12 7.83µs ± 0% 1.28µs ± 0% −83.61%
BlasGer/Sger_30x300_(10_3)-12 7.57µs ± 0% 1.19µs ± 0% −84.34%
BlasGer/Sger_30x300_(10_4)-12 6.67µs ± 0% 1.42µs ± 0% −78.65%
BlasGer/Sger_30x300_(10_10)-12 8.01µs ± 0% 1.37µs ± 0% −82.91%
BlasGer/Sger_30x1000_(1_1)-12 6.60µs ± 0% 4.31µs ± 0% −34.74%
BlasGer/Sger_30x1000_(1_2)-12 24.9µs ± 0% 4.8µs ± 0% −80.67%
BlasGer/Sger_30x1000_(1_3)-12 26.2µs ± 0% 4.0µs ± 0% −84.91%
BlasGer/Sger_30x1000_(1_4)-12 26.3µs ± 0% 4.2µs ± 0% −83.92%
BlasGer/Sger_30x1000_(1_10)-12 26.3µs ± 0% 6.0µs ± 0% −77.33%
BlasGer/Sger_30x1000_(2_1)-12 26.5µs ± 0% 4.8µs ± 0% −81.90%
BlasGer/Sger_30x1000_(2_2)-12 25.2µs ± 0% 4.7µs ± 0% −81.38%
BlasGer/Sger_30x1000_(2_3)-12 22.5µs ± 0% 4.0µs ± 0% −82.12%
BlasGer/Sger_30x1000_(2_4)-12 25.4µs ± 0% 5.3µs ± 0% −79.35%
BlasGer/Sger_30x1000_(2_10)-12 25.2µs ± 0% 5.2µs ± 0% −79.42%
BlasGer/Sger_30x1000_(3_1)-12 26.5µs ± 0% 4.8µs ± 0% −81.97%
BlasGer/Sger_30x1000_(3_2)-12 23.6µs ± 0% 4.8µs ± 0% −79.76%
BlasGer/Sger_30x1000_(3_3)-12 26.1µs ± 0% 4.8µs ± 0% −81.59%
BlasGer/Sger_30x1000_(3_4)-12 25.5µs ± 0% 4.8µs ± 0% −81.30%
BlasGer/Sger_30x1000_(3_10)-12 25.9µs ± 0% 5.4µs ± 0% −79.26%
BlasGer/Sger_30x1000_(4_1)-12 25.8µs ± 0% 4.6µs ± 0% −81.95%
BlasGer/Sger_30x1000_(4_2)-12 26.2µs ± 0% 4.4µs ± 0% −83.09%
BlasGer/Sger_30x1000_(4_3)-12 24.3µs ± 0% 4.7µs ± 0% −80.68%
BlasGer/Sger_30x1000_(4_4)-12 23.6µs ± 0% 4.8µs ± 0% −79.66%
BlasGer/Sger_30x1000_(4_10)-12 22.6µs ± 0% 5.7µs ± 0% −74.72%
BlasGer/Sger_30x1000_(10_1)-12 25.6µs ± 0% 5.7µs ± 0% −77.63%
BlasGer/Sger_30x1000_(10_2)-12 26.3µs ± 0% 4.8µs ± 0% −81.95%
BlasGer/Sger_30x1000_(10_3)-12 26.2µs ± 0% 4.2µs ± 0% −84.01%
BlasGer/Sger_30x1000_(10_4)-12 26.1µs ± 0% 5.3µs ± 0% −79.79%
BlasGer/Sger_30x1000_(10_10)-12 26.3µs ± 0% 5.8µs ± 0% −78.07%
BlasGer/Sger_100x3_(1_1)-12 1.12µs ± 0% 0.15µs ± 0% −86.79%
BlasGer/Sger_100x3_(1_2)-12 1.05µs ± 0% 0.13µs ± 0% −87.45%
BlasGer/Sger_100x3_(1_3)-12 1.22µs ± 0% 0.14µs ± 0% −88.44%
BlasGer/Sger_100x3_(1_4)-12 1.11µs ± 0% 0.13µs ± 0% −88.04%
BlasGer/Sger_100x3_(1_10)-12 1.20µs ± 0% 0.13µs ± 0% −89.07%
BlasGer/Sger_100x3_(2_1)-12 1.17µs ± 0% 0.15µs ± 0% −87.41%
BlasGer/Sger_100x3_(2_2)-12 1.21µs ± 0% 0.14µs ± 0% −88.75%
BlasGer/Sger_100x3_(2_3)-12 1.09µs ± 0% 0.14µs ± 0% −86.75%
BlasGer/Sger_100x3_(2_4)-12 1.20µs ± 0% 0.15µs ± 0% −87.21%
BlasGer/Sger_100x3_(2_10)-12 1.16µs ± 0% 0.15µs ± 0% −87.19%
BlasGer/Sger_100x3_(3_1)-12 1.16µs ± 0% 0.13µs ± 0% −88.70%
BlasGer/Sger_100x3_(3_2)-12 1.23µs ± 0% 0.13µs ± 0% −89.24%
BlasGer/Sger_100x3_(3_3)-12 1.03µs ± 0% 0.14µs ± 0% −86.80%
BlasGer/Sger_100x3_(3_4)-12 1.18µs ± 0% 0.15µs ± 0% −87.51%
BlasGer/Sger_100x3_(3_10)-12 1.06µs ± 0% 0.13µs ± 0% −87.68%
BlasGer/Sger_100x3_(4_1)-12 1.07µs ± 0% 0.15µs ± 0% −85.85%
BlasGer/Sger_100x3_(4_2)-12 1.18µs ± 0% 0.15µs ± 0% −86.99%
BlasGer/Sger_100x3_(4_3)-12 1.22µs ± 0% 0.16µs ± 0% −86.50%
BlasGer/Sger_100x3_(4_4)-12 1.03µs ± 0% 0.15µs ± 0% −85.41%
BlasGer/Sger_100x3_(4_10)-12 1.22µs ± 0% 0.15µs ± 0% −87.45%
BlasGer/Sger_100x3_(10_1)-12 1.12µs ± 0% 0.15µs ± 0% −86.58%
BlasGer/Sger_100x3_(10_2)-12 1.18µs ± 0% 0.14µs ± 0% −87.82%
BlasGer/Sger_100x3_(10_3)-12 1.20µs ± 0% 0.13µs ± 0% −89.26%
BlasGer/Sger_100x3_(10_4)-12 1.18µs ± 0% 0.13µs ± 0% −89.28%
BlasGer/Sger_100x3_(10_10)-12 1.20µs ± 0% 0.15µs ± 0% −87.59%
BlasGer/Sger_100x10_(1_1)-12 1.21µs ± 0% 0.21µs ± 0% −82.54%
BlasGer/Sger_100x10_(1_2)-12 1.58µs ± 0% 0.24µs ± 0% −84.80%
BlasGer/Sger_100x10_(1_3)-12 1.60µs ± 0% 0.21µs ± 0% −86.70%
BlasGer/Sger_100x10_(1_4)-12 1.58µs ± 0% 0.24µs ± 0% −84.84%
BlasGer/Sger_100x10_(1_10)-12 1.60µs ± 0% 0.24µs ± 0% −85.10%
BlasGer/Sger_100x10_(2_1)-12 1.60µs ± 0% 0.23µs ± 0% −85.65%
BlasGer/Sger_100x10_(2_2)-12 1.60µs ± 0% 0.24µs ± 0% −84.76%
BlasGer/Sger_100x10_(2_3)-12 2.01µs ± 0% 0.19µs ± 0% −90.59%
BlasGer/Sger_100x10_(2_4)-12 1.57µs ± 0% 0.21µs ± 0% −86.72%
BlasGer/Sger_100x10_(2_10)-12 1.52µs ± 0% 0.20µs ± 0% −86.67%
BlasGer/Sger_100x10_(3_1)-12 1.63µs ± 0% 0.22µs ± 0% −86.69%
BlasGer/Sger_100x10_(3_2)-12 1.56µs ± 0% 0.22µs ± 0% −86.18%
BlasGer/Sger_100x10_(3_3)-12 1.47µs ± 0% 0.23µs ± 0% −84.02%
BlasGer/Sger_100x10_(3_4)-12 1.34µs ± 0% 0.23µs ± 0% −82.52%
BlasGer/Sger_100x10_(3_10)-12 2.02µs ± 0% 0.21µs ± 0% −89.79%
BlasGer/Sger_100x10_(4_1)-12 1.57µs ± 0% 0.24µs ± 0% −84.70%
BlasGer/Sger_100x10_(4_2)-12 1.61µs ± 0% 0.22µs ± 0% −86.37%
BlasGer/Sger_100x10_(4_3)-12 1.59µs ± 0% 0.23µs ± 0% −85.44%
BlasGer/Sger_100x10_(4_4)-12 1.59µs ± 0% 0.21µs ± 0% −86.71%
BlasGer/Sger_100x10_(4_10)-12 1.81µs ± 0% 0.21µs ± 0% −88.21%
BlasGer/Sger_100x10_(10_1)-12 1.59µs ± 0% 0.24µs ± 0% −84.87%
BlasGer/Sger_100x10_(10_2)-12 1.70µs ± 0% 0.19µs ± 0% −88.95%
BlasGer/Sger_100x10_(10_3)-12 1.42µs ± 0% 0.26µs ± 0% −82.02%
BlasGer/Sger_100x10_(10_4)-12 1.61µs ± 0% 0.24µs ± 0% −85.01%
BlasGer/Sger_100x10_(10_10)-12 1.35µs ± 0% 0.21µs ± 0% −84.17%
BlasGer/Sger_100x30_(1_1)-12 1.60µs ± 0% 0.44µs ± 0% −72.69%
BlasGer/Sger_100x30_(1_2)-12 3.21µs ± 0% 0.53µs ± 0% −83.59%
BlasGer/Sger_100x30_(1_3)-12 3.24µs ± 0% 0.48µs ± 0% −85.27%
BlasGer/Sger_100x30_(1_4)-12 3.25µs ± 0% 0.51µs ± 0% −84.39%
BlasGer/Sger_100x30_(1_10)-12 2.84µs ± 0% 0.44µs ± 0% −84.52%
BlasGer/Sger_100x30_(2_1)-12 3.24µs ± 0% 0.44µs ± 0% −86.36%
BlasGer/Sger_100x30_(2_2)-12 3.47µs ± 0% 0.45µs ± 0% −86.99%
BlasGer/Sger_100x30_(2_3)-12 3.27µs ± 0% 0.44µs ± 0% −86.53%
BlasGer/Sger_100x30_(2_4)-12 3.14µs ± 0% 0.58µs ± 0% −81.60%
BlasGer/Sger_100x30_(2_10)-12 2.71µs ± 0% 0.47µs ± 0% −82.75%
BlasGer/Sger_100x30_(3_1)-12 3.64µs ± 0% 0.58µs ± 0% −84.08%
BlasGer/Sger_100x30_(3_2)-12 3.19µs ± 0% 0.52µs ± 0% −83.83%
BlasGer/Sger_100x30_(3_3)-12 3.25µs ± 0% 0.56µs ± 0% −82.81%
BlasGer/Sger_100x30_(3_4)-12 2.91µs ± 0% 0.44µs ± 0% −84.87%
BlasGer/Sger_100x30_(3_10)-12 3.24µs ± 0% 0.56µs ± 0% −82.77%
BlasGer/Sger_100x30_(4_1)-12 3.04µs ± 0% 0.52µs ± 0% −82.79%
BlasGer/Sger_100x30_(4_2)-12 2.83µs ± 0% 0.52µs ± 0% −81.51%
BlasGer/Sger_100x30_(4_3)-12 3.25µs ± 0% 0.44µs ± 0% −86.30%
BlasGer/Sger_100x30_(4_4)-12 3.28µs ± 0% 0.52µs ± 0% −84.14%
BlasGer/Sger_100x30_(4_10)-12 2.84µs ± 0% 0.58µs ± 0% −79.39%
BlasGer/Sger_100x30_(10_1)-12 3.15µs ± 0% 0.65µs ± 0% −79.24%
BlasGer/Sger_100x30_(10_2)-12 3.19µs ± 0% 0.52µs ± 0% −83.53%
BlasGer/Sger_100x30_(10_3)-12 3.38µs ± 0% 0.44µs ± 0% −86.94%
BlasGer/Sger_100x30_(10_4)-12 3.24µs ± 0% 0.57µs ± 0% −82.48%
BlasGer/Sger_100x30_(10_10)-12 3.41µs ± 0% 0.57µs ± 0% −83.29%
BlasGer/Sger_100x100_(1_1)-12 2.48µs ± 0% 1.39µs ± 0% −44.20%
BlasGer/Sger_100x100_(1_2)-12 8.61µs ± 0% 1.38µs ± 0% −83.99%
BlasGer/Sger_100x100_(1_3)-12 8.54µs ± 0% 1.38µs ± 0% −83.83%
BlasGer/Sger_100x100_(1_4)-12 9.04µs ± 0% 1.54µs ± 0% −82.98%
BlasGer/Sger_100x100_(1_10)-12 8.66µs ± 0% 1.56µs ± 0% −82.01%
BlasGer/Sger_100x100_(2_1)-12 8.63µs ± 0% 1.56µs ± 0% −81.87%
BlasGer/Sger_100x100_(2_2)-12 8.58µs ± 0% 1.60µs ± 0% −81.34%
BlasGer/Sger_100x100_(2_3)-12 8.41µs ± 0% 1.50µs ± 0% −82.11%
BlasGer/Sger_100x100_(2_4)-12 8.60µs ± 0% 1.60µs ± 0% −81.35%
BlasGer/Sger_100x100_(2_10)-12 8.77µs ± 0% 1.59µs ± 0% −81.88%
BlasGer/Sger_100x100_(3_1)-12 8.65µs ± 0% 1.44µs ± 0% −83.36%
BlasGer/Sger_100x100_(3_2)-12 8.65µs ± 0% 1.61µs ± 0% −81.34%
BlasGer/Sger_100x100_(3_3)-12 8.74µs ± 0% 1.78µs ± 0% −79.69%
BlasGer/Sger_100x100_(3_4)-12 8.61µs ± 0% 1.58µs ± 0% −81.68%
BlasGer/Sger_100x100_(3_10)-12 8.66µs ± 0% 1.61µs ± 0% −81.45%
BlasGer/Sger_100x100_(4_1)-12 8.60µs ± 0% 1.34µs ± 0% −84.45%
BlasGer/Sger_100x100_(4_2)-12 8.59µs ± 0% 1.59µs ± 0% −81.51%
BlasGer/Sger_100x100_(4_3)-12 8.57µs ± 0% 1.66µs ± 0% −80.65%
BlasGer/Sger_100x100_(4_4)-12 8.54µs ± 0% 1.51µs ± 0% −82.25%
BlasGer/Sger_100x100_(4_10)-12 8.45µs ± 0% 1.46µs ± 0% −82.69%
BlasGer/Sger_100x100_(10_1)-12 9.60µs ± 0% 1.62µs ± 0% −83.17%
BlasGer/Sger_100x100_(10_2)-12 8.89µs ± 0% 1.37µs ± 0% −84.61%
BlasGer/Sger_100x100_(10_3)-12 7.72µs ± 0% 1.70µs ± 0% −78.03%
BlasGer/Sger_100x100_(10_4)-12 7.65µs ± 0% 1.35µs ± 0% −82.39%
BlasGer/Sger_100x100_(10_10)-12 8.79µs ± 0% 1.56µs ± 0% −82.25%
BlasGer/Sger_100x300_(1_1)-12 6.51µs ± 0% 3.35µs ± 0% −48.47%
BlasGer/Sger_100x300_(1_2)-12 26.3µs ± 0% 4.9µs ± 0% −81.44%
BlasGer/Sger_100x300_(1_3)-12 26.9µs ± 0% 4.8µs ± 0% −82.15%
BlasGer/Sger_100x300_(1_4)-12 24.0µs ± 0% 5.3µs ± 0% −77.73%
BlasGer/Sger_100x300_(1_10)-12 22.2µs ± 0% 4.7µs ± 0% −78.65%
BlasGer/Sger_100x300_(2_1)-12 26.7µs ± 0% 4.0µs ± 0% −84.85%
BlasGer/Sger_100x300_(2_2)-12 26.5µs ± 0% 5.3µs ± 0% −79.87%
BlasGer/Sger_100x300_(2_3)-12 26.5µs ± 0% 5.2µs ± 0% −80.49%
BlasGer/Sger_100x300_(2_4)-12 24.1µs ± 0% 4.6µs ± 0% −81.01%
BlasGer/Sger_100x300_(2_10)-12 26.7µs ± 0% 4.2µs ± 0% −84.42%
BlasGer/Sger_100x300_(3_1)-12 25.4µs ± 0% 4.4µs ± 0% −82.90%
BlasGer/Sger_100x300_(3_2)-12 26.5µs ± 0% 4.3µs ± 0% −83.82%
BlasGer/Sger_100x300_(3_3)-12 26.8µs ± 0% 5.1µs ± 0% −80.84%
BlasGer/Sger_100x300_(3_4)-12 24.2µs ± 0% 5.1µs ± 0% −78.96%
BlasGer/Sger_100x300_(3_10)-12 22.7µs ± 0% 5.1µs ± 0% −77.45%
BlasGer/Sger_100x300_(4_1)-12 26.7µs ± 0% 4.8µs ± 0% −82.23%
BlasGer/Sger_100x300_(4_2)-12 23.8µs ± 0% 4.4µs ± 0% −81.68%
BlasGer/Sger_100x300_(4_3)-12 26.6µs ± 0% 4.8µs ± 0% −82.12%
BlasGer/Sger_100x300_(4_4)-12 26.4µs ± 0% 4.6µs ± 0% −82.50%
BlasGer/Sger_100x300_(4_10)-12 27.5µs ± 0% 4.5µs ± 0% −83.77%
BlasGer/Sger_100x300_(10_1)-12 21.8µs ± 0% 4.4µs ± 0% −79.90%
BlasGer/Sger_100x300_(10_2)-12 26.3µs ± 0% 4.4µs ± 0% −83.26%
BlasGer/Sger_100x300_(10_3)-12 22.1µs ± 0% 4.6µs ± 0% −79.34%
BlasGer/Sger_100x300_(10_4)-12 27.7µs ± 0% 4.3µs ± 0% −84.63%
BlasGer/Sger_100x300_(10_10)-12 26.0µs ± 0% 4.1µs ± 0% −84.15%
BlasGer/Sger_100x1000_(1_1)-12 24.3µs ± 0% 16.3µs ± 0% −32.73%
BlasGer/Sger_100x1000_(1_2)-12 82.2µs ± 0% 23.2µs ± 0% −71.75%
BlasGer/Sger_100x1000_(1_3)-12 80.4µs ± 0% 23.0µs ± 0% −71.41%
BlasGer/Sger_100x1000_(1_4)-12 71.9µs ± 0% 23.2µs ± 0% −67.80%
BlasGer/Sger_100x1000_(1_10)-12 82.9µs ± 0% 25.2µs ± 0% −69.67%
BlasGer/Sger_100x1000_(2_1)-12 87.8µs ± 0% 19.6µs ± 0% −77.71%
BlasGer/Sger_100x1000_(2_2)-12 82.6µs ± 0% 21.4µs ± 0% −74.05%
BlasGer/Sger_100x1000_(2_3)-12 80.3µs ± 0% 22.9µs ± 0% −71.47%
BlasGer/Sger_100x1000_(2_4)-12 82.2µs ± 0% 20.8µs ± 0% −74.68%
BlasGer/Sger_100x1000_(2_10)-12 75.5µs ± 0% 20.7µs ± 0% −72.65%
BlasGer/Sger_100x1000_(3_1)-12 80.9µs ± 0% 20.9µs ± 0% −74.19%
BlasGer/Sger_100x1000_(3_2)-12 82.5µs ± 0% 20.1µs ± 0% −75.63%
BlasGer/Sger_100x1000_(3_3)-12 90.7µs ± 0% 23.6µs ± 0% −73.95%
BlasGer/Sger_100x1000_(3_4)-12 85.1µs ± 0% 19.8µs ± 0% −76.78%
BlasGer/Sger_100x1000_(3_10)-12 70.3µs ± 0% 25.2µs ± 0% −64.15%
BlasGer/Sger_100x1000_(4_1)-12 83.8µs ± 0% 19.2µs ± 0% −77.12%
BlasGer/Sger_100x1000_(4_2)-12 75.0µs ± 0% 23.5µs ± 0% −68.68%
BlasGer/Sger_100x1000_(4_3)-12 82.1µs ± 0% 20.9µs ± 0% −74.54%
BlasGer/Sger_100x1000_(4_4)-12 82.2µs ± 0% 23.8µs ± 0% −71.04%
BlasGer/Sger_100x1000_(4_10)-12 70.0µs ± 0% 22.6µs ± 0% −67.76%
BlasGer/Sger_100x1000_(10_1)-12 79.9µs ± 0% 18.0µs ± 0% −77.50%
BlasGer/Sger_100x1000_(10_2)-12 71.8µs ± 0% 18.7µs ± 0% −73.87%
BlasGer/Sger_100x1000_(10_3)-12 82.5µs ± 0% 21.2µs ± 0% −74.26%
BlasGer/Sger_100x1000_(10_4)-12 82.8µs ± 0% 20.6µs ± 0% −75.14%
BlasGer/Sger_100x1000_(10_10)-12 82.9µs ± 0% 24.5µs ± 0% −70.45%
BlasGer/Sger_300x10_(1_1)-12 4.04µs ± 0% 0.61µs ± 0% −85.01%
BlasGer/Sger_300x10_(1_2)-12 4.49µs ± 0% 0.58µs ± 0% −87.01%
BlasGer/Sger_300x10_(1_3)-12 4.52µs ± 0% 0.69µs ± 0% −84.85%
BlasGer/Sger_300x10_(1_4)-12 4.33µs ± 0% 0.70µs ± 0% −83.95%
BlasGer/Sger_300x10_(1_10)-12 3.99µs ± 0% 0.69µs ± 0% −82.84%
BlasGer/Sger_300x10_(2_1)-12 4.02µs ± 0% 0.64µs ± 0% −83.99%
BlasGer/Sger_300x10_(2_2)-12 4.94µs ± 0% 0.58µs ± 0% −88.17%
BlasGer/Sger_300x10_(2_3)-12 4.00µs ± 0% 0.60µs ± 0% −84.93%
BlasGer/Sger_300x10_(2_4)-12 4.46µs ± 0% 0.77µs ± 0% −82.76%
BlasGer/Sger_300x10_(2_10)-12 4.89µs ± 0% 0.58µs ± 0% −88.05%
BlasGer/Sger_300x10_(3_1)-12 4.72µs ± 0% 0.69µs ± 0% −85.40%
BlasGer/Sger_300x10_(3_2)-12 4.81µs ± 0% 0.68µs ± 0% −85.98%
BlasGer/Sger_300x10_(3_3)-12 5.13µs ± 0% 0.62µs ± 0% −87.99%
BlasGer/Sger_300x10_(3_4)-12 5.15µs ± 0% 0.68µs ± 0% −86.75%
BlasGer/Sger_300x10_(3_10)-12 4.02µs ± 0% 0.56µs ± 0% −86.03%
BlasGer/Sger_300x10_(4_1)-12 4.00µs ± 0% 0.62µs ± 0% −84.53%
BlasGer/Sger_300x10_(4_2)-12 5.30µs ± 0% 0.61µs ± 0% −88.40%
BlasGer/Sger_300x10_(4_3)-12 5.01µs ± 0% 0.63µs ± 0% −87.36%
BlasGer/Sger_300x10_(4_4)-12 4.52µs ± 0% 0.61µs ± 0% −86.60%
BlasGer/Sger_300x10_(4_10)-12 5.31µs ± 0% 0.69µs ± 0% −86.97%
BlasGer/Sger_300x10_(10_1)-12 4.85µs ± 0% 0.57µs ± 0% −88.20%
BlasGer/Sger_300x10_(10_2)-12 4.09µs ± 0% 0.64µs ± 0% −84.43%
BlasGer/Sger_300x10_(10_3)-12 5.41µs ± 0% 0.67µs ± 0% −87.61%
BlasGer/Sger_300x10_(10_4)-12 4.86µs ± 0% 0.70µs ± 0% −85.65%
BlasGer/Sger_300x10_(10_10)-12 4.13µs ± 0% 0.68µs ± 0% −83.65%
BlasGer/Sger_300x30_(1_1)-12 6.18µs ± 0% 1.27µs ± 0% −79.52%
BlasGer/Sger_300x30_(1_2)-12 9.44µs ± 0% 1.58µs ± 0% −83.23%
BlasGer/Sger_300x30_(1_3)-12 9.27µs ± 0% 1.56µs ± 0% −83.13%
BlasGer/Sger_300x30_(1_4)-12 9.67µs ± 0% 1.56µs ± 0% −83.88%
BlasGer/Sger_300x30_(1_10)-12 10.3µs ± 0% 1.4µs ± 0% −86.91%
BlasGer/Sger_300x30_(2_1)-12 10.5µs ± 0% 1.3µs ± 0% −87.24%
BlasGer/Sger_300x30_(2_2)-12 9.34µs ± 0% 1.64µs ± 0% −82.48%
BlasGer/Sger_300x30_(2_3)-12 8.14µs ± 0% 1.39µs ± 0% −82.89%
BlasGer/Sger_300x30_(2_4)-12 9.37µs ± 0% 1.46µs ± 0% −84.44%
BlasGer/Sger_300x30_(2_10)-12 8.12µs ± 0% 1.59µs ± 0% −80.44%
BlasGer/Sger_300x30_(3_1)-12 9.46µs ± 0% 1.60µs ± 0% −83.12%
BlasGer/Sger_300x30_(3_2)-12 10.4µs ± 0% 1.6µs ± 0% −84.48%
BlasGer/Sger_300x30_(3_3)-12 10.8µs ± 0% 1.4µs ± 0% −87.20%
BlasGer/Sger_300x30_(3_4)-12 10.5µs ± 0% 1.6µs ± 0% −84.55%
BlasGer/Sger_300x30_(3_10)-12 10.7µs ± 0% 1.4µs ± 0% −87.09%
BlasGer/Sger_300x30_(4_1)-12 8.88µs ± 0% 1.34µs ± 0% −84.86%
BlasGer/Sger_300x30_(4_2)-12 8.13µs ± 0% 1.61µs ± 0% −80.23%
BlasGer/Sger_300x30_(4_3)-12 10.4µs ± 0% 1.5µs ± 0% −85.58%
BlasGer/Sger_300x30_(4_4)-12 9.33µs ± 0% 1.69µs ± 0% −81.89%
BlasGer/Sger_300x30_(4_10)-12 9.40µs ± 0% 1.60µs ± 0% −83.03%
BlasGer/Sger_300x30_(10_1)-12 9.53µs ± 0% 1.63µs ± 0% −82.86%
BlasGer/Sger_300x30_(10_2)-12 10.7µs ± 0% 1.6µs ± 0% −84.66%
BlasGer/Sger_300x30_(10_3)-12 10.8µs ± 0% 1.6µs ± 0% −84.82%
BlasGer/Sger_300x30_(10_4)-12 9.45µs ± 0% 1.44µs ± 0% −84.74%
BlasGer/Sger_300x30_(10_10)-12 10.6µs ± 0% 1.4µs ± 0% −86.96%
BlasGer/Sger_300x100_(1_1)-12 8.15µs ± 0% 3.48µs ± 0% −57.28%
BlasGer/Sger_300x100_(1_2)-12 24.6µs ± 0% 4.0µs ± 0% −83.70%
BlasGer/Sger_300x100_(1_3)-12 24.3µs ± 0% 4.5µs ± 0% −81.32%
BlasGer/Sger_300x100_(1_4)-12 25.5µs ± 0% 4.6µs ± 0% −81.83%
BlasGer/Sger_300x100_(1_10)-12 24.4µs ± 0% 5.3µs ± 0% −78.48%
BlasGer/Sger_300x100_(2_1)-12 23.5µs ± 0% 4.1µs ± 0% −82.75%
BlasGer/Sger_300x100_(2_2)-12 28.2µs ± 0% 4.9µs ± 0% −82.69%
BlasGer/Sger_300x100_(2_3)-12 23.1µs ± 0% 5.3µs ± 0% −76.84%
BlasGer/Sger_300x100_(2_4)-12 24.3µs ± 0% 5.4µs ± 0% −77.75%
BlasGer/Sger_300x100_(2_10)-12 25.1µs ± 0% 4.9µs ± 0% −80.59%
BlasGer/Sger_300x100_(3_1)-12 26.2µs ± 0% 5.1µs ± 0% −80.49%
BlasGer/Sger_300x100_(3_2)-12 27.0µs ± 0% 4.9µs ± 0% −81.93%
BlasGer/Sger_300x100_(3_3)-12 25.7µs ± 0% 5.4µs ± 0% −79.02%
BlasGer/Sger_300x100_(3_4)-12 24.0µs ± 0% 4.7µs ± 0% −80.43%
BlasGer/Sger_300x100_(3_10)-12 27.9µs ± 0% 4.2µs ± 0% −84.87%
BlasGer/Sger_300x100_(4_1)-12 25.7µs ± 0% 4.5µs ± 0% −82.40%
BlasGer/Sger_300x100_(4_2)-12 27.5µs ± 0% 5.1µs ± 0% −81.49%
BlasGer/Sger_300x100_(4_3)-12 27.2µs ± 0% 4.6µs ± 0% −83.08%
BlasGer/Sger_300x100_(4_4)-12 27.2µs ± 0% 5.1µs ± 0% −81.11%
BlasGer/Sger_300x100_(4_10)-12 27.1µs ± 0% 5.1µs ± 0% −81.23%
BlasGer/Sger_300x100_(10_1)-12 26.9µs ± 0% 5.2µs ± 0% −80.79%
BlasGer/Sger_300x100_(10_2)-12 24.7µs ± 0% 4.9µs ± 0% −80.25%
BlasGer/Sger_300x100_(10_3)-12 22.7µs ± 0% 4.3µs ± 0% −80.82%
BlasGer/Sger_300x100_(10_4)-12 25.0µs ± 0% 4.5µs ± 0% −82.16%
BlasGer/Sger_300x100_(10_10)-12 27.2µs ± 0% 4.9µs ± 0% −82.13%
BlasGer/Sger_300x300_(1_1)-12 20.4µs ± 0% 14.5µs ± 0% −28.74%
BlasGer/Sger_300x300_(1_2)-12 74.5µs ± 0% 18.5µs ± 0% −75.18%
BlasGer/Sger_300x300_(1_3)-12 65.9µs ± 0% 20.5µs ± 0% −68.87%
BlasGer/Sger_300x300_(1_4)-12 71.6µs ± 0% 21.4µs ± 0% −70.10%
BlasGer/Sger_300x300_(1_10)-12 77.3µs ± 0% 20.2µs ± 0% −73.87%
BlasGer/Sger_300x300_(2_1)-12 78.6µs ± 0% 17.6µs ± 0% −77.56%
BlasGer/Sger_300x300_(2_2)-12 77.8µs ± 0% 20.6µs ± 0% −73.54%
BlasGer/Sger_300x300_(2_3)-12 76.9µs ± 0% 20.9µs ± 0% −72.81%
BlasGer/Sger_300x300_(2_4)-12 77.1µs ± 0% 17.2µs ± 0% −77.70%
BlasGer/Sger_300x300_(2_10)-12 78.7µs ± 0% 17.7µs ± 0% −77.55%
BlasGer/Sger_300x300_(3_1)-12 75.5µs ± 0% 17.3µs ± 0% −77.11%
BlasGer/Sger_300x300_(3_2)-12 77.8µs ± 0% 15.2µs ± 0% −80.44%
BlasGer/Sger_300x300_(3_3)-12 76.0µs ± 0% 21.8µs ± 0% −71.33%
BlasGer/Sger_300x300_(3_4)-12 66.6µs ± 0% 15.9µs ± 0% −76.13%
BlasGer/Sger_300x300_(3_10)-12 77.4µs ± 0% 18.1µs ± 0% −76.67%
BlasGer/Sger_300x300_(4_1)-12 76.2µs ± 0% 16.9µs ± 0% −77.84%
BlasGer/Sger_300x300_(4_2)-12 66.2µs ± 0% 20.3µs ± 0% −69.36%
BlasGer/Sger_300x300_(4_3)-12 74.8µs ± 0% 17.7µs ± 0% −76.29%
BlasGer/Sger_300x300_(4_4)-12 76.9µs ± 0% 20.3µs ± 0% −73.58%
BlasGer/Sger_300x300_(4_10)-12 78.0µs ± 0% 16.7µs ± 0% −78.57%
BlasGer/Sger_300x300_(10_1)-12 76.4µs ± 0% 17.7µs ± 0% −76.77%
BlasGer/Sger_300x300_(10_2)-12 76.1µs ± 0% 17.3µs ± 0% −77.26%
BlasGer/Sger_300x300_(10_3)-12 70.9µs ± 0% 21.6µs ± 0% −69.47%
BlasGer/Sger_300x300_(10_4)-12 68.5µs ± 0% 18.3µs ± 0% −73.22%
BlasGer/Sger_300x300_(10_10)-12 75.9µs ± 0% 18.5µs ± 0% −75.69%
BlasGer/Sger_300x1000_(1_1)-12 68.7µs ± 0% 61.1µs ± 0% −11.03%
BlasGer/Sger_300x1000_(1_2)-12 259µs ± 0% 62µs ± 0% −75.92%
BlasGer/Sger_300x1000_(1_3)-12 253µs ± 0% 58µs ± 0% −76.96%
BlasGer/Sger_300x1000_(1_4)-12 230µs ± 0% 63µs ± 0% −72.86%
BlasGer/Sger_300x1000_(1_10)-12 255µs ± 0% 64µs ± 0% −74.87%
BlasGer/Sger_300x1000_(2_1)-12 245µs ± 0% 56µs ± 0% −77.23%
BlasGer/Sger_300x1000_(2_2)-12 260µs ± 0% 55µs ± 0% −78.68%
BlasGer/Sger_300x1000_(2_3)-12 210µs ± 0% 55µs ± 0% −73.83%
BlasGer/Sger_300x1000_(2_4)-12 210µs ± 0% 72µs ± 0% −65.82%
BlasGer/Sger_300x1000_(2_10)-12 233µs ± 0% 61µs ± 0% −73.95%
BlasGer/Sger_300x1000_(3_1)-12 234µs ± 0% 68µs ± 0% −70.96%
BlasGer/Sger_300x1000_(3_2)-12 258µs ± 0% 56µs ± 0% −78.34%
BlasGer/Sger_300x1000_(3_3)-12 260µs ± 0% 69µs ± 0% −73.41%
BlasGer/Sger_300x1000_(3_4)-12 284µs ± 0% 57µs ± 0% −79.88%
BlasGer/Sger_300x1000_(3_10)-12 286µs ± 0% 61µs ± 0% −78.50%
BlasGer/Sger_300x1000_(4_1)-12 256µs ± 0% 62µs ± 0% −75.95%
BlasGer/Sger_300x1000_(4_2)-12 247µs ± 0% 66µs ± 0% −73.23%
BlasGer/Sger_300x1000_(4_3)-12 232µs ± 0% 69µs ± 0% −70.09%
BlasGer/Sger_300x1000_(4_4)-12 216µs ± 0% 66µs ± 0% −69.28%
BlasGer/Sger_300x1000_(4_10)-12 231µs ± 0% 70µs ± 0% −69.76%
BlasGer/Sger_300x1000_(10_1)-12 257µs ± 0% 56µs ± 0% −78.31%
BlasGer/Sger_300x1000_(10_2)-12 259µs ± 0% 69µs ± 0% −73.38%
BlasGer/Sger_300x1000_(10_3)-12 213µs ± 0% 67µs ± 0% −68.50%
BlasGer/Sger_300x1000_(10_4)-12 253µs ± 0% 61µs ± 0% −75.89%
BlasGer/Sger_300x1000_(10_10)-12 282µs ± 0% 74µs ± 0% −73.60%
BlasGer/Sger_300x10000_(1_1)-12 751µs ± 0% 755µs ± 0% +0.50%
BlasGer/Sger_300x10000_(1_2)-12 2.17ms ± 0% 0.62ms ± 0% −71.36%
BlasGer/Sger_300x10000_(1_3)-12 2.46ms ± 0% 0.72ms ± 0% −70.60%
BlasGer/Sger_300x10000_(1_4)-12 2.17ms ± 0% 0.85ms ± 0% −60.76%
BlasGer/Sger_300x10000_(1_10)-12 3.64ms ± 0% 0.97ms ± 0% −73.47%
BlasGer/Sger_300x10000_(2_1)-12 2.41ms ± 0% 0.59ms ± 0% −75.42%
BlasGer/Sger_300x10000_(2_2)-12 2.55ms ± 0% 0.73ms ± 0% −71.53%
BlasGer/Sger_300x10000_(2_3)-12 2.48ms ± 0% 0.75ms ± 0% −69.80%
BlasGer/Sger_300x10000_(2_4)-12 2.14ms ± 0% 0.80ms ± 0% −62.78%
BlasGer/Sger_300x10000_(2_10)-12 3.43ms ± 0% 1.13ms ± 0% −66.97%
BlasGer/Sger_300x10000_(3_1)-12 2.35ms ± 0% 0.67ms ± 0% −71.31%
BlasGer/Sger_300x10000_(3_2)-12 2.47ms ± 0% 0.74ms ± 0% −69.95%
BlasGer/Sger_300x10000_(3_3)-12 2.13ms ± 0% 0.81ms ± 0% −61.68%
BlasGer/Sger_300x10000_(3_4)-12 2.49ms ± 0% 0.76ms ± 0% −69.24%
BlasGer/Sger_300x10000_(3_10)-12 3.34ms ± 0% 0.96ms ± 0% −71.26%
BlasGer/Sger_300x10000_(4_1)-12 2.45ms ± 0% 0.70ms ± 0% −71.42%
BlasGer/Sger_300x10000_(4_2)-12 2.12ms ± 0% 0.72ms ± 0% −66.23%
BlasGer/Sger_300x10000_(4_3)-12 2.50ms ± 0% 0.72ms ± 0% −71.34%
BlasGer/Sger_300x10000_(4_4)-12 2.44ms ± 0% 0.85ms ± 0% −65.03%
BlasGer/Sger_300x10000_(4_10)-12 3.64ms ± 0% 0.99ms ± 0% −72.85%
BlasGer/Sger_300x10000_(10_1)-12 2.20ms ± 0% 0.67ms ± 0% −69.41%
BlasGer/Sger_300x10000_(10_2)-12 2.45ms ± 0% 0.72ms ± 0% −70.58%
BlasGer/Sger_300x10000_(10_3)-12 2.49ms ± 0% 0.73ms ± 0% −70.81%
BlasGer/Sger_300x10000_(10_4)-12 2.51ms ± 0% 0.82ms ± 0% −67.21%
BlasGer/Sger_300x10000_(10_10)-12 3.32ms ± 0% 0.99ms ± 0% −70.16%
BlasGer/Sger_1000x30_(1_1)-12 21.1µs ± 0% 4.8µs ± 0% −77.21%
BlasGer/Sger_1000x30_(1_2)-12 35.3µs ± 0% 5.4µs ± 0% −84.71%
BlasGer/Sger_1000x30_(1_3)-12 35.4µs ± 0% 6.1µs ± 0% −82.86%
BlasGer/Sger_1000x30_(1_4)-12 31.1µs ± 0% 5.9µs ± 0% −81.12%
BlasGer/Sger_1000x30_(1_10)-12 32.1µs ± 0% 5.7µs ± 0% −82.31%
BlasGer/Sger_1000x30_(2_1)-12 32.2µs ± 0% 6.0µs ± 0% −81.46%
BlasGer/Sger_1000x30_(2_2)-12 32.1µs ± 0% 4.7µs ± 0% −85.49%
BlasGer/Sger_1000x30_(2_3)-12 32.5µs ± 0% 5.3µs ± 0% −83.79%
BlasGer/Sger_1000x30_(2_4)-12 34.2µs ± 0% 5.0µs ± 0% −85.49%
BlasGer/Sger_1000x30_(2_10)-12 32.3µs ± 0% 5.9µs ± 0% −81.89%
BlasGer/Sger_1000x30_(3_1)-12 34.8µs ± 0% 5.6µs ± 0% −84.01%
BlasGer/Sger_1000x30_(3_2)-12 31.0µs ± 0% 5.8µs ± 0% −81.47%
BlasGer/Sger_1000x30_(3_3)-12 32.5µs ± 0% 5.9µs ± 0% −81.94%
BlasGer/Sger_1000x30_(3_4)-12 38.3µs ± 0% 5.4µs ± 0% −85.94%
BlasGer/Sger_1000x30_(3_10)-12 33.7µs ± 0% 5.6µs ± 0% −83.49%
BlasGer/Sger_1000x30_(4_1)-12 35.4µs ± 0% 5.8µs ± 0% −83.53%
BlasGer/Sger_1000x30_(4_2)-12 35.7µs ± 0% 5.3µs ± 0% −85.18%
BlasGer/Sger_1000x30_(4_3)-12 31.8µs ± 0% 5.5µs ± 0% −82.69%
BlasGer/Sger_1000x30_(4_4)-12 29.4µs ± 0% 5.0µs ± 0% −82.91%
BlasGer/Sger_1000x30_(4_10)-12 27.1µs ± 0% 5.9µs ± 0% −78.12%
BlasGer/Sger_1000x30_(10_1)-12 35.5µs ± 0% 5.8µs ± 0% −83.66%
BlasGer/Sger_1000x30_(10_2)-12 27.8µs ± 0% 6.0µs ± 0% −78.55%
BlasGer/Sger_1000x30_(10_3)-12 32.5µs ± 0% 5.4µs ± 0% −83.37%
BlasGer/Sger_1000x30_(10_4)-12 35.6µs ± 0% 6.4µs ± 0% −81.99%
BlasGer/Sger_1000x30_(10_10)-12 35.4µs ± 0% 6.2µs ± 0% −82.47%
BlasGer/Sger_1000x100_(1_1)-12 27.9µs ± 0% 18.6µs ± 0% −33.45%
BlasGer/Sger_1000x100_(1_2)-12 87.2µs ± 0% 19.9µs ± 0% −77.12%
BlasGer/Sger_1000x100_(1_3)-12 76.2µs ± 0% 23.9µs ± 0% −68.60%
BlasGer/Sger_1000x100_(1_4)-12 86.6µs ± 0% 21.0µs ± 0% −75.78%
BlasGer/Sger_1000x100_(1_10)-12 86.7µs ± 0% 24.3µs ± 0% −72.02%
BlasGer/Sger_1000x100_(2_1)-12 83.7µs ± 0% 20.0µs ± 0% −76.06%
BlasGer/Sger_1000x100_(2_2)-12 86.9µs ± 0% 20.2µs ± 0% −76.73%
BlasGer/Sger_1000x100_(2_3)-12 85.9µs ± 0% 18.0µs ± 0% −78.99%
BlasGer/Sger_1000x100_(2_4)-12 83.9µs ± 0% 20.2µs ± 0% −75.98%
BlasGer/Sger_1000x100_(2_10)-12 87.0µs ± 0% 20.6µs ± 0% −76.28%
BlasGer/Sger_1000x100_(3_1)-12 79.0µs ± 0% 17.9µs ± 0% −77.35%
BlasGer/Sger_1000x100_(3_2)-12 86.8µs ± 0% 19.8µs ± 0% −77.19%
BlasGer/Sger_1000x100_(3_3)-12 84.8µs ± 0% 19.2µs ± 0% −77.38%
BlasGer/Sger_1000x100_(3_4)-12 103µs ± 0% 19µs ± 0% −81.82%
BlasGer/Sger_1000x100_(3_10)-12 86.3µs ± 0% 18.4µs ± 0% −78.73%
BlasGer/Sger_1000x100_(4_1)-12 86.5µs ± 0% 20.6µs ± 0% −76.17%
BlasGer/Sger_1000x100_(4_2)-12 86.7µs ± 0% 19.9µs ± 0% −77.04%
BlasGer/Sger_1000x100_(4_3)-12 107µs ± 0% 21µs ± 0% −80.50%
BlasGer/Sger_1000x100_(4_4)-12 83.5µs ± 0% 20.9µs ± 0% −74.94%
BlasGer/Sger_1000x100_(4_10)-12 79.8µs ± 0% 24.4µs ± 0% −69.45%
BlasGer/Sger_1000x100_(10_1)-12 106µs ± 0% 25µs ± 0% −76.49%
BlasGer/Sger_1000x100_(10_2)-12 73.7µs ± 0% 19.7µs ± 0% −73.20%
BlasGer/Sger_1000x100_(10_3)-12 84.9µs ± 0% 22.2µs ± 0% −73.88%
BlasGer/Sger_1000x100_(10_4)-12 85.8µs ± 0% 20.2µs ± 0% −76.44%
BlasGer/Sger_1000x100_(10_10)-12 88.7µs ± 0% 20.1µs ± 0% −77.29%
BlasGer/Sger_1000x300_(1_1)-12 72.9µs ± 0% 65.2µs ± 0% −10.64%
BlasGer/Sger_1000x300_(1_2)-12 263µs ± 0% 54µs ± 0% −79.35%
BlasGer/Sger_1000x300_(1_3)-12 222µs ± 0% 67µs ± 0% −69.71%
BlasGer/Sger_1000x300_(1_4)-12 267µs ± 0% 66µs ± 0% −75.13%
BlasGer/Sger_1000x300_(1_10)-12 270µs ± 0% 62µs ± 0% −77.00%
BlasGer/Sger_1000x300_(2_1)-12 239µs ± 0% 68µs ± 0% −71.66%
BlasGer/Sger_1000x300_(2_2)-12 265µs ± 0% 68µs ± 0% −74.35%
BlasGer/Sger_1000x300_(2_3)-12 243µs ± 0% 53µs ± 0% −78.37%
BlasGer/Sger_1000x300_(2_4)-12 223µs ± 0% 62µs ± 0% −72.05%
BlasGer/Sger_1000x300_(2_10)-12 255µs ± 0% 65µs ± 0% −74.73%
BlasGer/Sger_1000x300_(3_1)-12 222µs ± 0% 64µs ± 0% −71.00%
BlasGer/Sger_1000x300_(3_2)-12 224µs ± 0% 66µs ± 0% −70.34%
BlasGer/Sger_1000x300_(3_3)-12 236µs ± 0% 56µs ± 0% −76.09%
BlasGer/Sger_1000x300_(3_4)-12 237µs ± 0% 53µs ± 0% −77.45%
BlasGer/Sger_1000x300_(3_10)-12 260µs ± 0% 58µs ± 0% −77.51%
BlasGer/Sger_1000x300_(4_1)-12 296µs ± 0% 54µs ± 0% −81.72%
BlasGer/Sger_1000x300_(4_2)-12 267µs ± 0% 55µs ± 0% −79.44%
BlasGer/Sger_1000x300_(4_3)-12 266µs ± 0% 54µs ± 0% −79.62%
BlasGer/Sger_1000x300_(4_4)-12 266µs ± 0% 73µs ± 0% −72.45%
BlasGer/Sger_1000x300_(4_10)-12 237µs ± 0% 66µs ± 0% −72.37%
BlasGer/Sger_1000x300_(10_1)-12 221µs ± 0% 58µs ± 0% −74.00%
BlasGer/Sger_1000x300_(10_2)-12 222µs ± 0% 65µs ± 0% −70.95%
BlasGer/Sger_1000x300_(10_3)-12 241µs ± 0% 68µs ± 0% −71.75%
BlasGer/Sger_1000x300_(10_4)-12 253µs ± 0% 59µs ± 0% −76.76%
BlasGer/Sger_1000x300_(10_10)-12 265µs ± 0% 64µs ± 0% −75.70%
BlasGer/Sger_1000x1000_(1_1)-12 204µs ± 0% 214µs ± 0% +5.13%
BlasGer/Sger_1000x1000_(1_2)-12 818µs ± 0% 205µs ± 0% −74.93%
BlasGer/Sger_1000x1000_(1_3)-12 819µs ± 0% 211µs ± 0% −74.20%
BlasGer/Sger_1000x1000_(1_4)-12 723µs ± 0% 200µs ± 0% −72.31%
BlasGer/Sger_1000x1000_(1_10)-12 808µs ± 0% 199µs ± 0% −75.33%
BlasGer/Sger_1000x1000_(2_1)-12 791µs ± 0% 231µs ± 0% −70.79%
BlasGer/Sger_1000x1000_(2_2)-12 811µs ± 0% 191µs ± 0% −76.41%
BlasGer/Sger_1000x1000_(2_3)-12 809µs ± 0% 234µs ± 0% −71.02%
BlasGer/Sger_1000x1000_(2_4)-12 797µs ± 0% 237µs ± 0% −70.29%
BlasGer/Sger_1000x1000_(2_10)-12 783µs ± 0% 268µs ± 0% −65.77%
BlasGer/Sger_1000x1000_(3_1)-12 808µs ± 0% 210µs ± 0% −73.98%
BlasGer/Sger_1000x1000_(3_2)-12 810µs ± 0% 194µs ± 0% −76.04%
BlasGer/Sger_1000x1000_(3_3)-12 805µs ± 0% 235µs ± 0% −70.81%
BlasGer/Sger_1000x1000_(3_4)-12 813µs ± 0% 229µs ± 0% −71.80%
BlasGer/Sger_1000x1000_(3_10)-12 812µs ± 0% 224µs ± 0% −72.45%
BlasGer/Sger_1000x1000_(4_1)-12 806µs ± 0% 212µs ± 0% −73.65%
BlasGer/Sger_1000x1000_(4_2)-12 816µs ± 0% 253µs ± 0% −69.02%
BlasGer/Sger_1000x1000_(4_3)-12 766µs ± 0% 222µs ± 0% −71.00%
BlasGer/Sger_1000x1000_(4_4)-12 806µs ± 0% 234µs ± 0% −70.92%
BlasGer/Sger_1000x1000_(4_10)-12 817µs ± 0% 221µs ± 0% −72.94%
BlasGer/Sger_1000x1000_(10_1)-12 712µs ± 0% 220µs ± 0% −69.09%
BlasGer/Sger_1000x1000_(10_2)-12 807µs ± 0% 213µs ± 0% −73.67%
BlasGer/Sger_1000x1000_(10_3)-12 806µs ± 0% 226µs ± 0% −71.91%
BlasGer/Sger_1000x1000_(10_4)-12 810µs ± 0% 250µs ± 0% −69.12%
BlasGer/Sger_1000x1000_(10_10)-12 814µs ± 0% 249µs ± 0% −69.42%
BlasGer/Sger_1000x10000_(1_1)-12 3.98ms ± 0% 3.70ms ± 0% −6.99%
BlasGer/Sger_1000x10000_(1_2)-12 7.46ms ± 0% 3.74ms ± 0% −49.90%
BlasGer/Sger_1000x10000_(1_3)-12 8.16ms ± 0% 3.95ms ± 0% −51.60%
BlasGer/Sger_1000x10000_(1_4)-12 7.64ms ± 0% 3.68ms ± 0% −51.89%
BlasGer/Sger_1000x10000_(1_10)-12 12.5ms ± 0% 4.9ms ± 0% −61.06%
BlasGer/Sger_1000x10000_(2_1)-12 7.69ms ± 0% 3.40ms ± 0% −55.85%
BlasGer/Sger_1000x10000_(2_2)-12 8.02ms ± 0% 3.54ms ± 0% −55.85%
BlasGer/Sger_1000x10000_(2_3)-12 8.01ms ± 0% 3.97ms ± 0% −50.46%
BlasGer/Sger_1000x10000_(2_4)-12 7.67ms ± 0% 3.84ms ± 0% −49.87%
BlasGer/Sger_1000x10000_(2_10)-12 12.8ms ± 0% 4.8ms ± 0% −62.78%
BlasGer/Sger_1000x10000_(3_1)-12 8.13ms ± 0% 3.40ms ± 0% −58.17%
BlasGer/Sger_1000x10000_(3_2)-12 8.03ms ± 0% 3.56ms ± 0% −55.65%
BlasGer/Sger_1000x10000_(3_3)-12 7.98ms ± 0% 3.86ms ± 0% −51.60%
BlasGer/Sger_1000x10000_(3_4)-12 8.28ms ± 0% 4.03ms ± 0% −51.39%
BlasGer/Sger_1000x10000_(3_10)-12 12.8ms ± 0% 4.9ms ± 0% −62.11%
BlasGer/Sger_1000x10000_(4_1)-12 7.90ms ± 0% 3.80ms ± 0% −51.82%
BlasGer/Sger_1000x10000_(4_2)-12 8.17ms ± 0% 3.66ms ± 0% −55.13%
BlasGer/Sger_1000x10000_(4_3)-12 7.51ms ± 0% 3.84ms ± 0% −48.85%
BlasGer/Sger_1000x10000_(4_4)-12 8.08ms ± 0% 3.78ms ± 0% −53.21%
BlasGer/Sger_1000x10000_(4_10)-12 12.9ms ± 0% 4.9ms ± 0% −62.13%
BlasGer/Sger_1000x10000_(10_1)-12 8.03ms ± 0% 3.65ms ± 0% −54.52%
BlasGer/Sger_1000x10000_(10_2)-12 7.91ms ± 0% 3.68ms ± 0% −53.48%
BlasGer/Sger_1000x10000_(10_3)-12 8.03ms ± 0% 3.66ms ± 0% −54.38%
BlasGer/Sger_1000x10000_(10_4)-12 8.37ms ± 0% 3.67ms ± 0% −56.15%
BlasGer/Sger_1000x10000_(10_10)-12 13.1ms ± 0% 4.6ms ± 0% −64.68%
BlasGer/Sger_10000x300_(1_1)-12 733µs ± 0% 789µs ± 0% +7.67%
BlasGer/Sger_10000x300_(1_2)-12 2.56ms ± 0% 1.12ms ± 0% −56.19%
BlasGer/Sger_10000x300_(1_3)-12 2.54ms ± 0% 0.86ms ± 0% −65.97%
BlasGer/Sger_10000x300_(1_4)-12 2.58ms ± 0% 0.85ms ± 0% −67.24%
BlasGer/Sger_10000x300_(1_10)-12 2.43ms ± 0% 0.90ms ± 0% −62.99%
BlasGer/Sger_10000x300_(2_1)-12 2.36ms ± 0% 0.90ms ± 0% −61.79%
BlasGer/Sger_10000x300_(2_2)-12 2.50ms ± 0% 0.74ms ± 0% −70.22%
BlasGer/Sger_10000x300_(2_3)-12 2.28ms ± 0% 0.74ms ± 0% −67.39%
BlasGer/Sger_10000x300_(2_4)-12 2.60ms ± 0% 0.96ms ± 0% −63.22%
BlasGer/Sger_10000x300_(2_10)-12 2.36ms ± 0% 0.93ms ± 0% −60.48%
BlasGer/Sger_10000x300_(3_1)-12 2.23ms ± 0% 1.01ms ± 0% −54.91%
BlasGer/Sger_10000x300_(3_2)-12 2.42ms ± 0% 0.84ms ± 0% −65.47%
BlasGer/Sger_10000x300_(3_3)-12 2.20ms ± 0% 0.96ms ± 0% −56.24%
BlasGer/Sger_10000x300_(3_4)-12 2.56ms ± 0% 0.95ms ± 0% −62.84%
BlasGer/Sger_10000x300_(3_10)-12 2.27ms ± 0% 0.86ms ± 0% −61.95%
BlasGer/Sger_10000x300_(4_1)-12 2.55ms ± 0% 0.86ms ± 0% −66.49%
BlasGer/Sger_10000x300_(4_2)-12 2.40ms ± 0% 0.88ms ± 0% −63.59%
BlasGer/Sger_10000x300_(4_3)-12 2.56ms ± 0% 0.90ms ± 0% −64.69%
BlasGer/Sger_10000x300_(4_4)-12 2.53ms ± 0% 0.85ms ± 0% −66.36%
BlasGer/Sger_10000x300_(4_10)-12 2.26ms ± 0% 0.81ms ± 0% −64.04%
BlasGer/Sger_10000x300_(10_1)-12 2.90ms ± 0% 0.96ms ± 0% −66.85%
BlasGer/Sger_10000x300_(10_2)-12 2.57ms ± 0% 0.98ms ± 0% −61.82%
BlasGer/Sger_10000x300_(10_3)-12 2.45ms ± 0% 0.96ms ± 0% −60.93%
BlasGer/Sger_10000x300_(10_4)-12 2.63ms ± 0% 1.04ms ± 0% −60.26%
BlasGer/Sger_10000x300_(10_10)-12 2.56ms ± 0% 0.80ms ± 0% −68.83%
BlasGer/Sger_10000x1000_(1_1)-12 3.41ms ± 0% 3.90ms ± 0% +14.31%
BlasGer/Sger_10000x1000_(1_2)-12 8.10ms ± 0% 3.75ms ± 0% −53.72%
BlasGer/Sger_10000x1000_(1_3)-12 8.19ms ± 0% 3.72ms ± 0% −54.59%
BlasGer/Sger_10000x1000_(1_4)-12 8.13ms ± 0% 3.59ms ± 0% −55.89%
BlasGer/Sger_10000x1000_(1_10)-12 8.23ms ± 0% 4.11ms ± 0% −50.06%
BlasGer/Sger_10000x1000_(2_1)-12 8.11ms ± 0% 3.86ms ± 0% −52.40%
BlasGer/Sger_10000x1000_(2_2)-12 8.05ms ± 0% 3.92ms ± 0% −51.27%
BlasGer/Sger_10000x1000_(2_3)-12 7.57ms ± 0% 3.97ms ± 0% −47.55%
BlasGer/Sger_10000x1000_(2_4)-12 7.56ms ± 0% 4.00ms ± 0% −47.09%
BlasGer/Sger_10000x1000_(2_10)-12 8.24ms ± 0% 3.93ms ± 0% −52.32%
BlasGer/Sger_10000x1000_(3_1)-12 8.31ms ± 0% 3.65ms ± 0% −56.02%
BlasGer/Sger_10000x1000_(3_2)-12 8.15ms ± 0% 3.56ms ± 0% −56.38%
BlasGer/Sger_10000x1000_(3_3)-12 8.36ms ± 0% 3.99ms ± 0% −52.21%
BlasGer/Sger_10000x1000_(3_4)-12 8.34ms ± 0% 3.88ms ± 0% −53.42%
BlasGer/Sger_10000x1000_(3_10)-12 8.33ms ± 0% 3.68ms ± 0% −55.78%
BlasGer/Sger_10000x1000_(4_1)-12 8.32ms ± 0% 3.94ms ± 0% −52.65%
BlasGer/Sger_10000x1000_(4_2)-12 8.13ms ± 0% 3.58ms ± 0% −55.96%
BlasGer/Sger_10000x1000_(4_3)-12 8.21ms ± 0% 3.93ms ± 0% −52.15%
BlasGer/Sger_10000x1000_(4_4)-12 7.59ms ± 0% 3.85ms ± 0% −49.28%
BlasGer/Sger_10000x1000_(4_10)-12 8.07ms ± 0% 3.69ms ± 0% −54.20%
BlasGer/Sger_10000x1000_(10_1)-12 7.57ms ± 0% 3.69ms ± 0% −51.25%
BlasGer/Sger_10000x1000_(10_2)-12 8.10ms ± 0% 3.91ms ± 0% −51.79%
BlasGer/Sger_10000x1000_(10_3)-12 8.11ms ± 0% 3.62ms ± 0% −55.37%
BlasGer/Sger_10000x1000_(10_4)-12 8.18ms ± 0% 3.97ms ± 0% −51.49%
BlasGer/Sger_10000x1000_(10_10)-12 8.19ms ± 0% 4.10ms ± 0% −49.96%
BlasGer/Sger_10000x10000_(1_1)-12 35.0ms ± 0% 34.6ms ± 0% −1.19%
BlasGer/Sger_10000x10000_(1_2)-12 74.7ms ± 0% 36.5ms ± 0% −51.08%
BlasGer/Sger_10000x10000_(1_3)-12 75.3ms ± 0% 36.1ms ± 0% −52.08%
BlasGer/Sger_10000x10000_(1_4)-12 79.4ms ± 0% 39.5ms ± 0% −50.22%
BlasGer/Sger_10000x10000_(1_10)-12 121ms ± 0% 45ms ± 0% −62.88%
BlasGer/Sger_10000x10000_(2_1)-12 75.0ms ± 0% 34.4ms ± 0% −54.11%
BlasGer/Sger_10000x10000_(2_2)-12 86.7ms ± 0% 36.0ms ± 0% −58.51%
BlasGer/Sger_10000x10000_(2_3)-12 75.0ms ± 0% 36.1ms ± 0% −51.90%
BlasGer/Sger_10000x10000_(2_4)-12 75.4ms ± 0% 37.3ms ± 0% −50.48%
BlasGer/Sger_10000x10000_(2_10)-12 120ms ± 0% 45ms ± 0% −62.29%
BlasGer/Sger_10000x10000_(3_1)-12 74.7ms ± 0% 34.4ms ± 0% −53.90%
BlasGer/Sger_10000x10000_(3_2)-12 76.9ms ± 0% 35.1ms ± 0% −54.33%
BlasGer/Sger_10000x10000_(3_3)-12 74.8ms ± 0% 35.6ms ± 0% −52.36%
BlasGer/Sger_10000x10000_(3_4)-12 76.5ms ± 0% 37.3ms ± 0% −51.32%
BlasGer/Sger_10000x10000_(3_10)-12 122ms ± 0% 45ms ± 0% −62.82%
BlasGer/Sger_10000x10000_(4_1)-12 75.1ms ± 0% 34.2ms ± 0% −54.41%
BlasGer/Sger_10000x10000_(4_2)-12 74.8ms ± 0% 36.1ms ± 0% −51.79%
BlasGer/Sger_10000x10000_(4_3)-12 75.8ms ± 0% 36.0ms ± 0% −52.47%
BlasGer/Sger_10000x10000_(4_4)-12 75.9ms ± 0% 36.5ms ± 0% −51.92%
BlasGer/Sger_10000x10000_(4_10)-12 118ms ± 0% 45ms ± 0% −62.17%
BlasGer/Sger_10000x10000_(10_1)-12 74.6ms ± 0% 34.2ms ± 0% −54.18%
BlasGer/Sger_10000x10000_(10_2)-12 75.1ms ± 0% 35.0ms ± 0% −53.39%
BlasGer/Sger_10000x10000_(10_3)-12 74.8ms ± 0% 35.7ms ± 0% −52.24%
BlasGer/Sger_10000x10000_(10_4)-12 77.5ms ± 0% 36.9ms ± 0% −52.46%
BlasGer/Sger_10000x10000_(10_10)-12 122ms ± 0% 45ms ± 0% −63.39%
BlasGer/Sger_10000x100000_(1_1)-12 426ms ± 0% 350ms ± 0% −17.84%
BlasGer/Sger_10000x100000_(1_2)-12 748ms ± 0% 356ms ± 0% −52.37%
BlasGer/Sger_10000x100000_(1_3)-12 765ms ± 0% 374ms ± 0% −51.07%
BlasGer/Sger_10000x100000_(1_4)-12 833ms ± 0% 374ms ± 0% −55.12%
BlasGer/Sger_10000x100000_(1_10)-12 1.22s ± 0% 0.45s ± 0% −63.30%
BlasGer/Sger_10000x100000_(2_1)-12 744ms ± 0% 346ms ± 0% −53.50%
BlasGer/Sger_10000x100000_(2_2)-12 749ms ± 0% 356ms ± 0% −52.44%
BlasGer/Sger_10000x100000_(2_3)-12 762ms ± 0% 368ms ± 0% −51.72%
BlasGer/Sger_10000x100000_(2_4)-12 831ms ± 0% 375ms ± 0% −54.83%
BlasGer/Sger_10000x100000_(2_10)-12 1.22s ± 0% 0.46s ± 0% −62.73%
BlasGer/Sger_10000x100000_(3_1)-12 744ms ± 0% 348ms ± 0% −53.20%
BlasGer/Sger_10000x100000_(3_2)-12 747ms ± 0% 355ms ± 0% −52.52%
BlasGer/Sger_10000x100000_(3_3)-12 764ms ± 0% 370ms ± 0% −51.54%
BlasGer/Sger_10000x100000_(3_4)-12 840ms ± 0% 374ms ± 0% −55.49%
BlasGer/Sger_10000x100000_(3_10)-12 1.23s ± 0% 0.45s ± 0% −63.53%
BlasGer/Sger_10000x100000_(4_1)-12 757ms ± 0% 344ms ± 0% −54.50%
BlasGer/Sger_10000x100000_(4_2)-12 749ms ± 0% 364ms ± 0% −51.44%
BlasGer/Sger_10000x100000_(4_3)-12 766ms ± 0% 373ms ± 0% −51.30%
BlasGer/Sger_10000x100000_(4_4)-12 833ms ± 0% 376ms ± 0% −54.81%
BlasGer/Sger_10000x100000_(4_10)-12 1.22s ± 0% 0.45s ± 0% −62.99%
BlasGer/Sger_10000x100000_(10_1)-12 747ms ± 0% 344ms ± 0% −53.92%
BlasGer/Sger_10000x100000_(10_2)-12 759ms ± 0% 357ms ± 0% −52.94%
BlasGer/Sger_10000x100000_(10_3)-12 777ms ± 0% 368ms ± 0% −52.67%
BlasGer/Sger_10000x100000_(10_4)-12 837ms ± 0% 375ms ± 0% −55.24%
BlasGer/Sger_10000x100000_(10_10)-12 1.22s ± 0% 0.46s ± 0% −62.81%
BlasGer/Sger_100000x10000_(1_1)-12 349ms ± 0% 347ms ± 0% −0.44%
BlasGer/Sger_100000x10000_(1_2)-12 750ms ± 0% 349ms ± 0% −53.44%
BlasGer/Sger_100000x10000_(1_3)-12 752ms ± 0% 358ms ± 0% −52.40%
BlasGer/Sger_100000x10000_(1_4)-12 786ms ± 0% 374ms ± 0% −52.38%
BlasGer/Sger_100000x10000_(1_10)-12 1.22s ± 0% 0.45s ± 0% −62.98%
BlasGer/Sger_100000x10000_(2_1)-12 752ms ± 0% 344ms ± 0% −54.20%
BlasGer/Sger_100000x10000_(2_2)-12 753ms ± 0% 348ms ± 0% −53.70%
BlasGer/Sger_100000x10000_(2_3)-12 753ms ± 0% 363ms ± 0% −51.79%
BlasGer/Sger_100000x10000_(2_4)-12 752ms ± 0% 363ms ± 0% −51.69%
BlasGer/Sger_100000x10000_(2_10)-12 1.22s ± 0% 0.45s ± 0% −62.90%
BlasGer/Sger_100000x10000_(3_1)-12 747ms ± 0% 373ms ± 0% −50.15%
BlasGer/Sger_100000x10000_(3_2)-12 747ms ± 0% 353ms ± 0% −52.78%
BlasGer/Sger_100000x10000_(3_3)-12 753ms ± 0% 358ms ± 0% −52.50%
BlasGer/Sger_100000x10000_(3_4)-12 810ms ± 0% 366ms ± 0% −54.83%
BlasGer/Sger_100000x10000_(3_10)-12 1.20s ± 0% 0.45s ± 0% −62.52%
BlasGer/Sger_100000x10000_(4_1)-12 750ms ± 0% 341ms ± 0% −54.51%
BlasGer/Sger_100000x10000_(4_2)-12 749ms ± 0% 354ms ± 0% −52.71%
BlasGer/Sger_100000x10000_(4_3)-12 754ms ± 0% 367ms ± 0% −51.34%
BlasGer/Sger_100000x10000_(4_4)-12 772ms ± 0% 364ms ± 0% −52.82%
BlasGer/Sger_100000x10000_(4_10)-12 1.22s ± 0% 0.45s ± 0% −62.95%
BlasGer/Sger_100000x10000_(10_1)-12 746ms ± 0% 345ms ± 0% −53.79%
BlasGer/Sger_100000x10000_(10_2)-12 750ms ± 0% 351ms ± 0% −53.26%
BlasGer/Sger_100000x10000_(10_3)-12 799ms ± 0% 362ms ± 0% −54.68%
BlasGer/Sger_100000x10000_(10_4)-12 783ms ± 0% 364ms ± 0% −53.47%
BlasGer/Sger_100000x10000_(10_10)-12 1.23s ± 0% 0.45s ± 0% −63.46%
 

Blas incorporation is single-threaded (not tiled), so performance gains are still available.

Benchmark comparison to OpenBLAS
New Kernel OpenBLAS
time/op delta
BlasGer/Sger_3x3_(1_1)-12 11.5ns ± 0% 99.8ns ± 0% +767.83%
BlasGer/Sger_3x3_(1_2)-12 15.5ns ± 0% 120.0ns ± 0% +674.19%
BlasGer/Sger_3x3_(1_3)-12 14.2ns ± 0% 117.0ns ± 0% +723.94%
BlasGer/Sger_3x3_(1_4)-12 15.1ns ± 0% 107.0ns ± 0% +608.61%
BlasGer/Sger_3x3_(1_10)-12 15.2ns ± 0% 115.0ns ± 0% +656.58%
BlasGer/Sger_3x3_(2_1)-12 15.5ns ± 0% 117.0ns ± 0% +654.84%
BlasGer/Sger_3x3_(2_2)-12 14.0ns ± 0% 103.0ns ± 0% +635.71%
BlasGer/Sger_3x3_(2_3)-12 15.5ns ± 0% 99.3ns ± 0% +540.65%
BlasGer/Sger_3x3_(2_4)-12 15.2ns ± 0% 118.0ns ± 0% +676.32%
BlasGer/Sger_3x3_(2_10)-12 13.4ns ± 0% 113.0ns ± 0% +743.28%
BlasGer/Sger_3x3_(3_1)-12 16.1ns ± 0% 108.0ns ± 0% +570.81%
BlasGer/Sger_3x3_(3_2)-12 15.5ns ± 0% 116.0ns ± 0% +648.39%
BlasGer/Sger_3x3_(3_3)-12 15.1ns ± 0% 117.0ns ± 0% +674.83%
BlasGer/Sger_3x3_(3_4)-12 14.1ns ± 0% 90.8ns ± 0% +543.97%
BlasGer/Sger_3x3_(3_10)-12 15.6ns ± 0% 103.0ns ± 0% +560.26%
BlasGer/Sger_3x3_(4_1)-12 15.6ns ± 0% 98.3ns ± 0% +530.13%
BlasGer/Sger_3x3_(4_2)-12 15.6ns ± 0% 127.0ns ± 0% +714.10%
BlasGer/Sger_3x3_(4_3)-12 15.6ns ± 0% 104.0ns ± 0% +566.67%
BlasGer/Sger_3x3_(4_4)-12 13.3ns ± 0% 98.2ns ± 0% +638.35%
BlasGer/Sger_3x3_(4_10)-12 15.3ns ± 0% 103.0ns ± 0% +573.20%
BlasGer/Sger_3x3_(10_1)-12 14.0ns ± 0% 99.0ns ± 0% +607.14%
BlasGer/Sger_3x3_(10_2)-12 13.2ns ± 0% 108.0ns ± 0% +718.18%
BlasGer/Sger_3x3_(10_3)-12 15.2ns ± 0% 114.0ns ± 0% +650.00%
BlasGer/Sger_3x3_(10_4)-12 14.2ns ± 0% 117.0ns ± 0% +723.94%
BlasGer/Sger_3x3_(10_10)-12 13.7ns ± 0% 105.0ns ± 0% +666.42%
BlasGer/Sger_3x10_(1_1)-12 15.2ns ± 0% 124.0ns ± 0% +715.79%
BlasGer/Sger_3x10_(1_2)-12 18.3ns ± 0% 150.0ns ± 0% +719.67%
BlasGer/Sger_3x10_(1_3)-12 18.1ns ± 0% 118.0ns ± 0% +551.93%
BlasGer/Sger_3x10_(1_4)-12 21.6ns ± 0% 127.0ns ± 0% +487.96%
BlasGer/Sger_3x10_(1_10)-12 15.6ns ± 0% 118.0ns ± 0% +656.41%
BlasGer/Sger_3x10_(2_1)-12 16.1ns ± 0% 124.0ns ± 0% +670.19%
BlasGer/Sger_3x10_(2_2)-12 16.7ns ± 0% 102.0ns ± 0% +510.78%
BlasGer/Sger_3x10_(2_3)-12 18.5ns ± 0% 129.0ns ± 0% +597.30%
BlasGer/Sger_3x10_(2_4)-12 17.3ns ± 0% 119.0ns ± 0% +587.86%
BlasGer/Sger_3x10_(2_10)-12 15.5ns ± 0% 116.0ns ± 0% +648.39%
BlasGer/Sger_3x10_(3_1)-12 20.2ns ± 0% 115.0ns ± 0% +469.31%
BlasGer/Sger_3x10_(3_2)-12 16.6ns ± 0% 112.0ns ± 0% +574.70%
BlasGer/Sger_3x10_(3_3)-12 18.1ns ± 0% 113.0ns ± 0% +524.31%
BlasGer/Sger_3x10_(3_4)-12 17.4ns ± 0% 126.0ns ± 0% +624.14%
BlasGer/Sger_3x10_(3_10)-12 16.0ns ± 0% 103.0ns ± 0% +543.75%
BlasGer/Sger_3x10_(4_1)-12 17.9ns ± 0% 117.0ns ± 0% +553.63%
BlasGer/Sger_3x10_(4_2)-12 17.9ns ± 0% 129.0ns ± 0% +620.67%
BlasGer/Sger_3x10_(4_3)-12 17.7ns ± 0% 120.0ns ± 0% +577.97%
BlasGer/Sger_3x10_(4_4)-12 17.2ns ± 0% 130.0ns ± 0% +655.81%
BlasGer/Sger_3x10_(4_10)-12 18.0ns ± 0% 107.0ns ± 0% +494.44%
BlasGer/Sger_3x10_(10_1)-12 16.5ns ± 0% 111.0ns ± 0% +572.73%
BlasGer/Sger_3x10_(10_2)-12 16.9ns ± 0% 124.0ns ± 0% +633.73%
BlasGer/Sger_3x10_(10_3)-12 18.0ns ± 0% 120.0ns ± 0% +566.67%
BlasGer/Sger_3x10_(10_4)-12 17.4ns ± 0% 127.0ns ± 0% +629.89%
BlasGer/Sger_3x10_(10_10)-12 15.5ns ± 0% 117.0ns ± 0% +654.84%
BlasGer/Sger_3x30_(1_1)-12 21.8ns ± 0% 131.0ns ± 0% +500.92%
BlasGer/Sger_3x30_(1_2)-12 26.6ns ± 0% 175.0ns ± 0% +557.89%
BlasGer/Sger_3x30_(1_3)-12 31.4ns ± 0% 170.0ns ± 0% +441.40%
BlasGer/Sger_3x30_(1_4)-12 33.2ns ± 0% 154.0ns ± 0% +363.86%
BlasGer/Sger_3x30_(1_10)-12 31.9ns ± 0% 150.0ns ± 0% +370.22%
BlasGer/Sger_3x30_(2_1)-12 25.1ns ± 0% 127.0ns ± 0% +405.98%
BlasGer/Sger_3x30_(2_2)-12 28.5ns ± 0% 159.0ns ± 0% +457.89%
BlasGer/Sger_3x30_(2_3)-12 31.6ns ± 0% 156.0ns ± 0% +393.67%
BlasGer/Sger_3x30_(2_4)-12 33.9ns ± 0% 145.0ns ± 0% +327.73%
BlasGer/Sger_3x30_(2_10)-12 30.9ns ± 0% 170.0ns ± 0% +450.16%
BlasGer/Sger_3x30_(3_1)-12 25.7ns ± 0% 157.0ns ± 0% +510.89%
BlasGer/Sger_3x30_(3_2)-12 26.1ns ± 0% 160.0ns ± 0% +513.03%
BlasGer/Sger_3x30_(3_3)-12 27.7ns ± 0% 148.0ns ± 0% +434.30%
BlasGer/Sger_3x30_(3_4)-12 31.5ns ± 0% 167.0ns ± 0% +430.16%
BlasGer/Sger_3x30_(3_10)-12 34.0ns ± 0% 177.0ns ± 0% +420.59%
BlasGer/Sger_3x30_(4_1)-12 30.6ns ± 0% 133.0ns ± 0% +334.64%
BlasGer/Sger_3x30_(4_2)-12 30.7ns ± 0% 159.0ns ± 0% +417.92%
BlasGer/Sger_3x30_(4_3)-12 29.9ns ± 0% 154.0ns ± 0% +415.05%
BlasGer/Sger_3x30_(4_4)-12 33.0ns ± 0% 166.0ns ± 0% +403.03%
BlasGer/Sger_3x30_(4_10)-12 30.2ns ± 0% 167.0ns ± 0% +452.98%
BlasGer/Sger_3x30_(10_1)-12 29.4ns ± 0% 144.0ns ± 0% +389.80%
BlasGer/Sger_3x30_(10_2)-12 29.5ns ± 0% 188.0ns ± 0% +537.29%
BlasGer/Sger_3x30_(10_3)-12 34.3ns ± 0% 165.0ns ± 0% +381.05%
BlasGer/Sger_3x30_(10_4)-12 33.4ns ± 0% 172.0ns ± 0% +414.97%
BlasGer/Sger_3x30_(10_10)-12 29.3ns ± 0% 159.0ns ± 0% +442.66%
BlasGer/Sger_3x100_(1_1)-12 52.6ns ± 0% 132.0ns ± 0% +150.95%
BlasGer/Sger_3x100_(1_2)-12 62.6ns ± 0% 168.0ns ± 0% +168.37%
BlasGer/Sger_3x100_(1_3)-12 59.7ns ± 0% 176.0ns ± 0% +194.81%
BlasGer/Sger_3x100_(1_4)-12 65.2ns ± 0% 161.0ns ± 0% +146.93%
BlasGer/Sger_3x100_(1_10)-12 67.5ns ± 0% 144.0ns ± 0% +113.33%
BlasGer/Sger_3x100_(2_1)-12 64.0ns ± 0% 123.0ns ± 0% +92.19%
BlasGer/Sger_3x100_(2_2)-12 68.5ns ± 0% 170.0ns ± 0% +148.18%
BlasGer/Sger_3x100_(2_3)-12 66.5ns ± 0% 162.0ns ± 0% +143.61%
BlasGer/Sger_3x100_(2_4)-12 68.1ns ± 0% 165.0ns ± 0% +142.29%
BlasGer/Sger_3x100_(2_10)-12 71.8ns ± 0% 165.0ns ± 0% +129.81%
BlasGer/Sger_3x100_(3_1)-12 68.7ns ± 0% 119.0ns ± 0% +73.22%
BlasGer/Sger_3x100_(3_2)-12 67.4ns ± 0% 172.0ns ± 0% +155.19%
BlasGer/Sger_3x100_(3_3)-12 69.3ns ± 0% 159.0ns ± 0% +129.44%
BlasGer/Sger_3x100_(3_4)-12 66.5ns ± 0% 159.0ns ± 0% +139.10%
BlasGer/Sger_3x100_(3_10)-12 73.4ns ± 0% 162.0ns ± 0% +120.71%
BlasGer/Sger_3x100_(4_1)-12 64.5ns ± 0% 130.0ns ± 0% +101.55%
BlasGer/Sger_3x100_(4_2)-12 70.1ns ± 0% 140.0ns ± 0% +99.71%
BlasGer/Sger_3x100_(4_3)-12 71.9ns ± 0% 165.0ns ± 0% +129.49%
BlasGer/Sger_3x100_(4_4)-12 75.9ns ± 0% 167.0ns ± 0% +120.03%
BlasGer/Sger_3x100_(4_10)-12 75.0ns ± 0% 161.0ns ± 0% +114.67%
BlasGer/Sger_3x100_(10_1)-12 74.6ns ± 0% 127.0ns ± 0% +70.24%
BlasGer/Sger_3x100_(10_2)-12 72.9ns ± 0% 161.0ns ± 0% +120.85%
BlasGer/Sger_3x100_(10_3)-12 72.6ns ± 0% 173.0ns ± 0% +138.29%
BlasGer/Sger_3x100_(10_4)-12 72.8ns ± 0% 143.0ns ± 0% +96.43%
BlasGer/Sger_3x100_(10_10)-12 70.0ns ± 0% 153.0ns ± 0% +118.57%
BlasGer/Sger_10x3_(1_1)-12 20.4ns ± 0% 143.0ns ± 0% +600.98%
BlasGer/Sger_10x3_(1_2)-12 22.6ns ± 0% 157.0ns ± 0% +594.69%
BlasGer/Sger_10x3_(1_3)-12 20.9ns ± 0% 159.0ns ± 0% +660.77%
BlasGer/Sger_10x3_(1_4)-12 30.6ns ± 0% 135.0ns ± 0% +341.18%
BlasGer/Sger_10x3_(1_10)-12 26.0ns ± 0% 152.0ns ± 0% +484.62%
BlasGer/Sger_10x3_(2_1)-12 21.6ns ± 0% 149.0ns ± 0% +589.81%
BlasGer/Sger_10x3_(2_2)-12 20.7ns ± 0% 144.0ns ± 0% +595.65%
BlasGer/Sger_10x3_(2_3)-12 20.6ns ± 0% 131.0ns ± 0% +535.92%
BlasGer/Sger_10x3_(2_4)-12 22.8ns ± 0% 149.0ns ± 0% +553.51%
BlasGer/Sger_10x3_(2_10)-12 26.2ns ± 0% 144.0ns ± 0% +449.62%
BlasGer/Sger_10x3_(3_1)-12 26.1ns ± 0% 140.0ns ± 0% +436.40%
BlasGer/Sger_10x3_(3_2)-12 25.7ns ± 0% 137.0ns ± 0% +433.07%
BlasGer/Sger_10x3_(3_3)-12 25.4ns ± 0% 154.0ns ± 0% +506.30%
BlasGer/Sger_10x3_(3_4)-12 25.8ns ± 0% 153.0ns ± 0% +493.02%
BlasGer/Sger_10x3_(3_10)-12 23.9ns ± 0% 133.0ns ± 0% +456.49%
BlasGer/Sger_10x3_(4_1)-12 25.2ns ± 0% 127.0ns ± 0% +403.97%
BlasGer/Sger_10x3_(4_2)-12 26.0ns ± 0% 154.0ns ± 0% +492.31%
BlasGer/Sger_10x3_(4_3)-12 25.8ns ± 0% 143.0ns ± 0% +454.26%
BlasGer/Sger_10x3_(4_4)-12 21.3ns ± 0% 151.0ns ± 0% +608.92%
BlasGer/Sger_10x3_(4_10)-12 23.0ns ± 0% 149.0ns ± 0% +547.83%
BlasGer/Sger_10x3_(10_1)-12 23.4ns ± 0% 149.0ns ± 0% +536.75%
BlasGer/Sger_10x3_(10_2)-12 23.0ns ± 0% 159.0ns ± 0% +591.30%
BlasGer/Sger_10x3_(10_3)-12 25.6ns ± 0% 131.0ns ± 0% +411.72%
BlasGer/Sger_10x3_(10_4)-12 26.2ns ± 0% 158.0ns ± 0% +503.05%
BlasGer/Sger_10x3_(10_10)-12 25.6ns ± 0% 149.0ns ± 0% +482.03%
BlasGer/Sger_10x10_(1_1)-12 25.2ns ± 0% 226.0ns ± 0% +796.83%
BlasGer/Sger_10x10_(1_2)-12 34.3ns ± 0% 187.0ns ± 0% +445.19%
BlasGer/Sger_10x10_(1_3)-12 33.0ns ± 0% 192.0ns ± 0% +481.82%
BlasGer/Sger_10x10_(1_4)-12 33.2ns ± 0% 174.0ns ± 0% +424.10%
BlasGer/Sger_10x10_(1_10)-12 31.3ns ± 0% 194.0ns ± 0% +519.81%
BlasGer/Sger_10x10_(2_1)-12 37.8ns ± 0% 208.0ns ± 0% +450.26%
BlasGer/Sger_10x10_(2_2)-12 29.5ns ± 0% 225.0ns ± 0% +662.71%
BlasGer/Sger_10x10_(2_3)-12 33.5ns ± 0% 169.0ns ± 0% +404.48%
BlasGer/Sger_10x10_(2_4)-12 33.4ns ± 0% 186.0ns ± 0% +456.89%
BlasGer/Sger_10x10_(2_10)-12 29.7ns ± 0% 223.0ns ± 0% +650.84%
BlasGer/Sger_10x10_(3_1)-12 34.1ns ± 0% 213.0ns ± 0% +524.63%
BlasGer/Sger_10x10_(3_2)-12 28.4ns ± 0% 181.0ns ± 0% +537.32%
BlasGer/Sger_10x10_(3_3)-12 30.5ns ± 0% 168.0ns ± 0% +450.82%
BlasGer/Sger_10x10_(3_4)-12 28.7ns ± 0% 212.0ns ± 0% +638.68%
BlasGer/Sger_10x10_(3_10)-12 35.4ns ± 0% 190.0ns ± 0% +436.72%
BlasGer/Sger_10x10_(4_1)-12 29.5ns ± 0% 176.0ns ± 0% +496.61%
BlasGer/Sger_10x10_(4_2)-12 31.6ns ± 0% 170.0ns ± 0% +437.97%
BlasGer/Sger_10x10_(4_3)-12 32.3ns ± 0% 194.0ns ± 0% +500.62%
BlasGer/Sger_10x10_(4_4)-12 33.7ns ± 0% 183.0ns ± 0% +443.03%
BlasGer/Sger_10x10_(4_10)-12 30.3ns ± 0% 195.0ns ± 0% +543.56%
BlasGer/Sger_10x10_(10_1)-12 29.5ns ± 0% 215.0ns ± 0% +628.81%
BlasGer/Sger_10x10_(10_2)-12 33.7ns ± 0% 219.0ns ± 0% +549.85%
BlasGer/Sger_10x10_(10_3)-12 28.3ns ± 0% 193.0ns ± 0% +581.98%
BlasGer/Sger_10x10_(10_4)-12 31.2ns ± 0% 199.0ns ± 0% +537.82%
BlasGer/Sger_10x10_(10_10)-12 33.4ns ± 0% 224.0ns ± 0% +570.66%
BlasGer/Sger_10x30_(1_1)-12 50.5ns ± 0% 301.0ns ± 0% +496.04%
BlasGer/Sger_10x30_(1_2)-12 53.7ns ± 0% 329.0ns ± 0% +512.66%
BlasGer/Sger_10x30_(1_3)-12 52.9ns ± 0% 351.0ns ± 0% +563.52%
BlasGer/Sger_10x30_(1_4)-12 65.4ns ± 0% 301.0ns ± 0% +360.24%
BlasGer/Sger_10x30_(1_10)-12 67.0ns ± 0% 361.0ns ± 0% +438.81%
BlasGer/Sger_10x30_(2_1)-12 63.2ns ± 0% 314.0ns ± 0% +396.84%
BlasGer/Sger_10x30_(2_2)-12 52.2ns ± 0% 297.0ns ± 0% +468.97%
BlasGer/Sger_10x30_(2_3)-12 65.5ns ± 0% 361.0ns ± 0% +451.15%
BlasGer/Sger_10x30_(2_4)-12 60.7ns ± 0% 353.0ns ± 0% +481.55%
BlasGer/Sger_10x30_(2_10)-12 53.9ns ± 0% 314.0ns ± 0% +482.56%
BlasGer/Sger_10x30_(3_1)-12 60.2ns ± 0% 312.0ns ± 0% +418.27%
BlasGer/Sger_10x30_(3_2)-12 64.2ns ± 0% 378.0ns ± 0% +488.79%
BlasGer/Sger_10x30_(3_3)-12 59.7ns ± 0% 328.0ns ± 0% +449.41%
BlasGer/Sger_10x30_(3_4)-12 66.4ns ± 0% 298.0ns ± 0% +348.80%
BlasGer/Sger_10x30_(3_10)-12 67.4ns ± 0% 300.0ns ± 0% +345.10%
BlasGer/Sger_10x30_(4_1)-12 60.7ns ± 0% 314.0ns ± 0% +417.30%
BlasGer/Sger_10x30_(4_2)-12 56.6ns ± 0% 382.0ns ± 0% +574.91%
BlasGer/Sger_10x30_(4_3)-12 66.4ns ± 0% 360.0ns ± 0% +442.17%
BlasGer/Sger_10x30_(4_4)-12 67.0ns ± 0% 309.0ns ± 0% +361.19%
BlasGer/Sger_10x30_(4_10)-12 68.0ns ± 0% 366.0ns ± 0% +438.24%
BlasGer/Sger_10x30_(10_1)-12 66.6ns ± 0% 340.0ns ± 0% +410.51%
BlasGer/Sger_10x30_(10_2)-12 53.9ns ± 0% 291.0ns ± 0% +439.89%
BlasGer/Sger_10x30_(10_3)-12 57.5ns ± 0% 281.0ns ± 0% +388.70%
BlasGer/Sger_10x30_(10_4)-12 60.3ns ± 0% 352.0ns ± 0% +483.75%
BlasGer/Sger_10x30_(10_10)-12 61.0ns ± 0% 281.0ns ± 0% +360.66%
BlasGer/Sger_10x100_(1_1)-12 127ns ± 0% 262ns ± 0% +106.30%
BlasGer/Sger_10x100_(1_2)-12 166ns ± 0% 277ns ± 0% +66.87%
BlasGer/Sger_10x100_(1_3)-12 177ns ± 0% 243ns ± 0% +37.29%
BlasGer/Sger_10x100_(1_4)-12 168ns ± 0% 266ns ± 0% +58.33%
BlasGer/Sger_10x100_(1_10)-12 144ns ± 0% 269ns ± 0% +86.81%
BlasGer/Sger_10x100_(2_1)-12 168ns ± 0% 237ns ± 0% +41.07%
BlasGer/Sger_10x100_(2_2)-12 164ns ± 0% 272ns ± 0% +65.85%
BlasGer/Sger_10x100_(2_3)-12 163ns ± 0% 268ns ± 0% +64.42%
BlasGer/Sger_10x100_(2_4)-12 169ns ± 0% 237ns ± 0% +40.24%
BlasGer/Sger_10x100_(2_10)-12 156ns ± 0% 272ns ± 0% +74.36%
BlasGer/Sger_10x100_(3_1)-12 160ns ± 0% 200ns ± 0% +25.00%
BlasGer/Sger_10x100_(3_2)-12 167ns ± 0% 245ns ± 0% +46.71%
BlasGer/Sger_10x100_(3_3)-12 166ns ± 0% 217ns ± 0% +30.72%
BlasGer/Sger_10x100_(3_4)-12 166ns ± 0% 270ns ± 0% +62.65%
BlasGer/Sger_10x100_(3_10)-12 154ns ± 0% 266ns ± 0% +72.73%
BlasGer/Sger_10x100_(4_1)-12 148ns ± 0% 224ns ± 0% +51.35%
BlasGer/Sger_10x100_(4_2)-12 164ns ± 0% 237ns ± 0% +44.51%
BlasGer/Sger_10x100_(4_3)-12 160ns ± 0% 260ns ± 0% +62.50%
BlasGer/Sger_10x100_(4_4)-12 146ns ± 0% 267ns ± 0% +82.88%
BlasGer/Sger_10x100_(4_10)-12 144ns ± 0% 274ns ± 0% +90.28%
BlasGer/Sger_10x100_(10_1)-12 168ns ± 0% 207ns ± 0% +23.21%
BlasGer/Sger_10x100_(10_2)-12 149ns ± 0% 268ns ± 0% +79.87%
BlasGer/Sger_10x100_(10_3)-12 149ns ± 0% 247ns ± 0% +65.77%
BlasGer/Sger_10x100_(10_4)-12 166ns ± 0% 239ns ± 0% +43.98%
BlasGer/Sger_10x100_(10_10)-12 172ns ± 0% 275ns ± 0% +59.88%
BlasGer/Sger_10x300_(1_1)-12 392ns ± 0% 374ns ± 0% −4.59%
BlasGer/Sger_10x300_(1_2)-12 434ns ± 0% 516ns ± 0% +18.89%
BlasGer/Sger_10x300_(1_3)-12 453ns ± 0% 492ns ± 0% +8.61%
BlasGer/Sger_10x300_(1_4)-12 547ns ± 0% 515ns ± 0% −5.85%
BlasGer/Sger_10x300_(1_10)-12 428ns ± 0% 534ns ± 0% +24.77%
BlasGer/Sger_10x300_(2_1)-12 490ns ± 0% 408ns ± 0% −16.73%
BlasGer/Sger_10x300_(2_2)-12 486ns ± 0% 555ns ± 0% +14.20%
BlasGer/Sger_10x300_(2_3)-12 428ns ± 0% 547ns ± 0% +27.80%
BlasGer/Sger_10x300_(2_4)-12 438ns ± 0% 506ns ± 0% +15.53%
BlasGer/Sger_10x300_(2_10)-12 503ns ± 0% 535ns ± 0% +6.36%
BlasGer/Sger_10x300_(3_1)-12 448ns ± 0% 349ns ± 0% −22.10%
BlasGer/Sger_10x300_(3_2)-12 472ns ± 0% 522ns ± 0% +10.59%
BlasGer/Sger_10x300_(3_3)-12 440ns ± 0% 539ns ± 0% +22.50%
BlasGer/Sger_10x300_(3_4)-12 493ns ± 0% 549ns ± 0% +11.36%
BlasGer/Sger_10x300_(3_10)-12 414ns ± 0% 467ns ± 0% +12.80%
BlasGer/Sger_10x300_(4_1)-12 417ns ± 0% 415ns ± 0% −0.48%
BlasGer/Sger_10x300_(4_2)-12 420ns ± 0% 517ns ± 0% +23.10%
BlasGer/Sger_10x300_(4_3)-12 422ns ± 0% 571ns ± 0% +35.31%
BlasGer/Sger_10x300_(4_4)-12 433ns ± 0% 531ns ± 0% +22.63%
BlasGer/Sger_10x300_(4_10)-12 526ns ± 0% 553ns ± 0% +5.13%
BlasGer/Sger_10x300_(10_1)-12 409ns ± 0% 339ns ± 0% −17.11%
BlasGer/Sger_10x300_(10_2)-12 491ns ± 0% 529ns ± 0% +7.74%
BlasGer/Sger_10x300_(10_3)-12 503ns ± 0% 516ns ± 0% +2.58%
BlasGer/Sger_10x300_(10_4)-12 445ns ± 0% 583ns ± 0% +31.01%
BlasGer/Sger_10x300_(10_10)-12 483ns ± 0% 540ns ± 0% +11.80%
BlasGer/Sger_30x3_(1_1)-12 53.8ns ± 0% 279.0ns ± 0% +418.59%
BlasGer/Sger_30x3_(1_2)-12 53.6ns ± 0% 256.0ns ± 0% +377.61%
BlasGer/Sger_30x3_(1_3)-12 52.6ns ± 0% 304.0ns ± 0% +477.95%
BlasGer/Sger_30x3_(1_4)-12 44.8ns ± 0% 263.0ns ± 0% +487.05%
BlasGer/Sger_30x3_(1_10)-12 52.3ns ± 0% 287.0ns ± 0% +448.76%
BlasGer/Sger_30x3_(2_1)-12 53.1ns ± 0% 249.0ns ± 0% +368.93%
BlasGer/Sger_30x3_(2_2)-12 51.9ns ± 0% 237.0ns ± 0% +356.65%
BlasGer/Sger_30x3_(2_3)-12 50.6ns ± 0% 258.0ns ± 0% +409.88%
BlasGer/Sger_30x3_(2_4)-12 44.3ns ± 0% 236.0ns ± 0% +432.73%
BlasGer/Sger_30x3_(2_10)-12 53.3ns ± 0% 251.0ns ± 0% +370.92%
BlasGer/Sger_30x3_(3_1)-12 50.0ns ± 0% 270.0ns ± 0% +440.00%
BlasGer/Sger_30x3_(3_2)-12 48.6ns ± 0% 312.0ns ± 0% +541.98%
BlasGer/Sger_30x3_(3_3)-12 60.0ns ± 0% 270.0ns ± 0% +350.00%
BlasGer/Sger_30x3_(3_4)-12 51.6ns ± 0% 277.0ns ± 0% +436.82%
BlasGer/Sger_30x3_(3_10)-12 51.7ns ± 0% 278.0ns ± 0% +437.72%
BlasGer/Sger_30x3_(4_1)-12 55.4ns ± 0% 259.0ns ± 0% +367.51%
BlasGer/Sger_30x3_(4_2)-12 55.2ns ± 0% 251.0ns ± 0% +354.71%
BlasGer/Sger_30x3_(4_3)-12 49.5ns ± 0% 251.0ns ± 0% +407.07%
BlasGer/Sger_30x3_(4_4)-12 63.3ns ± 0% 234.0ns ± 0% +269.67%
BlasGer/Sger_30x3_(4_10)-12 48.4ns ± 0% 275.0ns ± 0% +468.18%
BlasGer/Sger_30x3_(10_1)-12 55.8ns ± 0% 269.0ns ± 0% +382.08%
BlasGer/Sger_30x3_(10_2)-12 53.6ns ± 0% 249.0ns ± 0% +364.55%
BlasGer/Sger_30x3_(10_3)-12 53.2ns ± 0% 261.0ns ± 0% +390.60%
BlasGer/Sger_30x3_(10_4)-12 55.5ns ± 0% 239.0ns ± 0% +330.63%
BlasGer/Sger_30x3_(10_10)-12 49.6ns ± 0% 277.0ns ± 0% +458.47%
BlasGer/Sger_30x10_(1_1)-12 60.6ns ± 0% 416.0ns ± 0% +586.47%
BlasGer/Sger_30x10_(1_2)-12 77.4ns ± 0% 429.0ns ± 0% +454.26%
BlasGer/Sger_30x10_(1_3)-12 75.0ns ± 0% 427.0ns ± 0% +469.33%
BlasGer/Sger_30x10_(1_4)-12 62.8ns ± 0% 351.0ns ± 0% +458.92%
BlasGer/Sger_30x10_(1_10)-12 75.4ns ± 0% 424.0ns ± 0% +462.33%
BlasGer/Sger_30x10_(2_1)-12 75.8ns ± 0% 353.0ns ± 0% +365.70%
BlasGer/Sger_30x10_(2_2)-12 64.4ns ± 0% 443.0ns ± 0% +587.89%
BlasGer/Sger_30x10_(2_3)-12 76.4ns ± 0% 455.0ns ± 0% +495.55%
BlasGer/Sger_30x10_(2_4)-12 74.6ns ± 0% 429.0ns ± 0% +475.07%
BlasGer/Sger_30x10_(2_10)-12 67.0ns ± 0% 360.0ns ± 0% +437.31%
BlasGer/Sger_30x10_(3_1)-12 83.2ns ± 0% 408.0ns ± 0% +390.38%
BlasGer/Sger_30x10_(3_2)-12 73.1ns ± 0% 445.0ns ± 0% +508.76%
BlasGer/Sger_30x10_(3_3)-12 77.2ns ± 0% 516.0ns ± 0% +568.39%
BlasGer/Sger_30x10_(3_4)-12 70.9ns ± 0% 391.0ns ± 0% +451.48%
BlasGer/Sger_30x10_(3_10)-12 75.0ns ± 0% 349.0ns ± 0% +365.33%
BlasGer/Sger_30x10_(4_1)-12 64.5ns ± 0% 377.0ns ± 0% +484.50%
BlasGer/Sger_30x10_(4_2)-12 64.3ns ± 0% 413.0ns ± 0% +542.30%
BlasGer/Sger_30x10_(4_3)-12 76.1ns ± 0% 340.0ns ± 0% +346.78%
BlasGer/Sger_30x10_(4_4)-12 71.3ns ± 0% 358.0ns ± 0% +402.10%
BlasGer/Sger_30x10_(4_10)-12 70.0ns ± 0% 425.0ns ± 0% +507.14%
BlasGer/Sger_30x10_(10_1)-12 73.9ns ± 0% 361.0ns ± 0% +388.50%
BlasGer/Sger_30x10_(10_2)-12 75.1ns ± 0% 350.0ns ± 0% +366.05%
BlasGer/Sger_30x10_(10_3)-12 69.6ns ± 0% 388.0ns ± 0% +457.47%
BlasGer/Sger_30x10_(10_4)-12 66.2ns ± 0% 404.0ns ± 0% +510.27%
BlasGer/Sger_30x10_(10_10)-12 63.9ns ± 0% 426.0ns ± 0% +566.67%
BlasGer/Sger_30x30_(1_1)-12 111ns ± 0% 768ns ± 0% +591.89%
BlasGer/Sger_30x30_(1_2)-12 147ns ± 0% 749ns ± 0% +409.52%
BlasGer/Sger_30x30_(1_3)-12 163ns ± 0% 756ns ± 0% +363.80%
BlasGer/Sger_30x30_(1_4)-12 165ns ± 0% 775ns ± 0% +369.70%
BlasGer/Sger_30x30_(1_10)-12 154ns ± 0% 771ns ± 0% +400.65%
BlasGer/Sger_30x30_(2_1)-12 153ns ± 0% 648ns ± 0% +323.53%
BlasGer/Sger_30x30_(2_2)-12 181ns ± 0% 684ns ± 0% +277.90%
BlasGer/Sger_30x30_(2_3)-12 162ns ± 0% 659ns ± 0% +306.79%
BlasGer/Sger_30x30_(2_4)-12 138ns ± 0% 663ns ± 0% +380.43%
BlasGer/Sger_30x30_(2_10)-12 167ns ± 0% 644ns ± 0% +285.63%
BlasGer/Sger_30x30_(3_1)-12 161ns ± 0% 629ns ± 0% +290.68%
BlasGer/Sger_30x30_(3_2)-12 158ns ± 0% 781ns ± 0% +394.30%
BlasGer/Sger_30x30_(3_3)-12 169ns ± 0% 650ns ± 0% +284.62%
BlasGer/Sger_30x30_(3_4)-12 157ns ± 0% 662ns ± 0% +321.66%
BlasGer/Sger_30x30_(3_10)-12 161ns ± 0% 774ns ± 0% +380.75%
BlasGer/Sger_30x30_(4_1)-12 159ns ± 0% 751ns ± 0% +372.33%
BlasGer/Sger_30x30_(4_2)-12 146ns ± 0% 724ns ± 0% +395.89%
BlasGer/Sger_30x30_(4_3)-12 158ns ± 0% 742ns ± 0% +369.62%
BlasGer/Sger_30x30_(4_4)-12 165ns ± 0% 755ns ± 0% +357.58%
BlasGer/Sger_30x30_(4_10)-12 166ns ± 0% 779ns ± 0% +369.28%
BlasGer/Sger_30x30_(10_1)-12 156ns ± 0% 758ns ± 0% +385.90%
BlasGer/Sger_30x30_(10_2)-12 153ns ± 0% 641ns ± 0% +318.95%
BlasGer/Sger_30x30_(10_3)-12 149ns ± 0% 698ns ± 0% +368.46%
BlasGer/Sger_30x30_(10_4)-12 165ns ± 0% 747ns ± 0% +352.73%
BlasGer/Sger_30x30_(10_10)-12 164ns ± 0% 702ns ± 0% +328.05%
BlasGer/Sger_30x100_(1_1)-12 387ns ± 0% 505ns ± 0% +30.49%
BlasGer/Sger_30x100_(1_2)-12 434ns ± 0% 440ns ± 0% +1.38%
BlasGer/Sger_30x100_(1_3)-12 428ns ± 0% 570ns ± 0% +33.18%
BlasGer/Sger_30x100_(1_4)-12 459ns ± 0% 484ns ± 0% +5.45%
BlasGer/Sger_30x100_(1_10)-12 412ns ± 0% 432ns ± 0% +4.85%
BlasGer/Sger_30x100_(2_1)-12 424ns ± 0% 455ns ± 0% +7.31%
BlasGer/Sger_30x100_(2_2)-12 434ns ± 0% 516ns ± 0% +18.89%
BlasGer/Sger_30x100_(2_3)-12 490ns ± 0% 514ns ± 0% +4.90%
BlasGer/Sger_30x100_(2_4)-12 483ns ± 0% 510ns ± 0% +5.59%
BlasGer/Sger_30x100_(2_10)-12 435ns ± 0% 529ns ± 0% +21.61%
BlasGer/Sger_30x100_(3_1)-12 490ns ± 0% 483ns ± 0% −1.43%
BlasGer/Sger_30x100_(3_2)-12 514ns ± 0% 542ns ± 0% +5.45%
BlasGer/Sger_30x100_(3_3)-12 484ns ± 0% 525ns ± 0% +8.47%
BlasGer/Sger_30x100_(3_4)-12 410ns ± 0% 539ns ± 0% +31.46%
BlasGer/Sger_30x100_(3_10)-12 491ns ± 0% 509ns ± 0% +3.67%
BlasGer/Sger_30x100_(4_1)-12 488ns ± 0% 471ns ± 0% −3.48%
BlasGer/Sger_30x100_(4_2)-12 487ns ± 0% 503ns ± 0% +3.29%
BlasGer/Sger_30x100_(4_3)-12 487ns ± 0% 545ns ± 0% +11.91%
BlasGer/Sger_30x100_(4_4)-12 453ns ± 0% 441ns ± 0% −2.65%
BlasGer/Sger_30x100_(4_10)-12 490ns ± 0% 543ns ± 0% +10.82%
BlasGer/Sger_30x100_(10_1)-12 489ns ± 0% 446ns ± 0% −8.79%
BlasGer/Sger_30x100_(10_2)-12 488ns ± 0% 518ns ± 0% +6.15%
BlasGer/Sger_30x100_(10_3)-12 470ns ± 0% 510ns ± 0% +8.51%
BlasGer/Sger_30x100_(10_4)-12 476ns ± 0% 454ns ± 0% −4.62%
BlasGer/Sger_30x100_(10_10)-12 431ns ± 0% 468ns ± 0% +8.58%
BlasGer/Sger_30x300_(1_1)-12 1.32µs ± 0% 1.30µs ± 0% −1.59%
BlasGer/Sger_30x300_(1_2)-12 1.40µs ± 0% 1.35µs ± 0% −3.50%
BlasGer/Sger_30x300_(1_3)-12 1.19µs ± 0% 1.38µs ± 0% +15.59%
BlasGer/Sger_30x300_(1_4)-12 1.44µs ± 0% 1.36µs ± 0% −6.16%
BlasGer/Sger_30x300_(1_10)-12 1.44µs ± 0% 1.36µs ± 0% −5.83%
BlasGer/Sger_30x300_(2_1)-12 1.41µs ± 0% 1.20µs ± 0% −14.70%
BlasGer/Sger_30x300_(2_2)-12 1.43µs ± 0% 1.35µs ± 0% −5.66%
BlasGer/Sger_30x300_(2_3)-12 1.19µs ± 0% 1.34µs ± 0% +13.14%
BlasGer/Sger_30x300_(2_4)-12 1.35µs ± 0% 1.32µs ± 0% −2.23%
BlasGer/Sger_30x300_(2_10)-12 1.33µs ± 0% 1.51µs ± 0% +13.22%
BlasGer/Sger_30x300_(3_1)-12 1.40µs ± 0% 1.19µs ± 0% −14.87%
BlasGer/Sger_30x300_(3_2)-12 1.44µs ± 0% 1.29µs ± 0% −10.74%
BlasGer/Sger_30x300_(3_3)-12 1.22µs ± 0% 1.28µs ± 0% +4.34%
BlasGer/Sger_30x300_(3_4)-12 1.44µs ± 0% 1.29µs ± 0% −10.48%
BlasGer/Sger_30x300_(3_10)-12 1.18µs ± 0% 1.34µs ± 0% +13.37%
BlasGer/Sger_30x300_(4_1)-12 1.27µs ± 0% 1.16µs ± 0% −8.95%
BlasGer/Sger_30x300_(4_2)-12 1.54µs ± 0% 1.25µs ± 0% −18.90%
BlasGer/Sger_30x300_(4_3)-12 1.43µs ± 0% 1.27µs ± 0% −11.21%
BlasGer/Sger_30x300_(4_4)-12 1.19µs ± 0% 1.30µs ± 0% +9.19%
BlasGer/Sger_30x300_(4_10)-12 1.22µs ± 0% 1.34µs ± 0% +9.98%
BlasGer/Sger_30x300_(10_1)-12 1.24µs ± 0% 1.14µs ± 0% −7.74%
BlasGer/Sger_30x300_(10_2)-12 1.28µs ± 0% 1.24µs ± 0% −3.50%
BlasGer/Sger_30x300_(10_3)-12 1.19µs ± 0% 1.27µs ± 0% +7.34%
BlasGer/Sger_30x300_(10_4)-12 1.42µs ± 0% 1.35µs ± 0% −5.33%
BlasGer/Sger_30x300_(10_10)-12 1.37µs ± 0% 1.32µs ± 0% −3.58%
BlasGer/Sger_30x1000_(1_1)-12 4.31µs ± 0% 1.44µs ± 0% −66.71%
BlasGer/Sger_30x1000_(1_2)-12 4.81µs ± 0% 2.09µs ± 0% −56.47%
BlasGer/Sger_30x1000_(1_3)-12 3.96µs ± 0% 2.18µs ± 0% −44.94%
BlasGer/Sger_30x1000_(1_4)-12 4.22µs ± 0% 2.31µs ± 0% −45.22%
BlasGer/Sger_30x1000_(1_10)-12 5.97µs ± 0% 2.50µs ± 0% −58.15%
BlasGer/Sger_30x1000_(2_1)-12 4.80µs ± 0% 1.59µs ± 0% −66.91%
BlasGer/Sger_30x1000_(2_2)-12 4.70µs ± 0% 2.13µs ± 0% −54.78%
BlasGer/Sger_30x1000_(2_3)-12 4.03µs ± 0% 2.25µs ± 0% −44.03%
BlasGer/Sger_30x1000_(2_4)-12 5.25µs ± 0% 2.17µs ± 0% −58.78%
BlasGer/Sger_30x1000_(2_10)-12 5.19µs ± 0% 2.40µs ± 0% −53.84%
BlasGer/Sger_30x1000_(3_1)-12 4.78µs ± 0% 1.52µs ± 0% −68.30%
BlasGer/Sger_30x1000_(3_2)-12 4.78µs ± 0% 2.17µs ± 0% −54.62%
BlasGer/Sger_30x1000_(3_3)-12 4.81µs ± 0% 2.17µs ± 0% −54.79%
BlasGer/Sger_30x1000_(3_4)-12 4.77µs ± 0% 2.10µs ± 0% −56.09%
BlasGer/Sger_30x1000_(3_10)-12 5.38µs ± 0% 2.39µs ± 0% −55.57%
BlasGer/Sger_30x1000_(4_1)-12 4.65µs ± 0% 1.53µs ± 0% −67.05%
BlasGer/Sger_30x1000_(4_2)-12 4.43µs ± 0% 2.17µs ± 0% −50.96%
BlasGer/Sger_30x1000_(4_3)-12 4.69µs ± 0% 2.14µs ± 0% −54.34%
BlasGer/Sger_30x1000_(4_4)-12 4.80µs ± 0% 2.11µs ± 0% −55.94%
BlasGer/Sger_30x1000_(4_10)-12 5.72µs ± 0% 2.44µs ± 0% −57.39%
BlasGer/Sger_30x1000_(10_1)-12 5.73µs ± 0% 1.56µs ± 0% −72.80%
BlasGer/Sger_30x1000_(10_2)-12 4.76µs ± 0% 2.19µs ± 0% −53.94%
BlasGer/Sger_30x1000_(10_3)-12 4.19µs ± 0% 2.11µs ± 0% −49.70%
BlasGer/Sger_30x1000_(10_4)-12 5.27µs ± 0% 2.20µs ± 0% −58.17%
BlasGer/Sger_30x1000_(10_10)-12 5.76µs ± 0% 2.52µs ± 0% −56.28%
BlasGer/Sger_100x3_(1_1)-12 148ns ± 0% 625ns ± 0% +322.30%
BlasGer/Sger_100x3_(1_2)-12 132ns ± 0% 660ns ± 0% +400.00%
BlasGer/Sger_100x3_(1_3)-12 141ns ± 0% 582ns ± 0% +312.77%
BlasGer/Sger_100x3_(1_4)-12 133ns ± 0% 700ns ± 0% +426.32%
BlasGer/Sger_100x3_(1_10)-12 131ns ± 0% 712ns ± 0% +443.51%
BlasGer/Sger_100x3_(2_1)-12 147ns ± 0% 689ns ± 0% +368.71%
BlasGer/Sger_100x3_(2_2)-12 136ns ± 0% 692ns ± 0% +408.82%
BlasGer/Sger_100x3_(2_3)-12 144ns ± 0% 713ns ± 0% +395.14%
BlasGer/Sger_100x3_(2_4)-12 153ns ± 0% 693ns ± 0% +352.94%
BlasGer/Sger_100x3_(2_10)-12 149ns ± 0% 679ns ± 0% +355.70%
BlasGer/Sger_100x3_(3_1)-12 131ns ± 0% 639ns ± 0% +387.79%
BlasGer/Sger_100x3_(3_2)-12 132ns ± 0% 628ns ± 0% +375.76%
BlasGer/Sger_100x3_(3_3)-12 136ns ± 0% 649ns ± 0% +377.21%
BlasGer/Sger_100x3_(3_4)-12 147ns ± 0% 648ns ± 0% +340.82%
BlasGer/Sger_100x3_(3_10)-12 130ns ± 0% 691ns ± 0% +431.54%
BlasGer/Sger_100x3_(4_1)-12 151ns ± 0% 642ns ± 0% +325.17%
BlasGer/Sger_100x3_(4_2)-12 153ns ± 0% 703ns ± 0% +359.48%
BlasGer/Sger_100x3_(4_3)-12 164ns ± 0% 578ns ± 0% +252.44%
BlasGer/Sger_100x3_(4_4)-12 150ns ± 0% 695ns ± 0% +363.33%
BlasGer/Sger_100x3_(4_10)-12 153ns ± 0% 621ns ± 0% +305.88%
BlasGer/Sger_100x3_(10_1)-12 151ns ± 0% 632ns ± 0% +318.54%
BlasGer/Sger_100x3_(10_2)-12 144ns ± 0% 690ns ± 0% +379.17%
BlasGer/Sger_100x3_(10_3)-12 129ns ± 0% 666ns ± 0% +416.28%
BlasGer/Sger_100x3_(10_4)-12 126ns ± 0% 680ns ± 0% +439.68%
BlasGer/Sger_100x3_(10_10)-12 149ns ± 0% 622ns ± 0% +317.45%
BlasGer/Sger_100x10_(1_1)-12 212ns ± 0% 1055ns ± 0% +397.64%
BlasGer/Sger_100x10_(1_2)-12 240ns ± 0% 1199ns ± 0% +399.58%
BlasGer/Sger_100x10_(1_3)-12 213ns ± 0% 1209ns ± 0% +467.61%
BlasGer/Sger_100x10_(1_4)-12 240ns ± 0% 1206ns ± 0% +402.50%
BlasGer/Sger_100x10_(1_10)-12 239ns ± 0% 1184ns ± 0% +395.40%
BlasGer/Sger_100x10_(2_1)-12 230ns ± 0% 1313ns ± 0% +470.87%
BlasGer/Sger_100x10_(2_2)-12 244ns ± 0% 1049ns ± 0% +329.92%
BlasGer/Sger_100x10_(2_3)-12 189ns ± 0% 1184ns ± 0% +526.46%
BlasGer/Sger_100x10_(2_4)-12 209ns ± 0% 1052ns ± 0% +403.35%
BlasGer/Sger_100x10_(2_10)-12 203ns ± 0% 1109ns ± 0% +446.31%
BlasGer/Sger_100x10_(3_1)-12 217ns ± 0% 1229ns ± 0% +466.36%
BlasGer/Sger_100x10_(3_2)-12 216ns ± 0% 1127ns ± 0% +421.76%
BlasGer/Sger_100x10_(3_3)-12 235ns ± 0% 1163ns ± 0% +394.89%
BlasGer/Sger_100x10_(3_4)-12 234ns ± 0% 1016ns ± 0% +334.19%
BlasGer/Sger_100x10_(3_10)-12 206ns ± 0% 1168ns ± 0% +466.99%
BlasGer/Sger_100x10_(4_1)-12 241ns ± 0% 1215ns ± 0% +404.15%
BlasGer/Sger_100x10_(4_2)-12 219ns ± 0% 1172ns ± 0% +435.16%
BlasGer/Sger_100x10_(4_3)-12 231ns ± 0% 1277ns ± 0% +452.81%
BlasGer/Sger_100x10_(4_4)-12 212ns ± 0% 1055ns ± 0% +397.64%
BlasGer/Sger_100x10_(4_10)-12 213ns ± 0% 1036ns ± 0% +386.38%
BlasGer/Sger_100x10_(10_1)-12 240ns ± 0% 1212ns ± 0% +405.00%
BlasGer/Sger_100x10_(10_2)-12 188ns ± 0% 1132ns ± 0% +502.13%
BlasGer/Sger_100x10_(10_3)-12 256ns ± 0% 1165ns ± 0% +355.08%
BlasGer/Sger_100x10_(10_4)-12 241ns ± 0% 964ns ± 0% +300.00%
BlasGer/Sger_100x10_(10_10)-12 214ns ± 0% 1118ns ± 0% +422.43%
BlasGer/Sger_100x30_(1_1)-12 437ns ± 0% 2266ns ± 0% +418.54%
BlasGer/Sger_100x30_(1_2)-12 527ns ± 0% 2417ns ± 0% +358.63%
BlasGer/Sger_100x30_(1_3)-12 478ns ± 0% 2456ns ± 0% +413.81%
BlasGer/Sger_100x30_(1_4)-12 507ns ± 0% 2710ns ± 0% +434.52%
BlasGer/Sger_100x30_(1_10)-12 440ns ± 0% 2285ns ± 0% +419.32%
BlasGer/Sger_100x30_(2_1)-12 442ns ± 0% 2243ns ± 0% +407.47%
BlasGer/Sger_100x30_(2_2)-12 451ns ± 0% 2061ns ± 0% +356.98%
BlasGer/Sger_100x30_(2_3)-12 441ns ± 0% 2497ns ± 0% +466.21%
BlasGer/Sger_100x30_(2_4)-12 578ns ± 0% 2297ns ± 0% +297.40%
BlasGer/Sger_100x30_(2_10)-12 468ns ± 0% 2282ns ± 0% +387.61%
BlasGer/Sger_100x30_(3_1)-12 579ns ± 0% 2351ns ± 0% +306.04%
BlasGer/Sger_100x30_(3_2)-12 516ns ± 0% 2191ns ± 0% +324.61%
BlasGer/Sger_100x30_(3_3)-12 558ns ± 0% 2202ns ± 0% +294.62%
BlasGer/Sger_100x30_(3_4)-12 440ns ± 0% 2517ns ± 0% +472.05%
BlasGer/Sger_100x30_(3_10)-12 559ns ± 0% 2128ns ± 0% +280.68%
BlasGer/Sger_100x30_(4_1)-12 524ns ± 0% 2428ns ± 0% +363.36%
BlasGer/Sger_100x30_(4_2)-12 524ns ± 0% 2097ns ± 0% +300.19%
BlasGer/Sger_100x30_(4_3)-12 445ns ± 0% 2021ns ± 0% +354.16%
BlasGer/Sger_100x30_(4_4)-12 521ns ± 0% 2013ns ± 0% +286.37%
BlasGer/Sger_100x30_(4_10)-12 585ns ± 0% 2149ns ± 0% +267.35%
BlasGer/Sger_100x30_(10_1)-12 654ns ± 0% 2143ns ± 0% +227.68%
BlasGer/Sger_100x30_(10_2)-12 525ns ± 0% 2158ns ± 0% +311.05%
BlasGer/Sger_100x30_(10_3)-12 441ns ± 0% 2379ns ± 0% +439.46%
BlasGer/Sger_100x30_(10_4)-12 568ns ± 0% 2505ns ± 0% +341.02%
BlasGer/Sger_100x30_(10_10)-12 570ns ± 0% 2217ns ± 0% +288.95%
BlasGer/Sger_100x100_(1_1)-12 1.39µs ± 0% 3.15µs ± 0% +127.42%
BlasGer/Sger_100x100_(1_2)-12 1.38µs ± 0% 5.01µs ± 0% +263.09%
BlasGer/Sger_100x100_(1_3)-12 1.38µs ± 0% 4.41µs ± 0% +219.35%
BlasGer/Sger_100x100_(1_4)-12 1.54µs ± 0% 3.89µs ± 0% +153.06%
BlasGer/Sger_100x100_(1_10)-12 1.56µs ± 0% 2.79µs ± 0% +79.32%
BlasGer/Sger_100x100_(2_1)-12 1.56µs ± 0% 2.32µs ± 0% +48.12%
BlasGer/Sger_100x100_(2_2)-12 1.60µs ± 0% 2.43µs ± 0% +51.59%
BlasGer/Sger_100x100_(2_3)-12 1.50µs ± 0% 3.04µs ± 0% +101.86%
BlasGer/Sger_100x100_(2_4)-12 1.60µs ± 0% 2.41µs ± 0% +50.00%
BlasGer/Sger_100x100_(2_10)-12 1.59µs ± 0% 3.06µs ± 0% +92.33%
BlasGer/Sger_100x100_(3_1)-12 1.44µs ± 0% 2.14µs ± 0% +48.40%
BlasGer/Sger_100x100_(3_2)-12 1.61µs ± 0% 2.96µs ± 0% +83.45%
BlasGer/Sger_100x100_(3_3)-12 1.78µs ± 0% 2.51µs ± 0% +41.16%
BlasGer/Sger_100x100_(3_4)-12 1.58µs ± 0% 2.40µs ± 0% +52.15%
BlasGer/Sger_100x100_(3_10)-12 1.61µs ± 0% 2.32µs ± 0% +44.46%
BlasGer/Sger_100x100_(4_1)-12 1.34µs ± 0% 2.19µs ± 0% +63.83%
BlasGer/Sger_100x100_(4_2)-12 1.59µs ± 0% 3.46µs ± 0% +117.90%
BlasGer/Sger_100x100_(4_3)-12 1.66µs ± 0% 2.27µs ± 0% +36.91%
BlasGer/Sger_100x100_(4_4)-12 1.51µs ± 0% 2.68µs ± 0% +76.96%
BlasGer/Sger_100x100_(4_10)-12 1.46µs ± 0% 2.63µs ± 0% +79.77%
BlasGer/Sger_100x100_(10_1)-12 1.62µs ± 0% 2.88µs ± 0% +78.34%
BlasGer/Sger_100x100_(10_2)-12 1.37µs ± 0% 2.74µs ± 0% +100.15%
BlasGer/Sger_100x100_(10_3)-12 1.70µs ± 0% 2.14µs ± 0% +25.87%
BlasGer/Sger_100x100_(10_4)-12 1.35µs ± 0% 2.25µs ± 0% +67.06%
BlasGer/Sger_100x100_(10_10)-12 1.56µs ± 0% 2.43µs ± 0% +55.83%
BlasGer/Sger_100x300_(1_1)-12 3.35µs ± 0% 2.76µs ± 0% −17.68%
BlasGer/Sger_100x300_(1_2)-12 4.88µs ± 0% 3.49µs ± 0% −28.38%
BlasGer/Sger_100x300_(1_3)-12 4.80µs ± 0% 3.08µs ± 0% −35.79%
BlasGer/Sger_100x300_(1_4)-12 5.35µs ± 0% 3.10µs ± 0% −41.93%
BlasGer/Sger_100x300_(1_10)-12 4.75µs ± 0% 3.15µs ± 0% −33.59%
BlasGer/Sger_100x300_(2_1)-12 4.04µs ± 0% 2.64µs ± 0% −34.83%
BlasGer/Sger_100x300_(2_2)-12 5.33µs ± 0% 3.05µs ± 0% −42.82%
BlasGer/Sger_100x300_(2_3)-12 5.17µs ± 0% 3.87µs ± 0% −25.09%
BlasGer/Sger_100x300_(2_4)-12 4.57µs ± 0% 3.83µs ± 0% −16.34%
BlasGer/Sger_100x300_(2_10)-12 4.16µs ± 0% 3.45µs ± 0% −16.93%
BlasGer/Sger_100x300_(3_1)-12 4.35µs ± 0% 2.62µs ± 0% −39.85%
BlasGer/Sger_100x300_(3_2)-12 4.29µs ± 0% 5.33µs ± 0% +24.20%
BlasGer/Sger_100x300_(3_3)-12 5.13µs ± 0% 9.07µs ± 0% +76.98%
BlasGer/Sger_100x300_(3_4)-12 5.09µs ± 0% 7.55µs ± 0% +48.46%
BlasGer/Sger_100x300_(3_10)-12 5.12µs ± 0% 4.93µs ± 0% −3.77%
BlasGer/Sger_100x300_(4_1)-12 4.75µs ± 0% 2.99µs ± 0% −37.08%
BlasGer/Sger_100x300_(4_2)-12 4.36µs ± 0% 3.34µs ± 0% −23.38%
BlasGer/Sger_100x300_(4_3)-12 4.76µs ± 0% 3.08µs ± 0% −35.36%
BlasGer/Sger_100x300_(4_4)-12 4.61µs ± 0% 3.03µs ± 0% −34.22%
BlasGer/Sger_100x300_(4_10)-12 4.46µs ± 0% 2.87µs ± 0% −35.62%
BlasGer/Sger_100x300_(10_1)-12 4.38µs ± 0% 2.77µs ± 0% −36.72%
BlasGer/Sger_100x300_(10_2)-12 4.40µs ± 0% 2.96µs ± 0% −32.77%
BlasGer/Sger_100x300_(10_3)-12 4.56µs ± 0% 2.97µs ± 0% −34.75%
BlasGer/Sger_100x300_(10_4)-12 4.26µs ± 0% 2.80µs ± 0% −34.33%
BlasGer/Sger_100x300_(10_10)-12 4.11µs ± 0% 3.20µs ± 0% −22.24%
BlasGer/Sger_100x1000_(1_1)-12 16.3µs ± 0% 5.0µs ± 0% −69.09%
BlasGer/Sger_100x1000_(1_2)-12 23.2µs ± 0% 6.4µs ± 0% −72.38%
BlasGer/Sger_100x1000_(1_3)-12 23.0µs ± 0% 5.6µs ± 0% −75.49%
BlasGer/Sger_100x1000_(1_4)-12 23.2µs ± 0% 5.5µs ± 0% −76.09%
BlasGer/Sger_100x1000_(1_10)-12 25.2µs ± 0% 7.2µs ± 0% −71.40%
BlasGer/Sger_100x1000_(2_1)-12 19.6µs ± 0% 4.6µs ± 0% −76.36%
BlasGer/Sger_100x1000_(2_2)-12 21.4µs ± 0% 6.0µs ± 0% −72.00%
BlasGer/Sger_100x1000_(2_3)-12 22.9µs ± 0% 5.4µs ± 0% −76.36%
BlasGer/Sger_100x1000_(2_4)-12 20.8µs ± 0% 5.8µs ± 0% −72.21%
BlasGer/Sger_100x1000_(2_10)-12 20.7µs ± 0% 5.9µs ± 0% −71.52%
BlasGer/Sger_100x1000_(3_1)-12 20.9µs ± 0% 4.4µs ± 0% −78.86%
BlasGer/Sger_100x1000_(3_2)-12 20.1µs ± 0% 5.8µs ± 0% −71.24%
BlasGer/Sger_100x1000_(3_3)-12 23.6µs ± 0% 4.6µs ± 0% −80.73%
BlasGer/Sger_100x1000_(3_4)-12 19.8µs ± 0% 4.7µs ± 0% −76.02%
BlasGer/Sger_100x1000_(3_10)-12 25.2µs ± 0% 5.7µs ± 0% −77.47%
BlasGer/Sger_100x1000_(4_1)-12 19.2µs ± 0% 3.4µs ± 0% −82.07%
BlasGer/Sger_100x1000_(4_2)-12 23.5µs ± 0% 4.9µs ± 0% −79.08%
BlasGer/Sger_100x1000_(4_3)-12 20.9µs ± 0% 4.6µs ± 0% −77.99%
BlasGer/Sger_100x1000_(4_4)-12 23.8µs ± 0% 4.7µs ± 0% −80.36%
BlasGer/Sger_100x1000_(4_10)-12 22.6µs ± 0% 5.3µs ± 0% −76.60%
BlasGer/Sger_100x1000_(10_1)-12 18.0µs ± 0% 4.9µs ± 0% −72.60%
BlasGer/Sger_100x1000_(10_2)-12 18.7µs ± 0% 11.6µs ± 0% −37.99%
BlasGer/Sger_100x1000_(10_3)-12 21.2µs ± 0% 8.9µs ± 0% −58.08%
BlasGer/Sger_100x1000_(10_4)-12 20.6µs ± 0% 8.5µs ± 0% −58.88%
BlasGer/Sger_100x1000_(10_10)-12 24.5µs ± 0% 8.9µs ± 0% −63.59%
BlasGer/Sger_300x10_(1_1)-12 606ns ± 0% 3296ns ± 0% +443.89%
BlasGer/Sger_300x10_(1_2)-12 583ns ± 0% 3530ns ± 0% +505.49%
BlasGer/Sger_300x10_(1_3)-12 685ns ± 0% 3110ns ± 0% +354.01%
BlasGer/Sger_300x10_(1_4)-12 695ns ± 0% 2795ns ± 0% +302.16%
BlasGer/Sger_300x10_(1_10)-12 685ns ± 0% 2669ns ± 0% +289.64%
BlasGer/Sger_300x10_(2_1)-12 644ns ± 0% 3209ns ± 0% +398.29%
BlasGer/Sger_300x10_(2_2)-12 584ns ± 0% 3156ns ± 0% +440.41%
BlasGer/Sger_300x10_(2_3)-12 603ns ± 0% 3540ns ± 0% +487.06%
BlasGer/Sger_300x10_(2_4)-12 769ns ± 0% 3184ns ± 0% +314.04%
BlasGer/Sger_300x10_(2_10)-12 585ns ± 0% 3485ns ± 0% +495.73%
BlasGer/Sger_300x10_(3_1)-12 689ns ± 0% 2998ns ± 0% +335.12%
BlasGer/Sger_300x10_(3_2)-12 675ns ± 0% 3061ns ± 0% +353.48%
BlasGer/Sger_300x10_(3_3)-12 616ns ± 0% 3186ns ± 0% +417.21%
BlasGer/Sger_300x10_(3_4)-12 683ns ± 0% 2667ns ± 0% +290.48%
BlasGer/Sger_300x10_(3_10)-12 561ns ± 0% 2780ns ± 0% +395.54%
BlasGer/Sger_300x10_(4_1)-12 619ns ± 0% 2893ns ± 0% +367.37%
BlasGer/Sger_300x10_(4_2)-12 615ns ± 0% 3266ns ± 0% +431.06%
BlasGer/Sger_300x10_(4_3)-12 634ns ± 0% 3187ns ± 0% +402.68%
BlasGer/Sger_300x10_(4_4)-12 606ns ± 0% 2772ns ± 0% +357.43%
BlasGer/Sger_300x10_(4_10)-12 692ns ± 0% 3053ns ± 0% +341.18%
BlasGer/Sger_300x10_(10_1)-12 573ns ± 0% 2676ns ± 0% +367.02%
BlasGer/Sger_300x10_(10_2)-12 637ns ± 0% 3497ns ± 0% +448.98%
BlasGer/Sger_300x10_(10_3)-12 670ns ± 0% 2904ns ± 0% +333.43%
BlasGer/Sger_300x10_(10_4)-12 697ns ± 0% 3188ns ± 0% +357.39%
BlasGer/Sger_300x10_(10_10)-12 676ns ± 0% 2683ns ± 0% +296.89%
BlasGer/Sger_300x30_(1_1)-12 1.27µs ± 0% 3.72µs ± 0% +194.00%
BlasGer/Sger_300x30_(1_2)-12 1.58µs ± 0% 3.70µs ± 0% +133.27%
BlasGer/Sger_300x30_(1_3)-12 1.56µs ± 0% 3.10µs ± 0% +98.46%
BlasGer/Sger_300x30_(1_4)-12 1.56µs ± 0% 4.20µs ± 0% +169.36%
BlasGer/Sger_300x30_(1_10)-12 1.35µs ± 0% 3.93µs ± 0% +190.24%
BlasGer/Sger_300x30_(2_1)-12 1.34µs ± 0% 3.39µs ± 0% +152.27%
BlasGer/Sger_300x30_(2_2)-12 1.64µs ± 0% 6.25µs ± 0% +281.92%
BlasGer/Sger_300x30_(2_3)-12 1.39µs ± 0% 6.62µs ± 0% +374.87%
BlasGer/Sger_300x30_(2_4)-12 1.46µs ± 0% 8.52µs ± 0% +484.64%
BlasGer/Sger_300x30_(2_10)-12 1.59µs ± 0% 6.15µs ± 0% +287.15%
BlasGer/Sger_300x30_(3_1)-12 1.60µs ± 0% 4.28µs ± 0% +168.23%
BlasGer/Sger_300x30_(3_2)-12 1.62µs ± 0% 3.27µs ± 0% +101.85%
BlasGer/Sger_300x30_(3_3)-12 1.38µs ± 0% 4.02µs ± 0% +191.09%
BlasGer/Sger_300x30_(3_4)-12 1.63µs ± 0% 5.88µs ± 0% +261.56%
BlasGer/Sger_300x30_(3_10)-12 1.38µs ± 0% 5.63µs ± 0% +306.87%
BlasGer/Sger_300x30_(4_1)-12 1.34µs ± 0% 4.36µs ± 0% +224.39%
BlasGer/Sger_300x30_(4_2)-12 1.61µs ± 0% 4.11µs ± 0% +155.47%
BlasGer/Sger_300x30_(4_3)-12 1.50µs ± 0% 6.62µs ± 0% +340.52%
BlasGer/Sger_300x30_(4_4)-12 1.69µs ± 0% 3.42µs ± 0% +102.37%
BlasGer/Sger_300x30_(4_10)-12 1.60µs ± 0% 3.48µs ± 0% +118.17%
BlasGer/Sger_300x30_(10_1)-12 1.63µs ± 0% 4.25µs ± 0% +159.91%
BlasGer/Sger_300x30_(10_2)-12 1.65µs ± 0% 3.69µs ± 0% +123.92%
BlasGer/Sger_300x30_(10_3)-12 1.64µs ± 0% 3.99µs ± 0% +143.39%
BlasGer/Sger_300x30_(10_4)-12 1.44µs ± 0% 2.97µs ± 0% +105.96%
BlasGer/Sger_300x30_(10_10)-12 1.38µs ± 0% 3.17µs ± 0% +129.05%
BlasGer/Sger_300x100_(1_1)-12 3.48µs ± 0% 2.83µs ± 0% −18.62%
BlasGer/Sger_300x100_(1_2)-12 4.01µs ± 0% 3.01µs ± 0% −25.05%
BlasGer/Sger_300x100_(1_3)-12 4.54µs ± 0% 2.76µs ± 0% −39.30%
BlasGer/Sger_300x100_(1_4)-12 4.64µs ± 0% 2.85µs ± 0% −38.56%
BlasGer/Sger_300x100_(1_10)-12 5.26µs ± 0% 3.14µs ± 0% −40.25%
BlasGer/Sger_300x100_(2_1)-12 4.05µs ± 0% 3.62µs ± 0% −10.73%
BlasGer/Sger_300x100_(2_2)-12 4.89µs ± 0% 3.61µs ± 0% −26.15%
BlasGer/Sger_300x100_(2_3)-12 5.34µs ± 0% 2.98µs ± 0% −44.15%
BlasGer/Sger_300x100_(2_4)-12 5.41µs ± 0% 3.04µs ± 0% −43.90%
BlasGer/Sger_300x100_(2_10)-12 4.87µs ± 0% 2.50µs ± 0% −48.64%
BlasGer/Sger_300x100_(3_1)-12 5.11µs ± 0% 2.99µs ± 0% −41.53%
BlasGer/Sger_300x100_(3_2)-12 4.89µs ± 0% 2.70µs ± 0% −44.75%
BlasGer/Sger_300x100_(3_3)-12 5.38µs ± 0% 2.91µs ± 0% −46.02%
BlasGer/Sger_300x100_(3_4)-12 4.70µs ± 0% 2.90µs ± 0% −38.44%
BlasGer/Sger_300x100_(3_10)-12 4.22µs ± 0% 3.50µs ± 0% −16.96%
BlasGer/Sger_300x100_(4_1)-12 4.52µs ± 0% 2.95µs ± 0% −34.73%
BlasGer/Sger_300x100_(4_2)-12 5.09µs ± 0% 2.92µs ± 0% −42.75%
BlasGer/Sger_300x100_(4_3)-12 4.60µs ± 0% 2.76µs ± 0% −39.92%
BlasGer/Sger_300x100_(4_4)-12 5.14µs ± 0% 2.55µs ± 0% −50.40%
BlasGer/Sger_300x100_(4_10)-12 5.09µs ± 0% 3.79µs ± 0% −25.68%
BlasGer/Sger_300x100_(10_1)-12 5.16µs ± 0% 6.01µs ± 0% +16.55%
BlasGer/Sger_300x100_(10_2)-12 4.87µs ± 0% 5.68µs ± 0% +16.63%
BlasGer/Sger_300x100_(10_3)-12 4.35µs ± 0% 6.14µs ± 0% +41.19%
BlasGer/Sger_300x100_(10_4)-12 4.47µs ± 0% 6.29µs ± 0% +40.76%
BlasGer/Sger_300x100_(10_10)-12 4.86µs ± 0% 2.59µs ± 0% −46.68%
BlasGer/Sger_300x300_(1_1)-12 14.5µs ± 0% 4.3µs ± 0% −70.11%
BlasGer/Sger_300x300_(1_2)-12 18.5µs ± 0% 4.6µs ± 0% −75.23%
BlasGer/Sger_300x300_(1_3)-12 20.5µs ± 0% 4.7µs ± 0% −77.24%
BlasGer/Sger_300x300_(1_4)-12 21.4µs ± 0% 6.3µs ± 0% −70.42%
BlasGer/Sger_300x300_(1_10)-12 20.2µs ± 0% 4.6µs ± 0% −77.25%
BlasGer/Sger_300x300_(2_1)-12 17.6µs ± 0% 3.9µs ± 0% −78.05%
BlasGer/Sger_300x300_(2_2)-12 20.6µs ± 0% 5.2µs ± 0% −74.65%
BlasGer/Sger_300x300_(2_3)-12 20.9µs ± 0% 5.3µs ± 0% −74.63%
BlasGer/Sger_300x300_(2_4)-12 17.2µs ± 0% 5.8µs ± 0% −66.28%
BlasGer/Sger_300x300_(2_10)-12 17.7µs ± 0% 4.8µs ± 0% −72.72%
BlasGer/Sger_300x300_(3_1)-12 17.3µs ± 0% 4.0µs ± 0% −76.98%
BlasGer/Sger_300x300_(3_2)-12 15.2µs ± 0% 3.9µs ± 0% −74.69%
BlasGer/Sger_300x300_(3_3)-12 21.8µs ± 0% 7.0µs ± 0% −67.89%
BlasGer/Sger_300x300_(3_4)-12 15.9µs ± 0% 4.7µs ± 0% −70.18%
BlasGer/Sger_300x300_(3_10)-12 18.1µs ± 0% 4.6µs ± 0% −74.40%
BlasGer/Sger_300x300_(4_1)-12 16.9µs ± 0% 4.7µs ± 0% −72.03%
BlasGer/Sger_300x300_(4_2)-12 20.3µs ± 0% 5.1µs ± 0% −75.08%
BlasGer/Sger_300x300_(4_3)-12 17.7µs ± 0% 4.7µs ± 0% −73.27%
BlasGer/Sger_300x300_(4_4)-12 20.3µs ± 0% 4.5µs ± 0% −77.83%
BlasGer/Sger_300x300_(4_10)-12 16.7µs ± 0% 4.8µs ± 0% −71.25%
BlasGer/Sger_300x300_(10_1)-12 17.7µs ± 0% 5.3µs ± 0% −70.15%
BlasGer/Sger_300x300_(10_2)-12 17.3µs ± 0% 5.0µs ± 0% −70.92%
BlasGer/Sger_300x300_(10_3)-12 21.6µs ± 0% 4.7µs ± 0% −78.10%
BlasGer/Sger_300x300_(10_4)-12 18.3µs ± 0% 6.6µs ± 0% −64.12%
BlasGer/Sger_300x300_(10_10)-12 18.5µs ± 0% 6.1µs ± 0% −66.88%
BlasGer/Sger_300x1000_(1_1)-12 61.1µs ± 0% 12.6µs ± 0% −79.33%
BlasGer/Sger_300x1000_(1_2)-12 62.3µs ± 0% 17.5µs ± 0% −71.90%
BlasGer/Sger_300x1000_(1_3)-12 58.3µs ± 0% 11.3µs ± 0% −80.65%
BlasGer/Sger_300x1000_(1_4)-12 62.5µs ± 0% 16.5µs ± 0% −73.67%
BlasGer/Sger_300x1000_(1_10)-12 64.0µs ± 0% 13.3µs ± 0% −79.20%
BlasGer/Sger_300x1000_(2_1)-12 55.7µs ± 0% 11.1µs ± 0% −80.08%
BlasGer/Sger_300x1000_(2_2)-12 55.4µs ± 0% 10.3µs ± 0% −81.40%
BlasGer/Sger_300x1000_(2_3)-12 54.9µs ± 0% 14.3µs ± 0% −73.91%
BlasGer/Sger_300x1000_(2_4)-12 71.9µs ± 0% 18.0µs ± 0% −75.00%
BlasGer/Sger_300x1000_(2_10)-12 60.6µs ± 0% 23.7µs ± 0% −60.90%
BlasGer/Sger_300x1000_(3_1)-12 67.8µs ± 0% 21.9µs ± 0% −67.70%
BlasGer/Sger_300x1000_(3_2)-12 55.8µs ± 0% 25.0µs ± 0% −55.14%
BlasGer/Sger_300x1000_(3_3)-12 69.1µs ± 0% 11.8µs ± 0% −82.94%
BlasGer/Sger_300x1000_(3_4)-12 57.2µs ± 0% 12.6µs ± 0% −77.93%
BlasGer/Sger_300x1000_(3_10)-12 61.4µs ± 0% 13.4µs ± 0% −78.24%
BlasGer/Sger_300x1000_(4_1)-12 61.5µs ± 0% 10.2µs ± 0% −83.41%
BlasGer/Sger_300x1000_(4_2)-12 66.2µs ± 0% 11.5µs ± 0% −82.58%
BlasGer/Sger_300x1000_(4_3)-12 69.3µs ± 0% 12.2µs ± 0% −82.37%
BlasGer/Sger_300x1000_(4_4)-12 66.3µs ± 0% 14.1µs ± 0% −78.75%
BlasGer/Sger_300x1000_(4_10)-12 69.7µs ± 0% 12.9µs ± 0% −81.44%
BlasGer/Sger_300x1000_(10_1)-12 55.8µs ± 0% 11.3µs ± 0% −79.72%
BlasGer/Sger_300x1000_(10_2)-12 68.9µs ± 0% 12.8µs ± 0% −81.41%
BlasGer/Sger_300x1000_(10_3)-12 67.1µs ± 0% 12.2µs ± 0% −81.77%
BlasGer/Sger_300x1000_(10_4)-12 61.1µs ± 0% 12.1µs ± 0% −80.14%
BlasGer/Sger_300x1000_(10_10)-12 74.3µs ± 0% 15.3µs ± 0% −79.47%
BlasGer/Sger_300x10000_(1_1)-12 755µs ± 0% 149µs ± 0% −80.30%
BlasGer/Sger_300x10000_(1_2)-12 622µs ± 0% 179µs ± 0% −71.16%
BlasGer/Sger_300x10000_(1_3)-12 723µs ± 0% 169µs ± 0% −76.69%
BlasGer/Sger_300x10000_(1_4)-12 852µs ± 0% 182µs ± 0% −78.62%
BlasGer/Sger_300x10000_(1_10)-12 966µs ± 0% 179µs ± 0% −81.45%
BlasGer/Sger_300x10000_(2_1)-12 592µs ± 0% 148µs ± 0% −75.03%
BlasGer/Sger_300x10000_(2_2)-12 726µs ± 0% 196µs ± 0% −72.95%
BlasGer/Sger_300x10000_(2_3)-12 749µs ± 0% 300µs ± 0% −60.00%
BlasGer/Sger_300x10000_(2_4)-12 796µs ± 0% 266µs ± 0% −66.63%
BlasGer/Sger_300x10000_(2_10)-12 1.13ms ± 0% 0.24ms ± 0% −79.20%
BlasGer/Sger_300x10000_(3_1)-12 674µs ± 0% 165µs ± 0% −75.53%
BlasGer/Sger_300x10000_(3_2)-12 741µs ± 0% 161µs ± 0% −78.24%
BlasGer/Sger_300x10000_(3_3)-12 814µs ± 0% 225µs ± 0% −72.42%
BlasGer/Sger_300x10000_(3_4)-12 765µs ± 0% 254µs ± 0% −66.77%
BlasGer/Sger_300x10000_(3_10)-12 959µs ± 0% 433µs ± 0% −54.80%
BlasGer/Sger_300x10000_(4_1)-12 701µs ± 0% 240µs ± 0% −65.78%
BlasGer/Sger_300x10000_(4_2)-12 715µs ± 0% 295µs ± 0% −58.83%
BlasGer/Sger_300x10000_(4_3)-12 715µs ± 0% 167µs ± 0% −76.59%
BlasGer/Sger_300x10000_(4_4)-12 853µs ± 0% 199µs ± 0% −76.69%
BlasGer/Sger_300x10000_(4_10)-12 989µs ± 0% 188µs ± 0% −80.98%
BlasGer/Sger_300x10000_(10_1)-12 672µs ± 0% 159µs ± 0% −76.42%
BlasGer/Sger_300x10000_(10_2)-12 720µs ± 0% 218µs ± 0% −69.66%
BlasGer/Sger_300x10000_(10_3)-12 726µs ± 0% 185µs ± 0% −74.54%
BlasGer/Sger_300x10000_(10_4)-12 823µs ± 0% 184µs ± 0% −77.68%
BlasGer/Sger_300x10000_(10_10)-12 990µs ± 0% 212µs ± 0% −78.64%
BlasGer/Sger_1000x30_(1_1)-12 4.81µs ± 0% 6.90µs ± 0% +43.53%
BlasGer/Sger_1000x30_(1_2)-12 5.40µs ± 0% 7.09µs ± 0% +31.33%
BlasGer/Sger_1000x30_(1_3)-12 6.06µs ± 0% 6.33µs ± 0% +4.42%
BlasGer/Sger_1000x30_(1_4)-12 5.87µs ± 0% 6.80µs ± 0% +15.80%
BlasGer/Sger_1000x30_(1_10)-12 5.68µs ± 0% 6.01µs ± 0% +5.79%
BlasGer/Sger_1000x30_(2_1)-12 5.96µs ± 0% 7.20µs ± 0% +20.72%
BlasGer/Sger_1000x30_(2_2)-12 4.66µs ± 0% 6.09µs ± 0% +30.68%
BlasGer/Sger_1000x30_(2_3)-12 5.26µs ± 0% 5.94µs ± 0% +12.83%
BlasGer/Sger_1000x30_(2_4)-12 4.96µs ± 0% 6.26µs ± 0% +26.20%
BlasGer/Sger_1000x30_(2_10)-12 5.85µs ± 0% 6.58µs ± 0% +12.47%
BlasGer/Sger_1000x30_(3_1)-12 5.56µs ± 0% 5.65µs ± 0% +1.73%
BlasGer/Sger_1000x30_(3_2)-12 5.75µs ± 0% 6.20µs ± 0% +7.70%
BlasGer/Sger_1000x30_(3_3)-12 5.87µs ± 0% 6.09µs ± 0% +3.76%
BlasGer/Sger_1000x30_(3_4)-12 5.38µs ± 0% 5.86µs ± 0% +8.92%
BlasGer/Sger_1000x30_(3_10)-12 5.56µs ± 0% 6.17µs ± 0% +10.84%
BlasGer/Sger_1000x30_(4_1)-12 5.83µs ± 0% 7.19µs ± 0% +23.40%
BlasGer/Sger_1000x30_(4_2)-12 5.29µs ± 0% 6.49µs ± 0% +22.69%
BlasGer/Sger_1000x30_(4_3)-12 5.50µs ± 0% 6.49µs ± 0% +18.03%
BlasGer/Sger_1000x30_(4_4)-12 5.02µs ± 0% 5.91µs ± 0% +17.77%
BlasGer/Sger_1000x30_(4_10)-12 5.92µs ± 0% 5.94µs ± 0% +0.25%
BlasGer/Sger_1000x30_(10_1)-12 5.80µs ± 0% 5.86µs ± 0% +1.09%
BlasGer/Sger_1000x30_(10_2)-12 5.97µs ± 0% 5.84µs ± 0% −2.13%
BlasGer/Sger_1000x30_(10_3)-12 5.41µs ± 0% 5.85µs ± 0% +8.19%
BlasGer/Sger_1000x30_(10_4)-12 6.41µs ± 0% 5.87µs ± 0% −8.47%
BlasGer/Sger_1000x30_(10_10)-12 6.20µs ± 0% 5.88µs ± 0% −5.13%
BlasGer/Sger_1000x100_(1_1)-12 18.6µs ± 0% 4.8µs ± 0% −73.90%
BlasGer/Sger_1000x100_(1_2)-12 19.9µs ± 0% 4.2µs ± 0% −78.98%
BlasGer/Sger_1000x100_(1_3)-12 23.9µs ± 0% 7.5µs ± 0% −68.53%
BlasGer/Sger_1000x100_(1_4)-12 21.0µs ± 0% 10.3µs ± 0% −50.76%
BlasGer/Sger_1000x100_(1_10)-12 24.3µs ± 0% 10.2µs ± 0% −57.83%
BlasGer/Sger_1000x100_(2_1)-12 20.0µs ± 0% 8.8µs ± 0% −56.35%
BlasGer/Sger_1000x100_(2_2)-12 20.2µs ± 0% 9.6µs ± 0% −52.62%
BlasGer/Sger_1000x100_(2_3)-12 18.0µs ± 0% 5.8µs ± 0% −67.86%
BlasGer/Sger_1000x100_(2_4)-12 20.2µs ± 0% 4.3µs ± 0% −78.57%
BlasGer/Sger_1000x100_(2_10)-12 20.6µs ± 0% 5.4µs ± 0% −73.64%
BlasGer/Sger_1000x100_(3_1)-12 17.9µs ± 0% 4.8µs ± 0% −73.04%
BlasGer/Sger_1000x100_(3_2)-12 19.8µs ± 0% 4.8µs ± 0% −75.62%
BlasGer/Sger_1000x100_(3_3)-12 19.2µs ± 0% 4.9µs ± 0% −74.66%
BlasGer/Sger_1000x100_(3_4)-12 18.7µs ± 0% 4.4µs ± 0% −76.73%
BlasGer/Sger_1000x100_(3_10)-12 18.4µs ± 0% 4.3µs ± 0% −76.54%
BlasGer/Sger_1000x100_(4_1)-12 20.6µs ± 0% 4.8µs ± 0% −76.90%
BlasGer/Sger_1000x100_(4_2)-12 19.9µs ± 0% 4.3µs ± 0% −78.34%
BlasGer/Sger_1000x100_(4_3)-12 20.8µs ± 0% 4.9µs ± 0% −76.41%
BlasGer/Sger_1000x100_(4_4)-12 20.9µs ± 0% 4.6µs ± 0% −78.01%
BlasGer/Sger_1000x100_(4_10)-12 24.4µs ± 0% 5.1µs ± 0% −79.17%
BlasGer/Sger_1000x100_(10_1)-12 24.9µs ± 0% 4.5µs ± 0% −81.74%
BlasGer/Sger_1000x100_(10_2)-12 19.7µs ± 0% 4.9µs ± 0% −75.39%
BlasGer/Sger_1000x100_(10_3)-12 22.2µs ± 0% 4.8µs ± 0% −78.42%
BlasGer/Sger_1000x100_(10_4)-12 20.2µs ± 0% 4.9µs ± 0% −75.56%
BlasGer/Sger_1000x100_(10_10)-12 20.1µs ± 0% 5.3µs ± 0% −73.69%
BlasGer/Sger_1000x300_(1_1)-12 65.2µs ± 0% 9.9µs ± 0% −84.87%
BlasGer/Sger_1000x300_(1_2)-12 54.3µs ± 0% 11.2µs ± 0% −79.46%
BlasGer/Sger_1000x300_(1_3)-12 67.4µs ± 0% 12.0µs ± 0% −82.25%
BlasGer/Sger_1000x300_(1_4)-12 66.3µs ± 0% 12.7µs ± 0% −80.89%
BlasGer/Sger_1000x300_(1_10)-12 62.1µs ± 0% 10.5µs ± 0% −83.11%
BlasGer/Sger_1000x300_(2_1)-12 67.7µs ± 0% 10.8µs ± 0% −84.08%
BlasGer/Sger_1000x300_(2_2)-12 67.9µs ± 0% 10.7µs ± 0% −84.27%
BlasGer/Sger_1000x300_(2_3)-12 52.7µs ± 0% 11.2µs ± 0% −78.71%
BlasGer/Sger_1000x300_(2_4)-12 62.2µs ± 0% 10.6µs ± 0% −83.02%
BlasGer/Sger_1000x300_(2_10)-12 64.5µs ± 0% 12.4µs ± 0% −80.77%
BlasGer/Sger_1000x300_(3_1)-12 64.3µs ± 0% 10.9µs ± 0% −83.05%
BlasGer/Sger_1000x300_(3_2)-12 66.3µs ± 0% 10.5µs ± 0% −84.20%
BlasGer/Sger_1000x300_(3_3)-12 56.5µs ± 0% 11.4µs ± 0% −79.80%
BlasGer/Sger_1000x300_(3_4)-12 53.4µs ± 0% 10.6µs ± 0% −80.07%
BlasGer/Sger_1000x300_(3_10)-12 58.5µs ± 0% 11.0µs ± 0% −81.22%
BlasGer/Sger_1000x300_(4_1)-12 54.2µs ± 0% 11.8µs ± 0% −78.27%
BlasGer/Sger_1000x300_(4_2)-12 54.9µs ± 0% 10.8µs ± 0% −80.40%
BlasGer/Sger_1000x300_(4_3)-12 54.2µs ± 0% 10.1µs ± 0% −81.44%
BlasGer/Sger_1000x300_(4_4)-12 73.2µs ± 0% 11.0µs ± 0% −85.03%
BlasGer/Sger_1000x300_(4_10)-12 65.6µs ± 0% 11.3µs ± 0% −82.72%
BlasGer/Sger_1000x300_(10_1)-12 57.6µs ± 0% 19.8µs ± 0% −65.65%
BlasGer/Sger_1000x300_(10_2)-12 64.6µs ± 0% 26.1µs ± 0% −59.60%
BlasGer/Sger_1000x300_(10_3)-12 68.1µs ± 0% 27.6µs ± 0% −59.45%
BlasGer/Sger_1000x300_(10_4)-12 58.9µs ± 0% 15.4µs ± 0% −73.94%
BlasGer/Sger_1000x300_(10_10)-12 64.5µs ± 0% 11.1µs ± 0% −82.82%
BlasGer/Sger_1000x1000_(1_1)-12 214µs ± 0% 40µs ± 0% −81.21%
BlasGer/Sger_1000x1000_(1_2)-12 205µs ± 0% 42µs ± 0% −79.73%
BlasGer/Sger_1000x1000_(1_3)-12 211µs ± 0% 41µs ± 0% −80.47%
BlasGer/Sger_1000x1000_(1_4)-12 200µs ± 0% 39µs ± 0% −80.72%
BlasGer/Sger_1000x1000_(1_10)-12 199µs ± 0% 38µs ± 0% −80.79%
BlasGer/Sger_1000x1000_(2_1)-12 231µs ± 0% 40µs ± 0% −82.74%
BlasGer/Sger_1000x1000_(2_2)-12 191µs ± 0% 43µs ± 0% −77.53%
BlasGer/Sger_1000x1000_(2_3)-12 234µs ± 0% 48µs ± 0% −79.49%
BlasGer/Sger_1000x1000_(2_4)-12 237µs ± 0% 37µs ± 0% −84.33%
BlasGer/Sger_1000x1000_(2_10)-12 268µs ± 0% 40µs ± 0% −84.93%
BlasGer/Sger_1000x1000_(3_1)-12 210µs ± 0% 39µs ± 0% −81.24%
BlasGer/Sger_1000x1000_(3_2)-12 194µs ± 0% 37µs ± 0% −81.16%
BlasGer/Sger_1000x1000_(3_3)-12 235µs ± 0% 41µs ± 0% −82.40%
BlasGer/Sger_1000x1000_(3_4)-12 229µs ± 0% 41µs ± 0% −82.05%
BlasGer/Sger_1000x1000_(3_10)-12 224µs ± 0% 40µs ± 0% −81.91%
BlasGer/Sger_1000x1000_(4_1)-12 212µs ± 0% 38µs ± 0% −81.91%
BlasGer/Sger_1000x1000_(4_2)-12 253µs ± 0% 41µs ± 0% −83.63%
BlasGer/Sger_1000x1000_(4_3)-12 222µs ± 0% 38µs ± 0% −82.80%
BlasGer/Sger_1000x1000_(4_4)-12 234µs ± 0% 43µs ± 0% −81.65%
BlasGer/Sger_1000x1000_(4_10)-12 221µs ± 0% 45µs ± 0% −79.56%
BlasGer/Sger_1000x1000_(10_1)-12 220µs ± 0% 37µs ± 0% −83.12%
BlasGer/Sger_1000x1000_(10_2)-12 213µs ± 0% 43µs ± 0% −79.96%
BlasGer/Sger_1000x1000_(10_3)-12 226µs ± 0% 40µs ± 0% −82.35%
BlasGer/Sger_1000x1000_(10_4)-12 250µs ± 0% 42µs ± 0% −83.32%
BlasGer/Sger_1000x1000_(10_10)-12 249µs ± 0% 39µs ± 0% −84.16%
BlasGer/Sger_1000x10000_(1_1)-12 3.70ms ± 0% 2.25ms ± 0% −39.33%
BlasGer/Sger_1000x10000_(1_2)-12 3.74ms ± 0% 2.71ms ± 0% −27.41%
BlasGer/Sger_1000x10000_(1_3)-12 3.95ms ± 0% 2.50ms ± 0% −36.78%
BlasGer/Sger_1000x10000_(1_4)-12 3.68ms ± 0% 2.43ms ± 0% −33.95%
BlasGer/Sger_1000x10000_(1_10)-12 4.89ms ± 0% 2.37ms ± 0% −51.44%
BlasGer/Sger_1000x10000_(2_1)-12 3.40ms ± 0% 2.15ms ± 0% −36.58%
BlasGer/Sger_1000x10000_(2_2)-12 3.54ms ± 0% 2.38ms ± 0% −32.71%
BlasGer/Sger_1000x10000_(2_3)-12 3.97ms ± 0% 2.34ms ± 0% −41.08%
BlasGer/Sger_1000x10000_(2_4)-12 3.84ms ± 0% 2.17ms ± 0% −43.52%
BlasGer/Sger_1000x10000_(2_10)-12 4.77ms ± 0% 2.33ms ± 0% −51.24%
BlasGer/Sger_1000x10000_(3_1)-12 3.40ms ± 0% 2.21ms ± 0% −35.15%
BlasGer/Sger_1000x10000_(3_2)-12 3.56ms ± 0% 2.20ms ± 0% −38.39%
BlasGer/Sger_1000x10000_(3_3)-12 3.86ms ± 0% 2.34ms ± 0% −39.26%
BlasGer/Sger_1000x10000_(3_4)-12 4.03ms ± 0% 2.24ms ± 0% −44.44%
BlasGer/Sger_1000x10000_(3_10)-12 4.85ms ± 0% 2.33ms ± 0% −52.10%
BlasGer/Sger_1000x10000_(4_1)-12 3.80ms ± 0% 2.50ms ± 0% −34.32%
BlasGer/Sger_1000x10000_(4_2)-12 3.66ms ± 0% 2.27ms ± 0% −38.08%
BlasGer/Sger_1000x10000_(4_3)-12 3.84ms ± 0% 2.28ms ± 0% −40.59%
BlasGer/Sger_1000x10000_(4_4)-12 3.78ms ± 0% 2.68ms ± 0% −29.07%
BlasGer/Sger_1000x10000_(4_10)-12 4.88ms ± 0% 3.01ms ± 0% −38.27%
BlasGer/Sger_1000x10000_(10_1)-12 3.65ms ± 0% 2.56ms ± 0% −30.02%
BlasGer/Sger_1000x10000_(10_2)-12 3.68ms ± 0% 2.35ms ± 0% −36.04%
BlasGer/Sger_1000x10000_(10_3)-12 3.66ms ± 0% 2.62ms ± 0% −28.58%
BlasGer/Sger_1000x10000_(10_4)-12 3.67ms ± 0% 3.13ms ± 0% −14.78%
BlasGer/Sger_1000x10000_(10_10)-12 4.64ms ± 0% 3.10ms ± 0% −33.17%
BlasGer/Sger_10000x300_(1_1)-12 789µs ± 0% 291µs ± 0% −63.08%
BlasGer/Sger_10000x300_(1_2)-12 1.12ms ± 0% 0.20ms ± 0% −81.82%
BlasGer/Sger_10000x300_(1_3)-12 864µs ± 0% 210µs ± 0% −75.73%
BlasGer/Sger_10000x300_(1_4)-12 846µs ± 0% 146µs ± 0% −82.75%
BlasGer/Sger_10000x300_(1_10)-12 899µs ± 0% 138µs ± 0% −84.69%
BlasGer/Sger_10000x300_(2_1)-12 903µs ± 0% 159µs ± 0% −82.34%
BlasGer/Sger_10000x300_(2_2)-12 744µs ± 0% 132µs ± 0% −82.21%
BlasGer/Sger_10000x300_(2_3)-12 745µs ± 0% 163µs ± 0% −78.13%
BlasGer/Sger_10000x300_(2_4)-12 957µs ± 0% 171µs ± 0% −82.17%
BlasGer/Sger_10000x300_(2_10)-12 931µs ± 0% 148µs ± 0% −84.05%
BlasGer/Sger_10000x300_(3_1)-12 1.01ms ± 0% 0.12ms ± 0% −87.99%
BlasGer/Sger_10000x300_(3_2)-12 835µs ± 0% 141µs ± 0% −83.09%
BlasGer/Sger_10000x300_(3_3)-12 964µs ± 0% 132µs ± 0% −86.35%
BlasGer/Sger_10000x300_(3_4)-12 951µs ± 0% 127µs ± 0% −86.62%
BlasGer/Sger_10000x300_(3_10)-12 863µs ± 0% 152µs ± 0% −82.43%
BlasGer/Sger_10000x300_(4_1)-12 855µs ± 0% 142µs ± 0% −83.39%
BlasGer/Sger_10000x300_(4_2)-12 875µs ± 0% 127µs ± 0% −85.44%
BlasGer/Sger_10000x300_(4_3)-12 905µs ± 0% 129µs ± 0% −85.71%
BlasGer/Sger_10000x300_(4_4)-12 851µs ± 0% 123µs ± 0% −85.56%
BlasGer/Sger_10000x300_(4_10)-12 812µs ± 0% 132µs ± 0% −83.71%
BlasGer/Sger_10000x300_(10_1)-12 963µs ± 0% 138µs ± 0% −85.62%
BlasGer/Sger_10000x300_(10_2)-12 983µs ± 0% 140µs ± 0% −85.80%
BlasGer/Sger_10000x300_(10_3)-12 958µs ± 0% 133µs ± 0% −86.11%
BlasGer/Sger_10000x300_(10_4)-12 1.04ms ± 0% 0.13ms ± 0% −87.17%
BlasGer/Sger_10000x300_(10_10)-12 798µs ± 0% 131µs ± 0% −83.60%
BlasGer/Sger_10000x1000_(1_1)-12 3.90ms ± 0% 2.17ms ± 0% −44.28%
BlasGer/Sger_10000x1000_(1_2)-12 3.75ms ± 0% 2.26ms ± 0% −39.76%
BlasGer/Sger_10000x1000_(1_3)-12 3.72ms ± 0% 2.12ms ± 0% −43.05%
BlasGer/Sger_10000x1000_(1_4)-12 3.59ms ± 0% 2.25ms ± 0% −37.32%
BlasGer/Sger_10000x1000_(1_10)-12 4.11ms ± 0% 2.31ms ± 0% −43.78%
BlasGer/Sger_10000x1000_(2_1)-12 3.86ms ± 0% 2.10ms ± 0% −45.60%
BlasGer/Sger_10000x1000_(2_2)-12 3.92ms ± 0% 2.11ms ± 0% −46.27%
BlasGer/Sger_10000x1000_(2_3)-12 3.97ms ± 0% 2.17ms ± 0% −45.24%
BlasGer/Sger_10000x1000_(2_4)-12 4.00ms ± 0% 2.77ms ± 0% −30.68%
BlasGer/Sger_10000x1000_(2_10)-12 3.93ms ± 0% 2.58ms ± 0% −34.44%
BlasGer/Sger_10000x1000_(3_1)-12 3.65ms ± 0% 2.95ms ± 0% −19.30%
BlasGer/Sger_10000x1000_(3_2)-12 3.56ms ± 0% 2.43ms ± 0% −31.77%
BlasGer/Sger_10000x1000_(3_3)-12 3.99ms ± 0% 2.56ms ± 0% −35.79%
BlasGer/Sger_10000x1000_(3_4)-12 3.88ms ± 0% 2.53ms ± 0% −34.83%
BlasGer/Sger_10000x1000_(3_10)-12 3.68ms ± 0% 2.20ms ± 0% −40.41%
BlasGer/Sger_10000x1000_(4_1)-12 3.94ms ± 0% 2.24ms ± 0% −43.04%
BlasGer/Sger_10000x1000_(4_2)-12 3.58ms ± 0% 2.27ms ± 0% −36.71%
BlasGer/Sger_10000x1000_(4_3)-12 3.93ms ± 0% 2.31ms ± 0% −41.13%
BlasGer/Sger_10000x1000_(4_4)-12 3.85ms ± 0% 2.54ms ± 0% −33.99%
BlasGer/Sger_10000x1000_(4_10)-12 3.69ms ± 0% 2.48ms ± 0% −32.76%
BlasGer/Sger_10000x1000_(10_1)-12 3.69ms ± 0% 2.17ms ± 0% −41.21%
BlasGer/Sger_10000x1000_(10_2)-12 3.91ms ± 0% 2.15ms ± 0% −45.08%
BlasGer/Sger_10000x1000_(10_3)-12 3.62ms ± 0% 2.15ms ± 0% −40.65%
BlasGer/Sger_10000x1000_(10_4)-12 3.97ms ± 0% 2.49ms ± 0% −37.32%
BlasGer/Sger_10000x1000_(10_10)-12 4.10ms ± 0% 2.49ms ± 0% −39.14%
BlasGer/Sger_10000x10000_(1_1)-12 34.6ms ± 0% 18.7ms ± 0% −45.88%
BlasGer/Sger_10000x10000_(1_2)-12 36.5ms ± 0% 19.0ms ± 0% −48.09%
BlasGer/Sger_10000x10000_(1_3)-12 36.1ms ± 0% 17.6ms ± 0% −51.35%
BlasGer/Sger_10000x10000_(1_4)-12 39.5ms ± 0% 17.9ms ± 0% −54.58%
BlasGer/Sger_10000x10000_(1_10)-12 45.0ms ± 0% 21.2ms ± 0% −52.87%
BlasGer/Sger_10000x10000_(2_1)-12 34.4ms ± 0% 20.1ms ± 0% −41.47%
BlasGer/Sger_10000x10000_(2_2)-12 36.0ms ± 0% 18.3ms ± 0% −49.20%
BlasGer/Sger_10000x10000_(2_3)-12 36.1ms ± 0% 18.5ms ± 0% −48.64%
BlasGer/Sger_10000x10000_(2_4)-12 37.3ms ± 0% 18.0ms ± 0% −51.70%
BlasGer/Sger_10000x10000_(2_10)-12 45.2ms ± 0% 18.3ms ± 0% −59.44%
BlasGer/Sger_10000x10000_(3_1)-12 34.4ms ± 0% 18.6ms ± 0% −46.10%
BlasGer/Sger_10000x10000_(3_2)-12 35.1ms ± 0% 17.7ms ± 0% −49.47%
BlasGer/Sger_10000x10000_(3_3)-12 35.6ms ± 0% 18.3ms ± 0% −48.52%
BlasGer/Sger_10000x10000_(3_4)-12 37.3ms ± 0% 17.9ms ± 0% −52.08%
BlasGer/Sger_10000x10000_(3_10)-12 45.3ms ± 0% 17.7ms ± 0% −60.96%
BlasGer/Sger_10000x10000_(4_1)-12 34.2ms ± 0% 21.3ms ± 0% −37.80%
BlasGer/Sger_10000x10000_(4_2)-12 36.1ms ± 0% 22.7ms ± 0% −37.07%
BlasGer/Sger_10000x10000_(4_3)-12 36.0ms ± 0% 22.6ms ± 0% −37.34%
BlasGer/Sger_10000x10000_(4_4)-12 36.5ms ± 0% 18.7ms ± 0% −48.85%
BlasGer/Sger_10000x10000_(4_10)-12 44.8ms ± 0% 19.2ms ± 0% −57.18%
BlasGer/Sger_10000x10000_(10_1)-12 34.2ms ± 0% 19.5ms ± 0% −43.07%
BlasGer/Sger_10000x10000_(10_2)-12 35.0ms ± 0% 20.3ms ± 0% −41.87%
BlasGer/Sger_10000x10000_(10_3)-12 35.7ms ± 0% 20.1ms ± 0% −43.76%
BlasGer/Sger_10000x10000_(10_4)-12 36.9ms ± 0% 19.8ms ± 0% −46.40%
BlasGer/Sger_10000x10000_(10_10)-12 44.6ms ± 0% 19.7ms ± 0% −55.85%
BlasGer/Sger_10000x100000_(1_1)-12 350ms ± 0% 173ms ± 0% −50.55%
BlasGer/Sger_10000x100000_(1_2)-12 356ms ± 0% 178ms ± 0% −49.92%
BlasGer/Sger_10000x100000_(1_3)-12 374ms ± 0% 181ms ± 0% −51.68%
BlasGer/Sger_10000x100000_(1_4)-12 374ms ± 0% 186ms ± 0% −50.19%
BlasGer/Sger_10000x100000_(1_10)-12 449ms ± 0% 176ms ± 0% −60.72%
BlasGer/Sger_10000x100000_(2_1)-12 346ms ± 0% 183ms ± 0% −46.95%
BlasGer/Sger_10000x100000_(2_2)-12 356ms ± 0% 179ms ± 0% −49.61%
BlasGer/Sger_10000x100000_(2_3)-12 368ms ± 0% 175ms ± 0% −52.39%
BlasGer/Sger_10000x100000_(2_4)-12 375ms ± 0% 189ms ± 0% −49.61%
BlasGer/Sger_10000x100000_(2_10)-12 455ms ± 0% 172ms ± 0% −62.23%
BlasGer/Sger_10000x100000_(3_1)-12 348ms ± 0% 171ms ± 0% −50.99%
BlasGer/Sger_10000x100000_(3_2)-12 355ms ± 0% 172ms ± 0% −51.50%
BlasGer/Sger_10000x100000_(3_3)-12 370ms ± 0% 191ms ± 0% −48.41%
BlasGer/Sger_10000x100000_(3_4)-12 374ms ± 0% 181ms ± 0% −51.73%
BlasGer/Sger_10000x100000_(3_10)-12 448ms ± 0% 173ms ± 0% −61.38%
BlasGer/Sger_10000x100000_(4_1)-12 344ms ± 0% 187ms ± 0% −45.80%
BlasGer/Sger_10000x100000_(4_2)-12 364ms ± 0% 187ms ± 0% −48.49%
BlasGer/Sger_10000x100000_(4_3)-12 373ms ± 0% 203ms ± 0% −45.63%
BlasGer/Sger_10000x100000_(4_4)-12 376ms ± 0% 192ms ± 0% −48.96%
BlasGer/Sger_10000x100000_(4_10)-12 453ms ± 0% 195ms ± 0% −57.00%
BlasGer/Sger_10000x100000_(10_1)-12 344ms ± 0% 188ms ± 0% −45.45%
BlasGer/Sger_10000x100000_(10_2)-12 357ms ± 0% 194ms ± 0% −45.84%
BlasGer/Sger_10000x100000_(10_3)-12 368ms ± 0% 191ms ± 0% −48.15%
BlasGer/Sger_10000x100000_(10_4)-12 375ms ± 0% 186ms ± 0% −50.26%
BlasGer/Sger_10000x100000_(10_10)-12 456ms ± 0% 192ms ± 0% −57.82%
BlasGer/Sger_100000x10000_(1_1)-12 347ms ± 0% 178ms ± 0% −48.62%
BlasGer/Sger_100000x10000_(1_2)-12 349ms ± 0% 179ms ± 0% −48.73%
BlasGer/Sger_100000x10000_(1_3)-12 358ms ± 0% 183ms ± 0% −49.02%
BlasGer/Sger_100000x10000_(1_4)-12 374ms ± 0% 180ms ± 0% −52.04%
BlasGer/Sger_100000x10000_(1_10)-12 450ms ± 0% 183ms ± 0% −59.28%
BlasGer/Sger_100000x10000_(2_1)-12 344ms ± 0% 179ms ± 0% −48.12%
BlasGer/Sger_100000x10000_(2_2)-12 348ms ± 0% 184ms ± 0% −47.32%
BlasGer/Sger_100000x10000_(2_3)-12 363ms ± 0% 178ms ± 0% −50.98%
BlasGer/Sger_100000x10000_(2_4)-12 363ms ± 0% 189ms ± 0% −47.83%
BlasGer/Sger_100000x10000_(2_10)-12 454ms ± 0% 182ms ± 0% −59.88%
BlasGer/Sger_100000x10000_(3_1)-12 373ms ± 0% 178ms ± 0% −52.12%
BlasGer/Sger_100000x10000_(3_2)-12 353ms ± 0% 185ms ± 0% −47.66%
BlasGer/Sger_100000x10000_(3_3)-12 358ms ± 0% 183ms ± 0% −48.91%
BlasGer/Sger_100000x10000_(3_4)-12 366ms ± 0% 178ms ± 0% −51.30%
BlasGer/Sger_100000x10000_(3_10)-12 451ms ± 0% 187ms ± 0% −58.67%
BlasGer/Sger_100000x10000_(4_1)-12 341ms ± 0% 178ms ± 0% −47.71%
BlasGer/Sger_100000x10000_(4_2)-12 354ms ± 0% 182ms ± 0% −48.73%
BlasGer/Sger_100000x10000_(4_3)-12 367ms ± 0% 185ms ± 0% −49.57%
BlasGer/Sger_100000x10000_(4_4)-12 364ms ± 0% 181ms ± 0% −50.42%
BlasGer/Sger_100000x10000_(4_10)-12 452ms ± 0% 181ms ± 0% −59.98%
BlasGer/Sger_100000x10000_(10_1)-12 345ms ± 0% 178ms ± 0% −48.50%
BlasGer/Sger_100000x10000_(10_2)-12 351ms ± 0% 181ms ± 0% −48.28%
BlasGer/Sger_100000x10000_(10_3)-12 362ms ± 0% 182ms ± 0% −49.64%
BlasGer/Sger_100000x10000_(10_4)-12 364ms ± 0% 177ms ± 0% −51.27%
BlasGer/Sger_100000x10000_(10_10)-12 449ms ± 0% 182ms ± 0% −59.59%
 

Merge request reports