## Dispatch cost (us per section, median over processes; the `load` rows are call time minus the 10 us of work in every share)
| team/hot | 0.76 | 0.77 | 0.56 | 3 |
| team/gap1ms | 5.25 | 6.75 | 6.42 | 3 |
| team/gap10ms | 6.08 | 5.33 | 7.17 | 3 |
| team/gap300ms | 27.71 | 27.42 | 26.62 | 3 |
| team/load10us-minus-10us | 3.03 | 2.22 | 1.40 | 3 |
| team/fork-hot | 1.37 | 0.83 | 0.63 | 3 |
| plain/outside-hot | 10.84 | 12.07 | 11.97 | 3 |
| plain/outside-gap1ms | 19.33 | 16.17 | 33.17 | 3 |
| plain/outside-gap10ms | 31.17 | 35.88 | 45.42 | 3 |
| plain/outside-load10us-minus-10us | 13.96 | 18.75 | 18.60 | 3 |
| external/outside-hot | 11.48 | 10.65 | 12.19 | 3 |
| external/outside-gap1ms | 18.29 | 20.04 | 21.25 | 3 |
| plain/in-pool | 0.03 | 0.11 | 0.20 | 3 |
| plain/in-pool-load10us-minus-10us | 6.08 | 10.80 | 5.56 | 3 |
## faer-off
### faer-off / gemm_sq
| 16x16x16 | 0.29 | 9.99 | 0.29 | 0.30 | 34.22 | 1.00 | 1.02 | 6 |
| 24x24x24 | 0.65 | 15.93 | 4.18 | 1.15 | 24.32 | 6.38 | 1.75 | 6 |
| 32x32x32 | 1.33 | 16.35 | 4.49 | 1.26 | 12.34 | 3.39 | 0.95 | 6 |
| 40x40x40 | 2.52 | 17.58 | 4.71 | 1.62 | 6.98 | 1.87 | 0.64 | 6 |
| 48x48x48 | 4.22 | 20.62 | 5.17 | 2.03 | 4.89 | 1.23 | 0.48 | 6 |
| 56x56x56 | 6.62 | 20.60 | 7.51 | 2.40 | 3.11 | 1.13 | 0.36 | 6 |
| 64x64x64 | 10.07 | 20.30 | 6.30 | 3.21 | 2.02 | 0.63 | 0.32 | 6 |
| 80x80x80 | 19.25 | 26.39 | 8.80 | 5.34 | 1.37 | 0.46 | 0.28 | 6 |
| 96x96x96 | 32.96 | 30.78 | 11.56 | 8.36 | 0.93 | 0.35 | 0.25 | 6 |
| 112x112x112 | 50.78 | 34.08 | 16.77 | 14.91 | 0.67 | 0.33 | 0.29 | 6 |
| 128x128x128 | 74.98 | 43.24 | 22.80 | 23.37 | 0.58 | 0.30 | 0.31 | 6 |
| 160x160x160 | 144 | 59.13 | 40.34 | 40.86 | 0.41 | 0.28 | 0.28 | 6 |
| 192x192x192 | 248 | 86.36 | 60.97 | 65.05 | 0.35 | 0.25 | 0.26 | 6 |
| 256x256x256 | 584 | 159 | 143 | 145 | 0.27 | 0.24 | 0.25 | 6 |
### faer-off / gemm_tall64
| 64x64x64 | 10.03 | 22.28 | 7.27 | 3.15 | 2.22 | 0.72 | 0.31 | 6 |
| 128x64x64 | 19.40 | 20.93 | 8.74 | 5.78 | 1.08 | 0.45 | 0.30 | 6 |
| 256x64x64 | 38.15 | 28.93 | 15.87 | 11.08 | 0.76 | 0.42 | 0.29 | 6 |
| 512x64x64 | 75.53 | 40.90 | 23.81 | 22.94 | 0.54 | 0.32 | 0.30 | 6 |
| 1024x64x64 | 151 | 64.64 | 42.50 | 40.62 | 0.43 | 0.28 | 0.27 | 6 |
| 2048x64x64 | 309 | 108 | 89.76 | 74.47 | 0.35 | 0.29 | 0.24 | 6 |
| 4096x64x64 | 623 | 211 | 202 | 142 | 0.34 | 0.32 | 0.23 | 6 |
| 8192x64x64 | 1267 | 429 | 419 | 279 | 0.34 | 0.33 | 0.22 | 6 |
### faer-off / gemm_k16
| 32x32x16 | 0.73 | 16.15 | 4.00 | 0.93 | 22.07 | 5.46 | 1.27 | 6 |
| 48x48x16 | 1.51 | 16.60 | 4.31 | 1.30 | 10.97 | 2.84 | 0.86 | 6 |
| 64x64x16 | 2.51 | 17.30 | 5.89 | 1.30 | 6.89 | 2.34 | 0.52 | 6 |
| 96x96x16 | 5.44 | 18.93 | 6.59 | 2.22 | 3.48 | 1.21 | 0.41 | 6 |
| 128x128x16 | 9.45 | 21.31 | 7.08 | 3.50 | 2.26 | 0.75 | 0.37 | 6 |
| 192x192x16 | 20.86 | 23.46 | 10.23 | 9.89 | 1.12 | 0.49 | 0.47 | 6 |
| 256x256x16 | 36.45 | 32.63 | 14.27 | 13.61 | 0.90 | 0.39 | 0.37 | 6 |
| 512x512x16 | 143 | 68.21 | 43.35 | 40.19 | 0.48 | 0.30 | 0.28 | 6 |
| 1024x1024x16 | 581 | 196 | 157 | 155 | 0.34 | 0.27 | 0.27 | 6 |
### faer-off / gemm_c64
| 16x16x16 | 0.62 | 10.88 | 0.63 | 0.68 | 17.47 | 1.01 | 1.10 | 6 |
| 24x24x24 | 2.25 | 16.87 | 4.94 | 1.56 | 7.50 | 2.20 | 0.69 | 6 |
| 32x32x32 | 5.30 | 18.26 | 5.39 | 2.04 | 3.45 | 1.02 | 0.38 | 6 |
| 48x48x48 | 18.79 | 21.91 | 8.93 | 5.03 | 1.17 | 0.48 | 0.27 | 6 |
| 64x64x64 | 45.00 | 32.59 | 13.53 | 11.29 | 0.72 | 0.30 | 0.25 | 6 |
| 96x96x96 | 146 | 57.89 | 36.77 | 35.02 | 0.40 | 0.25 | 0.24 | 6 |
| 128x128x128 | 346 | 95.83 | 77.84 | 83.66 | 0.28 | 0.22 | 0.24 | 6 |
### faer-off / lu_f64
| 32 | 2.17 | 12.56 | 2.15 | 2.28 | 5.79 | 0.99 | 1.05 | 6 |
| 48 | 5.25 | 35.37 | 25.12 | 8.61 | 6.73 | 4.78 | 1.64 | 6 |
| 64 | 10.24 | 53.04 | 43.51 | 15.31 | 5.18 | 4.25 | 1.50 | 6 |
| 96 | 26.74 | 109 | 101 | 37.90 | 4.09 | 3.77 | 1.42 | 6 |
| 128 | 53.55 | 184 | 162 | 61.23 | 3.44 | 3.02 | 1.14 | 6 |
| 192 | 148 | 348 | 330 | 147 | 2.34 | 2.22 | 0.99 | 6 |
| 256 | 320 | 564 | 515 | 242 | 1.76 | 1.61 | 0.76 | 6 |
| 384 | 915 | 998 | 916 | 502 | 1.09 | 1.00 | 0.55 | 6 |
| 512 | 2310 | 1875 | 1911 | 1095 | 0.81 | 0.83 | 0.47 | 6 |
| 768 | 6398 | 3440 | 3758 | 2662 | 0.54 | 0.59 | 0.42 | 6 |
### faer-off: break-even (first shape after which the mode stays below 0.95 x serial), with the serial time there
| gemm_sq | 96x96x96 (32.96 us) | 64x64x64 (10.07 us) | 40x40x40 (2.52 us) |
| gemm_tall64 | 256x64x64 (38.15 us) | 64x64x64 (10.03 us) | 64x64x64 (10.03 us) |
| gemm_k16 | 256x256x16 (36.45 us) | 128x128x16 (9.45 us) | 48x48x16 (1.51 us) |
| gemm_c64 | 64x64x64 (45.00 us) | 48x48x48 (18.79 us) | 24x24x24 (2.25 us) |
| lu_f64 | 512 (2310 us) | 512 (2310 us) | 256 (320 us) |
## faer-default
### faer-default / gemm_sq
| 16x16x16 | 0.29 | 9.70 | 0.30 | 0.29 | 33.05 | 1.02 | 1.00 | 2 |
| 24x24x24 | 0.66 | 10.32 | 0.66 | 0.66 | 15.76 | 1.01 | 1.01 | 2 |
| 32x32x32 | 1.32 | 11.59 | 1.32 | 1.32 | 8.76 | 1.00 | 0.99 | 2 |
| 40x40x40 | 2.51 | 12.41 | 2.51 | 2.51 | 4.94 | 1.00 | 1.00 | 2 |
| 48x48x48 | 4.19 | 14.24 | 4.22 | 4.20 | 3.40 | 1.01 | 1.00 | 2 |
| 56x56x56 | 6.67 | 15.84 | 6.55 | 6.57 | 2.38 | 0.98 | 0.99 | 2 |
| 64x64x64 | 10.32 | 16.31 | 9.95 | 10.06 | 1.58 | 0.96 | 0.97 | 2 |
| 80x80x80 | 19.60 | 26.88 | 19.10 | 19.13 | 1.37 | 0.97 | 0.98 | 2 |
| 96x96x96 | 32.96 | 31.36 | 11.84 | 6.99 | 0.95 | 0.36 | 0.21 | 2 |
| 112x112x112 | 51.00 | 36.34 | 16.90 | 13.30 | 0.71 | 0.33 | 0.26 | 2 |
| 128x128x128 | 74.88 | 43.02 | 25.75 | 20.98 | 0.57 | 0.34 | 0.28 | 2 |
| 160x160x160 | 145 | 63.10 | 41.75 | 37.55 | 0.43 | 0.29 | 0.26 | 2 |
| 192x192x192 | 247 | 99.27 | 65.51 | 57.72 | 0.40 | 0.26 | 0.23 | 2 |
| 256x256x256 | 582 | 166 | 143 | 134 | 0.29 | 0.25 | 0.23 | 2 |
### faer-default / gemm_tall64
| 64x64x64 | 9.96 | 18.89 | 10.17 | 10.67 | 1.90 | 1.02 | 1.07 | 2 |
| 128x64x64 | 19.31 | 27.84 | 19.60 | 20.54 | 1.44 | 1.02 | 1.06 | 2 |
| 256x64x64 | 37.73 | 32.76 | 16.41 | 9.46 | 0.87 | 0.43 | 0.25 | 2 |
| 512x64x64 | 74.93 | 43.89 | 25.12 | 20.20 | 0.59 | 0.34 | 0.27 | 2 |
| 1024x64x64 | 150 | 65.92 | 43.23 | 36.68 | 0.44 | 0.29 | 0.24 | 2 |
| 2048x64x64 | 303 | 107 | 87.63 | 69.71 | 0.35 | 0.29 | 0.23 | 2 |
| 4096x64x64 | 613 | 201 | 187 | 134 | 0.33 | 0.31 | 0.22 | 2 |
| 8192x64x64 | 1239 | 415 | 410 | 275 | 0.33 | 0.33 | 0.22 | 2 |
### faer-default / gemm_k16
| 32x32x16 | 0.72 | 10.00 | 0.73 | 0.76 | 13.81 | 1.01 | 1.05 | 2 |
| 48x48x16 | 1.51 | 10.62 | 1.52 | 1.58 | 7.04 | 1.01 | 1.05 | 2 |
| 64x64x16 | 2.51 | 11.88 | 2.55 | 2.53 | 4.74 | 1.02 | 1.01 | 2 |
| 96x96x16 | 5.40 | 14.72 | 5.48 | 5.39 | 2.73 | 1.02 | 1.00 | 2 |
| 128x128x16 | 9.36 | 15.55 | 9.39 | 9.34 | 1.66 | 1.00 | 1.00 | 2 |
| 192x192x16 | 20.66 | 22.83 | 9.41 | 8.98 | 1.10 | 0.46 | 0.43 | 2 |
| 256x256x16 | 36.30 | 32.43 | 12.88 | 12.37 | 0.89 | 0.35 | 0.34 | 2 |
| 512x512x16 | 142 | 74.76 | 40.63 | 36.64 | 0.53 | 0.29 | 0.26 | 2 |
| 1024x1024x16 | 574 | 190 | 140 | 140 | 0.33 | 0.24 | 0.24 | 2 |
### faer-default / gemm_c64
| 16x16x16 | 0.62 | 10.36 | 0.62 | 0.68 | 16.82 | 1.00 | 1.11 | 2 |
| 24x24x24 | 2.23 | 12.65 | 2.24 | 2.40 | 5.66 | 1.00 | 1.08 | 2 |
| 32x32x32 | 5.26 | 15.89 | 5.26 | 5.33 | 3.02 | 1.00 | 1.01 | 2 |
| 48x48x48 | 18.40 | 24.04 | 7.93 | 19.20 | 1.31 | 0.43 | 1.04 | 2 |
| 64x64x64 | 44.16 | 33.97 | 12.85 | 45.25 | 0.77 | 0.29 | 1.02 | 2 |
| 96x96x96 | 145 | 56.28 | 36.04 | 34.99 | 0.39 | 0.25 | 0.24 | 2 |
| 128x128x128 | 344 | 91.71 | 79.34 | 79.92 | 0.27 | 0.23 | 0.23 | 2 |
### faer-default / lu_f64
| 32 | 2.13 | 12.72 | 2.14 | 2.19 | 5.96 | 1.00 | 1.02 | 2 |
| 48 | 5.27 | 15.23 | 5.24 | 5.19 | 2.89 | 0.99 | 0.98 | 2 |
| 64 | 10.26 | 16.50 | 10.03 | 10.12 | 1.61 | 0.98 | 0.99 | 2 |
| 96 | 26.86 | 37.90 | 26.47 | 26.46 | 1.41 | 0.99 | 0.99 | 2 |
| 128 | 53.43 | 64.28 | 53.12 | 53.16 | 1.20 | 0.99 | 1.00 | 2 |
| 192 | 147 | 166 | 156 | 158 | 1.12 | 1.06 | 1.07 | 2 |
| 256 | 316 | 345 | 337 | 326 | 1.09 | 1.07 | 1.03 | 2 |
| 384 | 910 | 684 | 657 | 622 | 0.75 | 0.72 | 0.68 | 2 |
| 512 | 2280 | 1348 | 1329 | 1300 | 0.59 | 0.58 | 0.57 | 2 |
| 768 | 6324 | 2746 | 2844 | 2650 | 0.43 | 0.45 | 0.42 | 2 |
### faer-default: break-even (first shape after which the mode stays below 0.95 x serial), with the serial time there
| gemm_sq | 112x112x112 (51.00 us) | 96x96x96 (32.96 us) | 96x96x96 (32.96 us) |
| gemm_tall64 | 256x64x64 (37.73 us) | 256x64x64 (37.73 us) | 256x64x64 (37.73 us) |
| gemm_k16 | 256x256x16 (36.30 us) | 192x192x16 (20.66 us) | 192x192x16 (20.66 us) |
| gemm_c64 | 64x64x64 (44.16 us) | 48x48x48 (18.40 us) | 96x96x96 (145 us) |
| lu_f64 | 384 (910 us) | 384 (910 us) | 384 (910 us) |
## stack
### stack / copy
| 256 | 0.03 | 12.21 | 0.56 | 0.75 | 430.57 | 19.81 | 26.55 | 6 |
| 512 | 0.05 | 11.15 | 0.50 | 0.64 | 234.24 | 10.48 | 13.55 | 6 |
| 1024 | 0.10 | 11.29 | 0.66 | 1.35 | 114.61 | 6.67 | 13.71 | 6 |
| 2048 | 0.15 | 11.67 | 1.74 | 2.72 | 77.11 | 11.50 | 17.98 | 6 |
| 4096 | 0.27 | 13.80 | 4.54 | 3.22 | 50.28 | 16.52 | 11.73 | 6 |
| 8192 | 0.54 | 15.13 | 3.11 | 2.22 | 27.97 | 5.74 | 4.10 | 6 |
| 16384 | 1.07 | 15.39 | 3.57 | 2.08 | 14.42 | 3.35 | 1.95 | 6 |
| 32768 | 2.92 | 16.33 | 3.70 | 2.57 | 5.60 | 1.27 | 0.88 | 6 |
| 65536 | 5.82 | 17.59 | 5.26 | 3.66 | 3.03 | 0.90 | 0.63 | 6 |
| 131072 | 11.67 | 20.39 | 7.12 | 6.39 | 1.75 | 0.61 | 0.55 | 6 |
| 262144 | 25.06 | 24.39 | 15.10 | 11.19 | 0.97 | 0.60 | 0.45 | 6 |
| 524288 | 53.24 | 35.01 | 31.45 | 21.03 | 0.66 | 0.59 | 0.40 | 6 |
| 1048576 | 110 | 57.84 | 55.76 | 42.18 | 0.52 | 0.50 | 0.38 | 6 |
| 2097152 | 228 | 118 | 118 | 96.74 | 0.52 | 0.52 | 0.43 | 6 |
### stack / map
| 256 | 0.04 | 10.18 | 0.45 | 0.63 | 267.19 | 11.78 | 16.59 | 6 |
| 512 | 0.06 | 9.77 | 0.50 | 0.65 | 174.72 | 9.03 | 11.54 | 6 |
| 1024 | 0.10 | 10.25 | 0.60 | 0.77 | 102.86 | 6.06 | 7.70 | 6 |
| 2048 | 0.16 | 11.09 | 1.78 | 3.23 | 67.55 | 10.84 | 19.69 | 6 |
| 4096 | 0.30 | 13.69 | 4.41 | 3.30 | 45.56 | 14.69 | 11.00 | 6 |
| 8192 | 0.57 | 14.69 | 3.65 | 2.24 | 25.63 | 6.36 | 3.90 | 6 |
| 16384 | 1.27 | 15.26 | 3.90 | 2.50 | 12.05 | 3.08 | 1.97 | 6 |
| 32768 | 3.14 | 15.94 | 4.56 | 2.93 | 5.07 | 1.45 | 0.93 | 6 |
| 65536 | 6.35 | 17.18 | 5.31 | 3.80 | 2.71 | 0.84 | 0.60 | 6 |
| 131072 | 13.95 | 17.98 | 7.97 | 6.20 | 1.29 | 0.57 | 0.44 | 6 |
| 262144 | 33.82 | 24.09 | 15.51 | 11.17 | 0.71 | 0.46 | 0.33 | 6 |
| 524288 | 67.59 | 35.38 | 32.21 | 21.51 | 0.52 | 0.48 | 0.32 | 6 |
| 1048576 | 135 | 58.42 | 56.06 | 43.76 | 0.43 | 0.41 | 0.32 | 6 |
| 2097152 | 276 | 117 | 119 | 88.56 | 0.43 | 0.43 | 0.32 | 6 |
### stack / exp
| 256 | 0.37 | 11.11 | 1.73 | 0.72 | 29.81 | 4.63 | 1.94 | 6 |
| 512 | 0.73 | 11.03 | 2.23 | 1.36 | 15.14 | 3.06 | 1.86 | 6 |
| 1024 | 1.41 | 13.60 | 3.24 | 2.22 | 9.62 | 2.29 | 1.57 | 6 |
| 2048 | 2.84 | 14.68 | 5.27 | 5.13 | 5.17 | 1.86 | 1.80 | 6 |
| 4096 | 5.68 | 18.18 | 7.98 | 5.94 | 3.20 | 1.40 | 1.04 | 6 |
| 8192 | 11.46 | 18.41 | 6.72 | 6.07 | 1.61 | 0.59 | 0.53 | 6 |
| 16384 | 22.77 | 23.10 | 10.00 | 9.48 | 1.01 | 0.44 | 0.42 | 6 |
| 32768 | 46.20 | 28.28 | 23.91 | 16.95 | 0.61 | 0.52 | 0.37 | 6 |
| 65536 | 90.14 | 43.63 | 41.36 | 31.06 | 0.48 | 0.46 | 0.34 | 6 |
| 131072 | 180 | 72.76 | 70.81 | 58.49 | 0.40 | 0.39 | 0.32 | 6 |
| 262144 | 360 | 127 | 126 | 116 | 0.35 | 0.35 | 0.32 | 6 |
| 524288 | 719 | 231 | 229 | 226 | 0.32 | 0.32 | 0.31 | 6 |
| 1048576 | 1426 | 432 | 439 | 443 | 0.30 | 0.31 | 0.31 | 6 |
| 2097152 | 2831 | 865 | 879 | 892 | 0.31 | 0.31 | 0.32 | 6 |
### stack / sum
| 256 | 0.03 | 11.57 | 0.86 | 1.16 | 450.02 | 33.31 | 45.05 | 6 |
| 512 | 0.04 | 11.56 | 1.00 | 1.18 | 269.79 | 23.35 | 27.44 | 6 |
| 1024 | 0.08 | 11.59 | 1.21 | 1.12 | 142.11 | 14.85 | 13.73 | 6 |
| 2048 | 0.16 | 12.11 | 1.39 | 1.15 | 76.27 | 8.77 | 7.27 | 6 |
| 4096 | 0.32 | 12.27 | 1.68 | 1.19 | 38.34 | 5.23 | 3.71 | 6 |
| 8192 | 0.65 | 13.00 | 1.92 | 1.27 | 19.96 | 2.95 | 1.95 | 6 |
| 16384 | 1.33 | 13.84 | 2.84 | 1.39 | 10.44 | 2.15 | 1.05 | 6 |
| 32768 | 2.66 | 13.83 | 2.66 | 1.55 | 5.20 | 1.00 | 0.58 | 6 |
| 65536 | 5.35 | 14.59 | 3.11 | 2.22 | 2.72 | 0.58 | 0.42 | 6 |
| 131072 | 10.81 | 16.11 | 4.19 | 3.74 | 1.49 | 0.39 | 0.35 | 6 |
| 262144 | 21.36 | 19.28 | 7.02 | 5.97 | 0.90 | 0.33 | 0.28 | 6 |
| 524288 | 43.39 | 27.38 | 12.35 | 10.80 | 0.63 | 0.28 | 0.25 | 6 |
| 1048576 | 87.15 | 39.69 | 22.99 | 20.91 | 0.46 | 0.26 | 0.24 | 6 |
| 2097152 | 175 | 65.62 | 53.37 | 44.11 | 0.38 | 0.31 | 0.25 | 6 |
### stack / tc_off
| 8 | 0.08 | 9.93 | 0.11 | 7.87 | 126.31 | 1.38 | 100.02 | 6 |
| 12 | 0.16 | 9.72 | 0.19 | 7.78 | 60.75 | 1.19 | 48.65 | 6 |
| 16 | 0.32 | 9.59 | 0.35 | 8.08 | 30.06 | 1.09 | 25.31 | 6 |
| 24 | 0.70 | 10.13 | 0.73 | 8.23 | 14.43 | 1.03 | 11.71 | 6 |
| 32 | 1.38 | 11.08 | 1.39 | 9.26 | 8.05 | 1.01 | 6.73 | 6 |
| 48 | 4.34 | 14.03 | 4.33 | 15.52 | 3.23 | 1.00 | 3.57 | 6 |
| 64 | 10.23 | 16.45 | 10.21 | 19.38 | 1.61 | 1.00 | 1.89 | 6 |
| 96 | 33.16 | 30.23 | 13.22 | 32.71 | 0.91 | 0.40 | 0.99 | 6 |
| 128 | 74.97 | 42.04 | 26.26 | 42.25 | 0.56 | 0.35 | 0.56 | 6 |
| 192 | 251 | 83.24 | 66.27 | 88.54 | 0.33 | 0.26 | 0.35 | 6 |
| 256 | 582 | 164 | 143 | 172 | 0.28 | 0.25 | 0.30 | 6 |
| 384 | 1948 | 463 | 453 | 472 | 0.24 | 0.23 | 0.24 | 6 |
### stack / tc_default
| 8 | 0.08 | 0.08 | 0.08 | 0.08 | 1.01 | 1.01 | 1.08 | 6 |
| 12 | 0.16 | 0.16 | 0.16 | 0.18 | 0.99 | 1.01 | 1.11 | 6 |
| 16 | 0.32 | 0.32 | 0.32 | 0.35 | 1.00 | 1.01 | 1.10 | 6 |
| 24 | 0.70 | 0.70 | 0.69 | 0.70 | 1.01 | 1.00 | 1.00 | 6 |
| 32 | 1.36 | 1.38 | 1.37 | 1.37 | 1.01 | 1.01 | 1.01 | 6 |
| 48 | 4.29 | 4.30 | 4.29 | 4.30 | 1.00 | 1.00 | 1.00 | 6 |
| 64 | 10.22 | 10.17 | 10.21 | 10.10 | 0.99 | 1.00 | 0.99 | 6 |
| 96 | 32.84 | 30.92 | 22.31 | 21.72 | 0.94 | 0.68 | 0.66 | 6 |
| 128 | 75.67 | 44.86 | 27.94 | 41.85 | 0.59 | 0.37 | 0.55 | 6 |
| 192 | 250 | 83.29 | 65.72 | 93.64 | 0.33 | 0.26 | 0.37 | 6 |
| 256 | 587 | 165 | 145 | 167 | 0.28 | 0.25 | 0.28 | 6 |
| 384 | 1986 | 450 | 441 | 459 | 0.23 | 0.22 | 0.23 | 6 |
### stack: break-even (first shape after which the mode stays below 0.95 x serial), with the serial time there
| copy | 524288 (53.24 us) | 65536 (5.82 us) | 32768 (2.92 us) |
| map | 262144 (33.82 us) | 65536 (6.35 us) | 32768 (3.14 us) |
| exp | 32768 (46.20 us) | 8192 (11.46 us) | 8192 (11.46 us) |
| sum | 262144 (21.36 us) | 65536 (5.35 us) | 32768 (2.66 us) |
| tc_off | 96 (33.16 us) | 96 (33.16 us) | 128 (74.97 us) |
| tc_default | 96 (32.84 us) | 96 (32.84 us) | 96 (32.84 us) |
## lusweep
### lusweep / lu_pt16384_gf589824
| 128 | 54.50 | 56.04 | 55.38 | 1.03 | 1.02 | 4 |
| 192 | 151 | 161 | 164 | 1.06 | 1.08 | 4 |
| 256 | 326 | 357 | 334 | 1.09 | 1.02 | 4 |
| 384 | 938 | 778 | 584 | 0.83 | 0.62 | 4 |
| 512 | 2332 | 1494 | 1259 | 0.64 | 0.54 | 4 |
| 768 | 6601 | 3324 | 2518 | 0.50 | 0.38 | 4 |
### lusweep / lu_pt0_gf0
| 128 | 55.40 | 184 | 64.57 | 3.32 | 1.17 | 4 |
| 192 | 151 | 348 | 137 | 2.31 | 0.91 | 4 |
| 256 | 322 | 585 | 271 | 1.82 | 0.84 | 4 |
| 384 | 934 | 1040 | 557 | 1.11 | 0.60 | 4 |
| 512 | 2336 | 1934 | 1206 | 0.83 | 0.52 | 4 |
| 768 | 6455 | 3468 | 2624 | 0.54 | 0.41 | 4 |
### lusweep / lu_pt16384_gf0
| 128 | 55.74 | 54.04 | 55.64 | 0.97 | 1.00 | 4 |
| 192 | 156 | 160 | 164 | 1.03 | 1.05 | 4 |
| 256 | 332 | 346 | 345 | 1.04 | 1.04 | 4 |
| 384 | 940 | 705 | 633 | 0.75 | 0.67 | 4 |
| 512 | 2367 | 1560 | 1411 | 0.66 | 0.60 | 4 |
| 768 | 6683 | 3012 | 2799 | 0.45 | 0.42 | 4 |
### lusweep / lu_pt65536_gf0
| 128 | 54.91 | 53.84 | 56.02 | 0.98 | 1.02 | 4 |
| 192 | 152 | 149 | 152 | 0.98 | 1.00 | 4 |
| 256 | 329 | 320 | 332 | 0.97 | 1.01 | 4 |
| 384 | 964 | 955 | 1013 | 0.99 | 1.05 | 4 |
| 512 | 2366 | 2318 | 2520 | 0.98 | 1.06 | 4 |
| 768 | 6558 | 3548 | 3495 | 0.54 | 0.53 | 4 |
### lusweep / lu_pt262144_gf0
| 128 | 55.15 | 54.74 | 55.10 | 0.99 | 1.00 | 4 |
| 192 | 151 | 150 | 150 | 0.99 | 1.00 | 4 |
| 256 | 326 | 320 | 327 | 0.98 | 1.00 | 4 |
| 384 | 928 | 951 | 932 | 1.02 | 1.00 | 4 |
| 512 | 2299 | 2348 | 2334 | 1.02 | 1.02 | 4 |
| 768 | 6402 | 6588 | 6752 | 1.03 | 1.05 | 4 |
### lusweep / lu_pt65536_gf589824
| 128 | 53.94 | 54.74 | 54.98 | 1.01 | 1.02 | 4 |
| 192 | 149 | 149 | 150 | 1.00 | 1.01 | 4 |
| 256 | 322 | 321 | 324 | 1.00 | 1.01 | 4 |
| 384 | 944 | 941 | 962 | 1.00 | 1.02 | 4 |
| 512 | 2353 | 2346 | 2416 | 1.00 | 1.03 | 4 |
| 768 | 6445 | 3545 | 3512 | 0.55 | 0.54 | 4 |
### lusweep: break-even (first shape after which the mode stays below 0.95 x serial), with the serial time there
| lu_pt16384_gf589824 | 384 (938 us) | 384 (938 us) |
| lu_pt0_gf0 | 512 (2336 us) | 192 (151 us) |
| lu_pt16384_gf0 | 384 (940 us) | 384 (940 us) |
| lu_pt65536_gf0 | 768 (6558 us) | 768 (6558 us) |
| lu_pt262144_gf0 | none | none |
| lu_pt65536_gf589824 | 768 (6445 us) | 768 (6445 us) |
## Host state per process (busy = idle below 85 percent before or after; in-run samples include the benchmark's own 6 threads and are not used)
| dispatch-p1 | 87.65 | 9.1 | 86.24 | 41.3 WindowServer; 40.3 DeepSeek Harness Helper (Renderer); 17.7 DeepSeek Harness Helper | no |
| faer-off-p1 | 86.66 | 7.5 | 87.88 | 68.3 DeepSeek Harness Helper (Renderer); 30.6 WindowServer; 12.7 Claude Helper | no |
| stack-p1 | 86.6 | 7.1 | 85.65 | 43.8 DeepSeek Harness Helper (Renderer); 30.1 WindowServer; 20.8 sysmond | no |
| dispatch-p2 | 86.62 | 7.5 | 86.93 | 43.0 WindowServer; 40.2 DeepSeek Harness Helper (Renderer); 26.7 com.apple.WebKit.WebContent | no |
| faer-off-p2 | 85.79 | 7.8 | 86.88 | 61.6 mediaanalysisd; 40.0 DeepSeek Harness Helper (Renderer); 31.6 WindowServer | no |
| stack-p2 | 85.35 | 10.3 | 81.5 | 53.3 WindowServer; 38.0 DeepSeek Harness Helper (Renderer); 17.4 DeepSeek Harness Helper | **yes** |
| dispatch-p3 | 85.93 | 5.9 | 83.16 | 42.5 WindowServer; 39.1 DeepSeek Harness Helper (Renderer); 25.4 sysmond | **yes** |
| faer-off-p3 | 85.8 | 6.8 | 79.66 | 58.3 DeepSeek Harness Helper (Renderer); 54.7 WindowServer; 17.8 CursorUIViewService | **yes** |
| stack-p3 | 86.69 | 7.3 | 82.5 | 44.5 WindowServer; 38.7 DeepSeek Harness Helper (Renderer); 15.0 Claude Helper | **yes** |
| faer-default-p1 | 87.78 | 9.2 | 67.28 | 39.5 DeepSeek Harness Helper (Renderer); 31.7 WindowServer; 16.2 DeepSeek Harness Helper | **yes** |
| lusweep-p1 | 87.3 | 6.8 | 82.94 | 75.7 DeepSeek Harness Helper (Renderer); 43.2 WindowServer; 16.2 DeepSeek Harness Helper | **yes** |
| lusweep-p2 | 85.22 | 6.2 | 86.86 | 57.5 WindowServer; 52.7 com.apple.WebKit.WebContent; 50.5 DeepSeek Harness Helper (Renderer) | no |