cd $BASILISK/src/test/
CFLAGS='-DSINGLE_PRECISION -DBENCHMARK -fopenmp' make lid.ctst
cd lid
OpenGL renderer string: Intel Core i7
for i in 64 128 256 512 1024 2048; do OMP_NUM_THREADS=8 ./lid $i 2> /dev/null | grep steps; done
# Title: Inteli7
# Multigrid, 300 steps, 4.42842 CPU, 0.5561 real, 2.21e+06 points.step/s, 15 var
# Multigrid, 300 steps, 6.59194 CPU, 0.8264 real, 5.95e+06 points.step/s, 15 var
# Multigrid, 300 steps, 12.8195 CPU, 1.604 real, 1.23e+07 points.step/s, 15 var
# Multigrid, 300 steps, 41.0347 CPU, 5.131 real, 1.53e+07 points.step/s, 15 var
# Multigrid, 300 steps, 181.98 CPU, 22.75 real, 1.38e+07 points.step/s, 15 var
# Multigrid, 300 steps, 868.664 CPU, 108.6 real, 1.16e+07 points.step/s, 15 var
cd ..
CFLAGS=-DBENCHMARK make lid.gpu.tst
cd lid.gpu
OpenGL renderer string: Mesa Intel(R) UHD Graphics (TGL GT1) (0x9a60)
Video memory: 3072MB
for i in 64 128 256 512 1024 2048; do ./lid.gpu $i | grep steps; done
# Title: IntelHUD
# Multigrid (GPU), 300 steps, 0.77206 CPU, 1.645 real, 7.47e+05 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 0.873143 CPU, 2.017 real, 2.44e+06 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 1.36386 CPU, 4.26 real, 4.62e+06 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 3.32161 CPU, 12.2 real, 6.44e+06 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 6.25396 CPU, 40.75 real, 7.72e+06 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 17.7899 CPU, 182.2 real, 6.91e+06 points.step/s, 17 var
OpenGL renderer string: NVIDIA GeForce RTX 3050 Ti Laptop GPU/PCIe/SSE2
Dedicated video memory: 4096 MB
for i in 64 128 256 512 1024 2048; do nvidia ./lid.gpu $i | grep steps; done
# Title: RTX3050
# Multigrid (GPU), 300 steps, 0.540176 CPU, 0.5402 real, 2.27e+06 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 0.597448 CPU, 0.5975 real, 8.23e+06 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 0.931005 CPU, 0.931 real, 2.11e+07 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 1.99594 CPU, 2.002 real, 3.93e+07 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 6.48733 CPU, 6.488 real, 4.85e+07 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 24.9791 CPU, 24.98 real, 5.04e+07 points.step/s, 17 var
# Title: RTX3050(cuda)
# Multigrid (cuda), 300 steps, 0.803599 CPU, 0.8033 real, 1.53e+06 points.step/s, 17 var
# Multigrid (cuda), 300 steps, 0.948767 CPU, 0.9485 real, 5.18e+06 points.step/s, 17 var
# Multigrid (cuda), 300 steps, 1.26331 CPU, 1.263 real, 1.56e+07 points.step/s, 17 var
# Multigrid (cuda), 300 steps, 2.36983 CPU, 2.369 real, 3.32e+07 points.step/s, 17 var
# Multigrid (cuda), 300 steps, 6.25535 CPU, 6.253 real, 5.03e+07 points.step/s, 17 var
# Multigrid (cuda), 300 steps, 22.4501 CPU, 22.44 real, 5.61e+07 points.step/s, 17 var
OpenGL renderer string: Quadro RTX 6000/PCIe/SSE2
Dedicated video memory: 24576 MB
for i in 64 128 256 512 1024 2048; do ./lid.gpu $i | grep steps; done
# Title: RTX6000
# Multigrid (GPU), 300 steps, 0.427174 CPU, 0.4272 real, 2.88e+06 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 0.430609 CPU, 0.4306 real, 1.14e+07 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 0.600736 CPU, 0.6008 real, 3.27e+07 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 1.03794 CPU, 1.038 real, 7.58e+07 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 2.47181 CPU, 2.472 real, 1.27e+08 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 8.35382 CPU, 8.354 real, 1.51e+08 points.step/s, 17 var
OpenGL renderer string: NVIDIA GeForce RTX 4090/PCIe/SSE2
Dedicated video memory: 24564 MB
for i in 64 128 256 512 1024 2048; do ./lid.gpu $i | grep steps; done
# Title: RTX4090
# Multigrid (gpu), 300 steps, 0.838243 CPU, 0.8383 real, 1.47e+06 points.step/s, 17 var
# Multigrid (gpu), 300 steps, 0.853601 CPU, 0.8538 real, 5.76e+06 points.step/s, 17 var
# Multigrid (gpu), 300 steps, 1.19769 CPU, 1.198 real, 1.64e+07 points.step/s, 17 var
# Multigrid (gpu), 300 steps, 1.73587 CPU, 1.736 real, 4.53e+07 points.step/s, 17 var
# Multigrid (gpu), 300 steps, 2.73267 CPU, 2.733 real, 1.15e+08 points.step/s, 17 var
# Multigrid (gpu), 300 steps, 6.74105 CPU, 6.742 real, 1.87e+08 points.step/s, 17 var
# Title: RTX4090(cuda)
# Multigrid (cuda), 300 steps, 0.82454 CPU, 0.8244 real, 1.49e+06 points.step/s, 17 var
# Multigrid (cuda), 300 steps, 0.851484 CPU, 0.8513 real, 5.77e+06 points.step/s, 17 var
# Multigrid (cuda), 300 steps, 1.17222 CPU, 1.172 real, 1.68e+07 points.step/s, 17 var
# Multigrid (cuda), 300 steps, 1.58465 CPU, 1.584 real, 4.96e+07 points.step/s, 17 var
# Multigrid (cuda), 300 steps, 2.25285 CPU, 2.252 real, 1.4e+08 points.step/s, 17 var
# Multigrid (cuda), 300 steps, 5.088 CPU, 5.087 real, 2.47e+08 points.step/s, 17 var
OpenGL renderer string: NVIDIA GeForce RTX 4090 D/PCIe/SSE2 (stokes.lmm.jussieu.fr)
Dedicated video memory: 24564 MB
for i in 64 128 256 512 1024 2048; do ./lid.gpu $i | grep steps; done
# Title: RTX4090D
# Multigrid (GPU), 300 steps, 0.997049 CPU, 0.9971 real, 1.23e+06 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 1.14402 CPU, 1.144 real, 4.3e+06 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 1.60657 CPU, 1.607 real, 1.22e+07 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 2.2603 CPU, 2.26 real, 3.48e+07 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 3.4422 CPU, 3.442 real, 9.14e+07 points.step/s, 17 var
# Multigrid (GPU), 300 steps, 7.70925 CPU, 7.709 real, 1.63e+08 points.step/s, 17 var