cd $BASILISK/src/test/
CFLAGS=-fopenmp make advection.ctst
OMP_NUM_THREADS=8 ./advection/advection 2048 2> /dev/null | grep steps
OpenGL renderer string: Intel Core i7
# Title: Inteli7
# Multigrid, 412 steps, 0.275058 CPU, 0.03633 real, 4.65e+07 points.step/s, 7 var
# Multigrid, 796 steps, 0.928328 CPU, 0.1163 real, 1.12e+08 points.step/s, 7 var
# Multigrid, 1566 steps, 5.72297 CPU, 0.7265 real, 1.41e+08 points.step/s, 7 var
# Multigrid, 3100 steps, 40.9198 CPU, 5.116 real, 1.59e+08 points.step/s, 7 var
# Multigrid, 6163 steps, 538.689 CPU, 67.34 real, 9.6e+07 points.step/s, 7 var
# Multigrid, 12280 steps, 4756.62 CPU, 594.6 real, 8.66e+07 points.step/s, 7 var
make advection.gpu.tst
OpenGL renderer string: Mesa Intel(R) UHD Graphics (TGL GT1) (0x9a60)
Video memory: 3072MB
./advection.gpu/advection.gpu 2048 2> /dev/null | grep steps
# Title: IntelHUD
# Multigrid (GPU), 412 steps, 0.052281 CPU, 0.08927 real, 1.89e+07 points.step/s, 8 var
# Multigrid (GPU), 796 steps, 0.061498 CPU, 0.1933 real, 6.75e+07 points.step/s, 8 var
# Multigrid (GPU), 1566 steps, 0.174751 CPU, 0.9532 real, 1.08e+08 points.step/s, 8 var
# Multigrid (GPU), 3100 steps, 0.766669 CPU, 6.167 real, 1.32e+08 points.step/s, 8 var
# Multigrid (GPU), 6163 steps, 3.10575 CPU, 43.14 real, 1.5e+08 points.step/s, 8 var
# Multigrid (GPU), 12280 steps, 7.34859 CPU, 340 real, 1.51e+08 points.step/s, 8 var
OpenGL renderer string: NVIDIA GeForce RTX 3050 Ti Laptop GPU/PCIe/SSE2
Dedicated video memory: 4096 MB
__NV_PRIME_RENDER_OFFLOAD=1 __GLX_VENDOR_LIBRARY_NAME=nvidia ./advection.gpu/advection.gpu 2048 2> /dev/null | grep steps
# Title: RTX3050
# Multigrid (gpu), 412 steps, 0.042306 CPU, 0.04542 real, 3.72e+07 points.step/s, 8 var
# Multigrid (gpu), 799 steps, 0.07919 CPU, 0.08108 real, 1.61e+08 points.step/s, 8 var
# Multigrid (gpu), 1569 steps, 0.214872 CPU, 0.2262 real, 4.55e+08 points.step/s, 8 var
# Multigrid (gpu), 3102 steps, 0.740162 CPU, 0.7527 real, 1.08e+09 points.step/s, 8 var
# Multigrid (gpu), 6165 steps, 4.46272 CPU, 4.474 real, 1.45e+09 points.step/s, 8 var
# Multigrid (gpu), 12284 steps, 33.3466 CPU, 33.38 real, 1.54e+09 points.step/s, 8 var
./advection.cuda/advection.cuda 2048 2> /dev/null | grep steps
# Title: RTX3050(cuda)
# Multigrid (cuda), 412 steps, 0.295274 CPU, 0.2952 real, 5.72e+06 points.step/s, 8 var
# Multigrid (cuda), 799 steps, 0.051436 CPU, 0.05144 real, 2.55e+08 points.step/s, 8 var
# Multigrid (cuda), 1569 steps, 0.137294 CPU, 0.1372 real, 7.49e+08 points.step/s, 8 var
# Multigrid (cuda), 3102 steps, 0.631017 CPU, 0.6308 real, 1.29e+09 points.step/s, 8 var
# Multigrid (cuda), 6165 steps, 4.29459 CPU, 4.293 real, 1.51e+09 points.step/s, 8 var
# Multigrid (cuda), 12284 steps, 34.0705 CPU, 34.06 real, 1.51e+09 points.step/s, 8 var
OpenGL renderer string: Quadro RTX 6000/PCIe/SSE2
Dedicated video memory: 24576 MB
./advection.gpu/advection.gpu 2048 2> /dev/null | grep steps
# Title: RTX6000
# Multigrid (GPU), 412 steps, 0.031947 CPU, 0.03195 real, 5.28e+07 points.step/s, 8 var
# Multigrid (GPU), 796 steps, 0.049988 CPU, 0.04999 real, 2.61e+08 points.step/s, 8 var
# Multigrid (GPU), 1566 steps, 0.124451 CPU, 0.1245 real, 8.25e+08 points.step/s, 8 var
# Multigrid (GPU), 3100 steps, 0.289842 CPU, 0.2898 real, 2.8e+09 points.step/s, 8 var
# Multigrid (GPU), 6163 steps, 1.49181 CPU, 1.492 real, 4.33e+09 points.step/s, 8 var
# Multigrid (GPU), 12282 steps, 10.7582 CPU, 10.76 real, 4.79e+09 points.step/s, 8 var
OpenGL renderer string: NVIDIA GeForce RTX 4090/PCIe/SSE2
Dedicated video memory: 24564 MB
./advection.gpu/advection.gpu 2048 2> /dev/null | grep steps
# Title: RTX4090
# Multigrid (GPU), 412 steps, 0.039116 CPU, 0.03912 real, 4.31e+07 points.step/s, 8 var
# Multigrid (GPU), 799 steps, 0.055118 CPU, 0.05512 real, 2.38e+08 points.step/s, 8 var
# Multigrid (GPU), 1569 steps, 0.122464 CPU, 0.1225 real, 8.4e+08 points.step/s, 8 var
# Multigrid (GPU), 3102 steps, 0.248881 CPU, 0.2489 real, 3.27e+09 points.step/s, 8 var
# Multigrid (GPU), 6165 steps, 0.73287 CPU, 0.7329 real, 8.82e+09 points.step/s, 8 var
# Multigrid (GPU), 12284 steps, 4.92467 CPU, 4.925 real, 1.05e+10 points.step/s, 8 var
./advection.cuda/advection.cuda 2048 2> /dev/null | grep steps
# Title: RTX4090(cuda)
# Multigrid (cuda), 412 steps, 0.102746 CPU, 0.1027 real, 1.64e+07 points.step/s, 8 var
# Multigrid (cuda), 799 steps, 0.077778 CPU, 0.07776 real, 1.68e+08 points.step/s, 8 var
# Multigrid (cuda), 1569 steps, 0.171465 CPU, 0.1714 real, 6e+08 points.step/s, 8 var
# Multigrid (cuda), 3102 steps, 0.292987 CPU, 0.2929 real, 2.78e+09 points.step/s, 8 var
# Multigrid (cuda), 6165 steps, 0.611425 CPU, 0.6113 real, 1.06e+10 points.step/s, 8 var
# Multigrid (cuda), 12284 steps, 4.36559 CPU, 4.365 real, 1.18e+10 points.step/s, 8 var
/advection.ocl/advection.ocl 2048 2> /dev/null | grep steps
# Title: RTX4090(opencl)
# Multigrid (opencl), 412 steps, 0.06094 CPU, 0.06092 real, 2.77e+07 points.step/s, 8 var
# Multigrid (opencl), 799 steps, 0.106572 CPU, 0.1066 real, 1.23e+08 points.step/s, 8 var
# Multigrid (opencl), 1569 steps, 0.227728 CPU, 0.2277 real, 4.52e+08 points.step/s, 8 var
# Multigrid (opencl), 3102 steps, 0.426498 CPU, 0.4264 real, 1.91e+09 points.step/s, 8 var
# Multigrid (opencl), 6165 steps, 1.06909 CPU, 1.069 real, 6.05e+09 points.step/s, 8 var
# Multigrid (opencl), 12284 steps, 5.32657 CPU, 5.325 real, 9.67e+09 points.step/s, 8 var
OpenGL renderer string: NVIDIA GeForce RTX 4090 D/PCIe/SSE2 (stokes.lmm.jussieu.fr)
Dedicated video memory: 24564 MB
./advection.gpu/advection.gpu 2048 2> /dev/null | grep steps
# Title: RTX4090D
# Multigrid (GPU), 412 steps, 0.058459 CPU, 0.05846 real, 2.89e+07 points.step/s, 8 var
# Multigrid (GPU), 799 steps, 0.103384 CPU, 0.1034 real, 1.27e+08 points.step/s, 8 var
# Multigrid (GPU), 1569 steps, 0.231444 CPU, 0.2315 real, 4.44e+08 points.step/s, 8 var
# Multigrid (GPU), 3102 steps, 0.439395 CPU, 0.4394 real, 1.85e+09 points.step/s, 8 var
# Multigrid (GPU), 6165 steps, 1.11655 CPU, 1.117 real, 5.79e+09 points.step/s, 8 var
# Multigrid (GPU), 12284 steps, 5.46386 CPU, 5.464 real, 9.43e+09 points.step/s, 8 var
