Hashmode: 5500 - NetNTLMv1 / NetNTLMv1+ESS Hashmode: 22000 - WPA-PBKDF2-PMKID+EAPOL (Iterations: 4095) OpenCL API (OpenCL 2.1 AMD-APP (3180.7)) - Platform #1 To disable the optimized kernel code in benchmark mode, use the -w option. Note: Using optimized kernel code limits the maximum supported password length. You can use it in your cracking session by setting the -O option. Quick Benchmark hashcat (v6.1.1-275-g057de100d+) starting in benchmark mode.īenchmarking uses hand-optimized kernel code by default. HMU to untangle this if you have bigger linux gennies Look, I was just happy to get numbers.rocminfo and the rocm-smi were cool though and worked fine.it seems like ROCm is not relevant to 50 series navi cards?.Fedora 33 with rocm and what I could make of this:.rocm builds in aur - though this thread looks promising ROCm-Developer-Tools/aomp#187.Kernels 5.4.0-52, 5.8.0-something from kernel-mainline builds.21.10 release with combinations of amdgpu-install, amdgpu-pro-install, -opencl=rocr,legacy or both, -headless and not, -no-dkms and dkms.Things I tried which did not get OpenCL working:.GP has gots to get his hashcat on pronto ).Outside the chroot, in the archlinux install there is nothing special- no dkms builds of drivers just amdgpu from the kernel package, no opencl-amd, no special firmwares.Uses a messy chroot containing hashcat, amdpro bits and who knows what else that I am trying to narrow down to the sufficient conditions.Software: Hashcat v6.1.1-275-g057de100d+, Archlinux, Linux 5.11.16Īccelerator: 1 x AMD Radeon RX 6700 XT, XFX Speedster QICK 319 Notes STORAGE: Samsung m.AMD Radeon RX 6700 XT (XFX QICK 319) Hashcat Benchmarks RAM: HyperX Savage 16GB (2X8GB) DDR4 2400MHz Hopefully, these sort of improvements can find their ways into software in a more timely manner - Hawaii-based dGPUs made their debut in late 2013. relative performance differs by scene - the complex indoor environment of 'classroom' favours AMD cards NVIDIA cards show better results with the studio shot of 'bmw27' the GTX 970 (1664 CUDA cores) can beat the GTX 1060 (1280 CUDA cores) in 'classroom' & 'pavillon' despite its lower FP32 performanceĪll in all, it's great that now Radeon users can fully utilize their dGPUs in Blender. one would have expected stronger performance from the R9 380X - numerous workstation cards and the 2014 / 2015 edition of the iMac Retina 5Ks wield Tonga-variants.the RX 560 falls short in the fight with the GTX 1050 - no repeat of the strong showing of the big Polaris chip.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |