It might just be that I'm more proficient analyzing and working around #GPU quirks (happens, when you do mostly #GPGPU for more than a decade) than #CPU, but there's so many weird things happening on this machine that I don't know where to start from.
@VileLasagna Has a blog post on the relative speed of different #GPU compute frameworks on the same hardware and driver.
Tl;dr: on an #Nvidia card, with Nvidia drivers, #CUDA is the slowest, by far. Fastest is our old stalwart #OpenCL - almost twice as fast when used only for compute. #Vulcan is good, and the least affected by using the card for your desktop at the same time. Read it - it's good.
Today was one of the last lessons of my #GPGPU course at the University, and I double-checked with my students if by their account we only had one more scheduled lessons to go (on Monday). I discovered that one of my students kept an accurate log of all the lessons. Problem is, I was sure today we were at the 23rd lessons (so one more lesson to go after today), and he was sure we were at lesson #22 (two lessons to go).
(Some of you may already be guessing where this is going.)
OK so I'm ready for today's #GPGPU lesson with the new laptop. My only gripe for the lesson will be that #Rusticl in #Mesa 23.2 doesn't support #profiling information. Apparently the feature was merged at a later commit https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24101
and I even tried upgrading to my distro's experimental 23.3-rc1 packages, but trying to use rusticl on those packages segfaults. So either I've messed up something with this mixed upgrade, or I've hit an actual bug.
Met with a student today to discuss their project for the course I teach (#GPGPU). They plan on implementing a solver for a very interesting puzzle a colleague of theirs came up with, and even if the solver itself is conceptually trivial, there's a lot of interesting challenges in managing the intermediate data needed by the algorithm, which won't fit in a consumer GPU for even moderate problem sizes.
„Despite being widely used and relying on secret cryptography, TETRA had never been subjected to in-depth public security research in its 20+ year history as a result of this secrecy.“
But then again noone pays me to fix it, so it's not my problem.
Spoiler: The proper fix is to abolish all #proprietary shit and demand a fully #OpenSource'd communications system, since everything else violates #KerckhoffsPrinciple and is thus inherently and unfixably insecure by design!