|
why does opencl suck so bad why can't i easily have a hypervisor thing that lets me play video games on windows and then alt-tab over to linux to do cuda poo poo, without involving a second drat graphics card so one can sit idle while the other does stuff why atomicthumbs fucked around with this message at 21:05 on Oct 3, 2017 |
#
¿
Oct 3, 2017 21:03
|
|
|
|
| # ¿ May 19, 2024 02:58 |
|
|
ate all the Oreos posted:so I'm trying to get ffmpeg running nice and fast on my dinky old server box just for fun and I notice it has CUDA support, great cool. I go and get my old graphics card and shove it in (only like 5 years old at this point) and no actually it's the wrong kind of CUDA you need the kind of CUDA that does video, which is only available on the newest cards. what the heck? "cuda" might not actually be cuda in this case; graphics cards have video encoding and decoding blocks built in these days, and their capabilities differ by generation
|
|
#
¿
Oct 3, 2017 21:19
|
|
|
back before someone recycled a GTX 780 i discovered that for machine learning poo poo (restricted to opencl) running stuff on the integrated GPU in my 4790K using beignet was surprisingly a third as fast as doing it on my amd 7870. interesting and neat
|
|
#
¿
Oct 3, 2017 21:22
|
|
|
Suspicious Dish posted:You'd think they'd decode some magical proprietary format if that was the case but nah it's just the same H.264 that you can already decode in software. doing decoding on quicksync saves a lot of battery power on a laptop and i think the encoder in my gtx 780 is faster than doing it on the cpu (either via quicksync or software) atomicthumbs fucked around with this message at 21:47 on Oct 3, 2017 |
|
#
¿
Oct 3, 2017 21:36
|
|
|
let me phrase it this way: why does CUDA blow opencl's pants off
|
|
#
¿
Oct 3, 2017 21:50
|
|
|
ate all the Oreos posted:yeah that's what I was saying in that post, and then what we've discussed for like two days since that post lol well excuuuuuuuuuse me mister "i read the entire thread for context before responding"
|
|
#
¿
Oct 3, 2017 21:57
|
|
|
somewhere in my boxes of boards i have what was an extremely expensive PCI 3D accelerator made by evans and sutherland. i oughta build a 90s pc workstation someday freedom graphics or realimage 2000 or something? it has tiny fans on the main chips, a daughterboard, and at least one simm slot looked somewhat similar to this AccelGraphics Eclipse, which is E&S based:  I also have some prototype of some "realizm" card or other IIRC. Intergraph, not the later 3DLabs ones atomicthumbs fucked around with this message at 02:27 on Oct 5, 2017 |
|
#
¿
Oct 5, 2017 02:22
|
|
|
what's the deal with radeon instinct also graphics cards (like everything else) were better back when they were just a pcb completely backed with purple-and-gold ceramic-packaged chips on both sides
|
|
#
¿
Oct 6, 2017 02:03
|
|
|
Bloody posted:ceramic packaged chips are awful i've never seen a more wrong opinion in my life
|
|
#
¿
Oct 6, 2017 04:02
|
|
|
feedmegin posted:Na man DIP packaging for life 80s style well, yeah, as long as it's ceramic. purple and gold are best but white and gold is also extremely good. basically as long as it's brazed  
|
|
#
¿
Oct 6, 2017 23:58
|
|
|
ate all the Oreos posted:sorry, could you also explain what the hell a 1-bit texture is or how that's useful? is it just for like, masking? 
|
|
#
¿
Oct 19, 2017 02:05
|
|
|
acetone vapor is nature's antialiasing
|
|
#
¿
Oct 19, 2017 02:17
|
|
|
feedmegin posted:Wayland is a windowing system/compositor, not HPC or anything, so in fact its largest target GPU market is Intel. Plenty of people out there running without discrete GPUs because they don't need them to shuffle windows around their desktop. i'm running with an integrated gpu so that cuda can have all the gtx 780's ram to itself because boy howdy does it need it
|
|
#
¿
Nov 15, 2017 05:01
|
|
|
ate all the Oreos posted:i tried to do this on my server box and the motherboard hates the idea, you either have to set it to use the integrated graphics adapter by default (in which case the GPU doesn't even show up at all once the system boots) or in any other configuration the integrated graphics adapter is never started and the only display signal is coming out of the card itself what Operating System are you using because it sounds like a Piece of poo poo
|
|
#
¿
Nov 16, 2017 11:06
|
|
|
feedmegin posted:This is why the fact that Intel is actually the largest vendor, as I mentioned earlier, is important. For people who don't need high performance OpenGL, Nvidia doesn't get to dictate this at all. Even with Nvidia hardware nouveau is enough to run a desktop. Also, there hasn't heretofore been a competent, good, production-level replacement for X11 made by the people who previously maintained X11, backed as mentioned by major distros. I'm not even sure what you could be thinking of from like the mid 90s as a project to replace X on Linux. intel is the largest vendor and minix is the most-deployed operating system
|
|
#
¿
Nov 16, 2017 21:34
|
|
|
|
| # ¿ May 19, 2024 02:58 |
|
|
XFree86
|
|
#
¿
Nov 18, 2017 08:02
|
|