vulkan secretly run better in software rendering, how do you accept it without sounding mad?
>>110108265What exactly am I looking at?
>>110108265spam thread, DNI
Fix your presentation you spastic fuck. Nobody knows what the fuck you're actually talking about unless you elaborate more.
>>110108292Just look at the fps?
Sorry my AI helper doesn't read video files I cant respond
>>110108367Yeah they're both abysmal
>>110108404What do you expect? If both of them run at 60fps what are we gonna refute?
>>110108345It is a benchmark
>>110108463Of what? On what? Why is the fps so bad? Why is the recording so cluttered and incoherent? Take your fucking meds so you can focus long enough to actually say something.
>>110108420>If I artificially adjust the fps on these two ""tests"" you can see that they're different.Retard
>>110108504Then you test it on your own rig to see whether it is a bullshit. If you think the test is fraud.
>>110108492What are you talking, i’m taking my own free time to do the test, it is a personal test, and it take effort to make a benchmark, yet you have the privilege to observe the result without effort and you are mad with it.
>>110108265saar
>>110108265this would be a better thread if you would just stop vague posting.
>>110108398Try another one
>>110108764It is better framerate
>>110108542Figure your shit out, jeet.
>>110108265>GPU 0% for bothI can't quite put my finger on it, but something tells me both of those are using a software renderer.
>>110109334>GPU 0%I can’t vouch for this, it is likely the software can’t recognise the gpu and how much is in used, it would only recognise reputable gpu (amd or etc) when attached.
>>110109517Maybe poor Vulkan support by whatever that GPU vendor is, is the problem then? My best bet.
>>110109558nah op is a retard
>>110108265stop spamming this board nigger
previously in vagueposting schizo theories>>110077789
>>110111262This is exactly what AI would say too, which is why I find it baffling.
>>110108265Have you tried not running an unoptimized emulator on a toaster?
>>110111917I haven’t do a lot of tests, but I prefer to test it on simple software first before going into heavier one. So far I’ve been impressed by the result.
There is no appreciable difference between a schizophrenic and a Hindu.
>>110112369install vulkan driver, you are using cpu rendering
>>110108520Ok I did and software is slower than hardware and you're gay.
>>110112737post screenshot
>>110108325I WILL engage
>>110108542>it is a personal test saar, vulkan secretly run better in software rendering saar
you might need actual vga setup
>>110111262The cpu could pull some 4000 to 6000 fps when done it with xvnc bundled if donr right, I could be wrong though. Which is very much as close as to gpu rendering.
after vibecoding for a while, i finally get the recording fix, now it render in real time without lag, and i think it is as best as it could get at this point.
>>110108265It's doing software rendering on both
>>110116437I would do the hardware rendering later, it is a test not a finalized product.
>>110109132>>110113194you two are actual retards. especially the second one parroting the first retard with precisely zero critical thinking attempts applied
>>110108265>running on a raspberry pi>surprised software rendering does better. Are you mentally handicapped?
>>110116168there's no way the pi3 is that slowis it?pretty sure the 486 could render faster than that in
>>110108265Consumer CPUs don't have the bandwidth for it. And server ones don't tend to include an iGPU./thread
>>110118299>don't have the bandwidthHigher bandwidth generate more heat, modern cpu is literally unusable for anything serious. I can stress test to my heart content and it would still not broken.
>>110108265Mario kart is very demanding
fucking state of this bored. I swear there wasn't this many ESL years ago
>>110119415>bored.kek
>>110108265>10 whole FPS emulating a fucking DS gameYour """benchmark""" is SHIT. Fuck off.
>>110119588Dude you do not see the potential of this, it can double fps on literally any game, video is just a proof of concept.
>>110118156pi5 could get just around 200fps to 300fps, it is norms.
>>110118360Lmao your cooling sucks.
android tv boxamlogic s905x3ddr3
another test, left one is running 20% faster and 15% more fps compare to the right one, at complete identical setting and emulator. it can be observe when run from the start screen point to end, right one delayed by as much as 30 seconds in a 4 minutes video.
