xLOCKnLOADx

@luancastro_lc @NasiLemakTech Yeah then scene didn't fuck up here, it's 792p on 8E/9500/8EG5 iirc according to geekerwan too so matches up You had GSR off tho right? That bumps the resolution up

@xLOCKnLOADx @NasiLemakTech Bruh *-*

Gonna have a watermark in future gaming tests because our videos always get screenshotted and shared without credit :( BTW, NTE runs at only 720p on Android lmao. Optimization is terrible as well.

@NasiLemakTech @luancastro_lc There's a trick to get the render res in some gacha games though Simply go to the camera mode, take a picture with no borders, see what resolution it says (doesn't work for wuwa iirc, nte genshin do work)

@NasiLemakTech @luancastro_lc It's just the window resolution, not the game's render res (it can be higher, usually is, lower is unlikely as that's supersampling) Scene just picks up the UI resolution basically which devs try to make sharp and not blurry like the 144p renders

@Unknown14539139 @eggsee_ Dw, that guy can't use Google somehow and it would deeply hurt him to do so (what he said was entirely wrong to sound smart 😭)

@eggsee_ Your embarrassing bro 😭

> talk about stuff unrelated to mobile > didn't even play on mobile > didn't even do test and provide data to backed up what he yap in the video if you're not in any way knowledgeable about stuff, you should just shut up and not spread misinformation man @RyuukuSensei

@RyuukuSensei >doesn't even watch the whole vid (as he claimed there was no data, but there clearly is plenty) >sticks to a line there just to build a flow (it talked about wuwa in general anyway, without even mentioning RT) Clearly, someone wanted likes in your name 🤭

What are you even talking about? 1. You're talking about when I mentioned leaf/water physics, etc. in the video intro as a throwaway line about the consistent graphical improvements. After that, I never mentioned them because they're not the problems that were brought up; CPU-side culling failure, pixel overdraw, 99% prism rejection rate and overuse of FP32 shaders. 2. This is a moot point because I don't have to play on mobile to read data. 3. Yeah, **I** didn't do the test because I don't have a phone with a Gen5 Snapdragon 8 Elite. But what do you mean I didn't provide data? It's literally in the video description and I showed the graphs on screen when I was talking about them. (Screenshot of one of them, showing both me and the data.) You then go on about ray tracing and a bunch of other things I never even mentioned outside of the intro making me wonder if you even watched the video past the first 30 seconds.

@eggsee_ @RyuukuSensei One thing I will add though, it becomes rather evident that you've never really touched a computer in your life and basically got into this "hobby" for wuwa configs as you seem to lack awareness of very very basic facts about benchmarking, SoCs and the OS you're working with

@eggsee_ @RyuukuSensei "What you said still doesn't agree with what you claim." That's a funny response considering you had no counters left Which makes sense as you were arguing about a topic which you didn't have much awareness of (didn't even know what threads are) Sure tho, end of discussion

If the data I've seen is correct, mobile has a lot more efficiency problems, for different reasons, than PC. It wasn't meant for lower-end devices, sure. But there are certainly a lot more devices that could easily keep up if certain things were changed. Working on a video now.

@eggsee_ @RyuukuSensei That's... Quite literally what a CPU bottleneck is, especially due to a heavy thread

@eggsee_ @RyuukuSensei on top of that, my point would hold perfectly? The i5 12400f and 12450h are similarly powerful in MC, far behind in ST And they can't saturate the GPU, the CPU isn't at 100% either (which means all cores didn't hit full load, which would be the case in perfect multithreading)

@eggsee_ @RyuukuSensei throttling shouldn't happen, capping the frequencies on just the prime cores somehow hurts frame rates (literally have a video of that sent before) An i5 12450h can't hit good gpu util, cpu util isn't near max either But sure, somehow that heavy thread isn't the primary factor!

@eggsee_ @RyuukuSensei It's really not that hard to read an image and if you ever saw a perfetto trace in your life you'd know that the 100ms limit would hurt any attempts to gather good data

@eggsee_ @RyuukuSensei x.com/i/status/20480… Because you're referring to information that can never actually be accurate and is only guesswork Which I've shown through a video (which we established, higher freq for same load=lower util)and even the same command scene's measurements will be limited by

@xLOCKnLOADx @RyuukuSensei Looks like the utilization is even across all cores (with some spikes here and there), even the smaller cores are on higher utilization. Again, your claim doesn't really agree with what you're showing.

@eggsee_ @RyuukuSensei and if you're still interested, here's proof the in-built android toybox (which scene will be forced to use without root) is quite literally limited to 100ms samples which will include durations where the CPU was done working and drop the util in an average (even if it was 100%)

@eggsee_ @RyuukuSensei "There's no fastest 1 core getting 100% utilization in any of the proper test graph" You can't measure that quite literally Unless you rooted your phone and wrote your own command Android's own toybox will only land you an average Making thread stats far more accurate

@eggsee_ @RyuukuSensei Anything here isn't actually that complicated, but you seem to lack a basic understanding of how these measurement tools or hardware actually works and are hellbent on sticking to that idea for some reason (misconceptions a basic google search could clear up)

@eggsee_ @RyuukuSensei If it were scaling well across all cores, as your tweet for a few likes claimed, total CPU usage would be near saturation. (Close to 100%) Instead, uneven thread load + low GPU usage points to a CPU-side bottleneck