RainingTacco

Use something like Riva Tuner to limit the framerate to 144Hz, it might smooth out the frametimes.

That's not double, don't spread misinformation.Generally speaking junction temp will be 50-57 in idle, and GPU temp will be similar. Only under load the junction temp goes +20 degrees higher than gpu. So generally speaking GPU temp =70 degree, and junction will be 90.

AMD Finewine happening on every GPU generation is a myth.

If you have warranty then 2070 super is no brainer

Your display is fine? Have you tried another like TV? Im 99% sure it wont help, but you should rule most things and write it down in RMA ticket.

No one asked it, but do you run your card at stock? Temps? It shouldnt do that either way, but that should still be ruled out. I would RMA it.

Apparently the newer driver version doesn't feature that. Annoying, but i guess newer driver is worth it.

Anyone knows a separate program or a setting in AMD adrenalin software to do that? Just put GPU in more performance oriented mode and less focused on power saving.

Antec P8 is very closed case, you need these three front fan slots occupied with high air static pressure fans and on quite high rpm do push lot of airflow. Your best bet is going negative pressure route. With 6 fans though you should be fine even with such case. GPU temp is not important for you, just watch the edge/junction temp not going over 90 degrees if possible. It can go as high as 105, but i doubt the card longevity[AMD says its fine, but...]

What's your PC case and how much fans you have?

I wish nvidia released gpus as strong as SUPER series but without the RT cores, thus making them cheaper.

Die and PCB/SMD cost is probably around 400 dollars, overall i would say that 50% of the GPU is just pure income for nvidia.

Im sure its not. It's texture loading problem, not two texture overlapping and fighting each other. It last much longer than 5-50 miliseconds. As i said it only happens in few games, so it might be related to their engines and AMD driver.

Well ive tried both stock and OC and it still happens. It is pretty rare like every few minutes, but it happens, and not in every game. And in some games, i also get occassional stutter when moving fast/loading area[like in far cry 5, and its not autosave!], im sure its not overloading VRAM, since ive monitored vram and it doesnt even get close to full capacity[like using 3-4 gb only]. Im using game on NVME SSD, 16gb of ram with 3600mhz and cl17, and ryzen 3600 so theoretically there should be no problems whatsoever. PSU is corsair rmx 850 so its pretty much overkill.

Well then buy g102 as i said its the same sensor, its the same shape, so braided cable is the only bonus here.

So here's the topic: And a picture I'm on 19.12.3 but i can't see any "Graphics Profile" at the tray. I only have "Video Profile". Basically i want my GPU to always run full clocks when there's any load, or substantial load if its possible[>20%]. Any way to do that?

Well if its for asia then i must have g203 Either way they are the same mouse, and i dont have braided cable!

I have Logitech G102, and from what i see the shell of the mouse is the same, so if you like the shape go for the one you like more feature wise. They both have the very good sensor -mercury sensor is some kind of knockoff pmw 3330 or 3360 if i recall correctly, so that's pretty much esports tier in a cheap mouse. If you are casual gamer, search for palm mouse that is more comfy. Preferably with pinkie rest or special shape that is suited for right handed people. One minus are the switches. The switches used in the mouse are pretty cheap so expect them to go bad after only two years of usage. You can resolder a new one, or use the trick with opening the switch and bending it back

Ive ran the unigine heaven, havent seen any artifact. Care to check my other topic? https://linustechtips.com/main/topic/1137070-textureshadows-occassional-poppingflicker/

So in some games i get occasional black texture popping in like for a miliseconds. Sometimes textures flicker or pop/move for a millisecond and return to normal position. It doesn't happen in every game -i get this in Kingdom Come Deliverance and to less extent in Mass Effect Andromeda[no black textures, just some texture popping]. It's particularly noticeable in GPU rendered cutscenes where scenery changes fast. I thought that my VRAM is bad, but i've done some tests and can't see anything wrong.Here's more info

If there's no direct access how manufacturers decide that there's a bad VRAM? Surely they need to have a method that doesn't use a human and his subjective opinion? So i don't get any artifacts in 3dmark, furmark and most of combustor test. In kombustor when i turn on "artifact scanner" i get the artifact number only when doing "msi-01 burn in mode" and its some whack number like 1 mil / 4 mil. I don't see anything on the screeen though. Is the program fubar? Any other good program with non-visual artifact detector/scanner? Here are screens.

This is a visual confirmation type of program. I want hard written data.

Anyone heard of such program? I would like to test my vram for troubles, because i suspect it might be corrupted slightly.

You know what would be amazing? A fucking robotic PC case which would automatically block the top with some kind of cover, after you power down your pc. It could slide the flexible piece of material from the side, or use some telescopic/accordion alike material that would stretch just before the system shutdown. This of course is not a problem for people using thick rads with push and pull config, since its unlikely that a dust would get through it into the pc, though cleaning rads with fans on both sides is also a pain.