Jump to content
Search In
  • More options...
Find results that contain...
Find results in...


  • Content Count

  • Joined

  • Last visited


This user doesn't have any awards

About DJ46

  • Title
    Geek on a budget
  • Birthday 1998-06-04

Profile Information

  • Gender
  • Location
  • Interests
    Computers, games, internet, doing the best I can in highschool.


  • CPU
    Intel Core i7-4790
  • Motherboard
    Gigabyte B85N
  • RAM
    Adata 4GB + Kingston 8GB
  • GPU
    XFX GTR RX 480 8GB
  • Case
    Lian Li PC-A05NB
  • Storage
    Samsung 850 EVO 500GB
  • PSU
    Corsair RM1000i
  • Display(s)
    HP Compaq LA2306x
  • Cooling
    Scythe Ninja 4
  • Keyboard
    Drevo Gramr (modded with real Cherry MX Blues)
  • Mouse
    Logitech G502 Proteus Core
  • Sound
    Sennheiser HD555
  • Operating System
    Windows 10 (+StartIsBack)

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. I used a 75% keyboard for a long time and absolutely loved it. It's not much bigger than a 60% but you get basically every key that a TKL has.
  2. What speed are the GPU core and memory running at?
  3. Your GPU and CPU combination seems pretty balanced, shouldn't be that bad of a bottleneck. Regardless, increasing the resolution increases the GPU usage, so it would be less of a bottleneck if it is caused by the GPU. Your FPS will probably drop though, as 1440p is quite taxing. The RX590 can handle it, but you probably won't be maxing out the details.
  4. If you don't need the iGPU (can be useful when trouble shooting), you can look for a motherboard with the P67 chipset. Otherwise you'll want the Z68 chipset or the newer Z77 chipset. Look for a higher grade motherboard (GD-45 and higher from Msi, Z77-V or Premium or Deluxe from Asus) with 2 heatsinks on the CPU VRM (that form an L shape). It doesn't tell the whole story but it's good as a first step in filtering the used market.
  5. I know that feeling too well I'm not sure if uneven thermal paste would cause that, a part of the die would probably heat up a lot faster than that. Without knowing the state of the thermal pads on the VRAM and VRM it's hard to tell. It could also be a different component that isn't cooled directly, unstable due to some kind off damage.
  6. The tape does seem pretty odd. I assumed there was no warranty. Since there is, I'd just send it back. Hopefully they don't mind you taking the card apart.
  7. If it was watercooled it also could've gotten wet. There might've been a leak from the CPU block onto the back of the PCB and only specific components/traces got damaged. So it works fine until those get utilised. You could thoroughly inspect the PCB or even clean it with a soft brush and some isopropyl alcohol, let it dry completely and try the repaste a few times. Since the alternative is throwing it away, you can just view it as a project/experiment and work on it whenever you feel like it.
  8. You can always put it on the shelf, get some thermal paste later and try it a couple times. I'm not saying it will 100% save the card, but for me it worked on like the 5th attempt, as I said before. I've had a reason to take the card apart since but honestly I'm too scared to do it. I don't want to anger the demon in it that seems to be chilling for now
  9. You have to remember that the temperature sensor is somewhere on the die, specifically on the edge. New AMD cards have multiple sensors, one on the edge and then a bunch spread around the core, the hottest of which gets reported as the "hotspot" temperature. When you think about the size of a GPU die, that sensor is useless if you have an air gap the size of a dust particle somewhere, with a bunch of transistors under it getting basically no cooling. After I realised that my issues probably came from something like that, I was terrified. The GPU was showing artifacts and I knew I tried upping the voltage first and I had been playing VR games in the mean time. I was completely sure I ruined it. There was a tiny dent that looked like a spec of dust in the die, and I thought it was a burned out spot of silicon or something. I guess it's just a scratch or something, since the card works fine now. That's why you have to take great care when spreading the paste and exactly the reason EK recommends that giant amount of paste - their mounting system can probably create enough pressure to actually push the extra out from the gap, like a CPU cooler would. My issue could've also been caused by the fact that I was using the Arctic MX-2 thermal paste. It's a pretty thick one. MX-4 is easier to spread as far as I know, same with the Thermal Grizzly stuff I've seen.
  10. I wasn't sure how low you could set the power limit. Bummer. If you need to get the card to work, you could try repasting it. I thought I knew how to do it but the previous comment of mine says otherwise... The mounting pressure of my GPU's cooler is probably weaker than I'm used to from CPUs, as the issue ended up being (probably) too much paste. When it comes to GPUs, I am an advocate of spreading the paste. You have to be careful and check the coverage (I do it by checking if any part of it reflects light like naked silicon, preferably use a magnifying glass to look for small gaps). I just stick my finger in a plastic bag and spread the paste with my finger. I even tried the method EK recommends (looks like a plus AND a cross combined) and only accomplished making a mess. The last time I put on barely enough paste and worked on spreading it for a long time, making sure every part of the core was covered but with as thin a layer as possible. It's hard to describe, but hey, you can just try a few times. Not like there's much to lose if the GPU is useless anyway.
  11. The fact that it happens only in 3D workloads is a pretty typical issue, but they way it is crashing is quite odd indeed. I had plenty of very interesting issues with my GPU and right now it's somehow working 100% fine after the roughly 5th repaste. You could also try lowering the power limit even more and seeing if it ends up being completely stable at like 50% or something.
  12. I don't know anything about CeX but from looking at their website they seem to sell used stuff or something like that. I'd guess someone sold them that GPU already broken, they checked it by booting to desktop and thought it was fine.
  13. So, did the card work perfectly in the other machine? What are the specs of that one? (specifically CPU and PSU)
  14. Yes. If you're numbering them from the 24pin toward the socket you should use the 1st and 3rd slot.
  15. Boy do I love this helpful forum. He literally quoted the manual, but this section is often a little confusing because the channels on one page imply a different module layout than the next page recommends.