Search the Community

Hey first time poster just don't know where else to go, My 3090 FE has suddenly started crashing and I can't identify why. After around 10 minutes of gaming both my screens go black and the gpu's fans peg to 100%, I looked around online and saw that the most likely cause of this problem is memory overheating as my gpu temps in task manager were reading normal (around 64 celcius when gaming at 1440p) so I launched HWmonitor and watched my memory temps, They got up to around 100 celcius and then the crash happened. Looked around online and everywhere said the thermal pads needed replacing and the chip needed re-pasting so I watched a bunch of tutorials and got all the recomended pads and thermal paste, took it apart took the old pads off and cleaned the memory with isopropyl alcohol and applied the new pads. Did the same with the chip and closed it all up, booted up and temps were looking way better. I was now idling at around 28-29 celcius on the chip whereas idle before was around 40 celcius and the memory was down to idling at around 40 celcius and before it was idling around 60 so I thought great everything worked and I should be fine, Launched a game and was playing for about 10-15 minutes before the same crash happened again and after turning my pc off and letting it sit for a while and then turning it back on and letting it idle for a while my temps were exactly the same as before I repasted and replaced the thermal pads so i'm really at a loss as to how to fix this or if i just didn't actually make a difference repasting and re-padding the gpu. Any help would be greatly appreciated as I truly have no idea what to do and I live in the uk so there's not really any good pc stores I can take it to for them to look at or fix for me. Sorry for long post, Thanks.

Hi all, I am trying to build a computer and I am getting the error code D6 when turning it on for the first time. Here is the build: ROG Strix Z690-E Gaming WiFi Intel i5-13600K Trident Z5 RGB DDR5 (2x16GB) Gigabyte Gaming OC 12GB 4070Ti Seasonic Prime TX-1000 WD_Black SN850X 1TB NVMe SSD Noctua NH-D15 NZXT H5 Flow When I try posting the machine after building it, the graphics card lights up and the fans begin to spin, then maybe 5 seconds later the fans stop spinning and the lights turn off. The motherboard then finally lands on error code D6. (Note: the only things plugged in are: the power cable, DisplayPort cable, usb keyboard, usb wireless mouse dongle) Things I have tried so far: Unplugging the GPU and using the iGPU as well as unplugging and reseating the GPU. Result: error D6 Using the Seasonic included 12VHPWR as well as the Nvidia branded 12VHPWR to pcie psu cables. Result: error D6 Reseating the RAM as well as using only one stick of RAM. Result: error D6 Reflashing the bios using BIOS Flaskback as well as clearing the CMOS using the Clear CMOS button. Result: error D6 I don’t have a separate motherboard that works to test the GPU, but because I get the same error message when I try to use the integrated graphics on the cpu, I suspect it is not the graphics card fault but instead the motherboard. Is there any other troubleshooting measures I can try to take before RMAing the motherboard? Thank you in advance EDIT: The issue has been resolved but I do not know exactly what fixed it so Ill just say what happened in case someone has a similar problem. After trying different things to no avail, I tried turning off the hybrid mode (no fan) on the power supply and suddenly the iGPU started working. I went through the Windows 11 install process and updated the nvidia drivers (Windows was detecting the gpu but when I tried to switch to it, it did not work). Then I tried moving the HDMI cable plugged into the motherboard down to the GPU (I had primarily been using DisplayPort on the GPU) and it worked. I then tried unplugging the DisplayPort cable and plugging it back in and it finally worked. Im not sure where along the way the GPU started working but hopefully this helps

Please excuse if this is accidentally a duplicate thread, I have done some research but not found a specific answer so far. I see a lot of talk about whether to use a 2, 3 or 4 PCIe connectors to a 16-Pin adapter but nothing about the PCIe adapter vs. a 16-Pin M-M. I was wondering when using a PSU with a 16-Pin connector built in, is there an advantage to using the M-M cable included over a PCIe adapter? Or is there an advantage to using a (3 or 4)PCIe to 16-Pin adapter? If there is, is it a safety advantage, performance etc.? Obviously the 16-Pin M-M would have a cable management advantage but I was wondering if there was others or if they otherwise work identically or is a specific option desired. To create an example to work with, using Cooler Master GX GOLD 1250 ATX 3.0 and the Asus ROG Strix GeForce RTX™ 4090 OC Edition 24GB GDDR6X Appreciate your time, thank you.

Hi, been having issues with certain games (Usually more intensive games). The two games i've specifically had issues with are just cause 3 and bullets per minute What ends up happening is the game freezes, music keeps playing, and my GPU usage lowers significantly, from around 90-95% to around 30%. It doesnt seem to happen in less intensive games, I've tried borderlands 2 and it runs fine. Oddly bullets per minute WAS running fine the other day but as of yesterday it's started doing this. Otherwise the games run perfectly fine, well over 60 FPS, no lag or anything just this stuttering issue. I have tried: Uninstalling Graphics drivers using DDU and reinstalling Changing settings in Nvidia control panel, also resetting back to defaults Installing on different drives, Both my WD 2TB hardrive and my Sandisk 1TB SSD Various settings in both games, highest and lowest. No effect Disabling all audio devices that aren't the main one Turning off Ingame stuff like XBOX game bar and Nvidias In game overlay Computer Specs: Windows 10 Pro Sandisk 1 Terabyte Sata Boot drive Evga 500 W1, 80+ White 500W, Power Supply 100-W1-0500-K3 B450 TOMAHAWK MAX Bios version 7C02v3G1 (Latest) Ryzen 5 2600 No overclocking Geforce RTX 3060 No overclocking 2x8 GB Corsair CMZ16GX3M2A1600C9 DDR4 3000 MHz RAM XMP profile 1 Thanks for the help in advance and feel free to ask if i've missed something 2023-04-21 22-51-48.mp4

I've switched back to water cooling after a long hiatus. And I've run into an issue that I haven't experienced before. My GPU is drawing less power i.e. Watts than before, which has impacted the FPS in gaming. This is something I haven't experienced before. The GPU is a XFX THICC III 5700XT Ultra. I did switch out the old thermal pads for thermal grizzly, but I don't think that could cause any issues, as the PC powers on and there are no black screens / hangs / crashes. Google search on this issue only brings up how to lower power consumption, which doesn't directly deal with what I am experiencing. I'm using Hardware Monitor, MSI Afterburner and AMD Radeon to monitor the power consumption, which is driving me nuts. Even a full Furmark / Prime95 Test is not pushing past 60 watts of power consumption. Before, this use to be around the 200-255 watt mark. Specs: Ryzen 5 5600x CPU 32GB Corsair Ram 3600MHZ Ram XFX THICC III 5700XT Ultra (Owned for almost 5 years now) Gigabyte B550 Aorus Elite Motherboard Power: Bequiet Straight Power 11 Gold 750 watt. What I've done so far to try and fix it: 1. Swapping out the previous VGA / PCIE. Before, it was one 8pin to two separate 6+2. Now its 2x 8pins to 6+2. 2. DDU Fresh Drivers / Rollbacks 3. Default Performance settings for the Card 4. Bios Update 5. Turned off / on Re-size Bar and AMD Smart Access Memory I'm just short of draining my water build and opening the Graphics Card to have a look if there is anything unusual. But I'm not sure that's the issue. Everything seems to running well, even as I type this. The issues are in multiple games I've tested over the last 2-3 Days. None of my games, with the lowest settings set, are able to crack 30fps minimum. I'm honestly boggled. Short of draining, reapplying paste and giving it another go, I might have to fork out a few thousand for a new GPU, (Australian Prices for GPUs are shocking) which I'm trying to avoid at all costs. Looking to see if the community has bumped into this issue. I feel like I've attacked multiple issue solving angles here. Not sure. Provide your input and I'll be happy to answer any questions / queries. I'll be checking this throughout the day too. Thanks for the Read!

I'm very new to Overclocking lets just put that put there lol. However, I have heard the term "Silicon Lottery" used a thousand times. I brought out my old GTX 1060 as I was going to give it to a friend for free who was just getting into PC gaming . When I installed it into his PC he was a bit underwhelmed with his performance in modern titles so I decided to overclock the old thing to see if it could give him some better performance. Over the course of the next couple of hours, I was shocked to see just how far I could push this card! Please let me know is this is a normal experience to have but I will include the overclocking stats below. CORE VOLTAGE: 100% (Included just for completion sake) CORE CLOCK: +240 MHZ MEMORY CLOCK: + 1000 MHZ POWER LIMIT: + 116% TEMP LIMIT: + 92 Degrees Once again the main aspect I am looking at is the core clocks and the memory clocks. Also this thing runs at around 75-80 degrees under sustained load in heaven benchmark for quite some time (mind you I live in Canada and its freezing up here in the winter)

EVGA RTX 2060 6gb Model EVGA208 REV 0.2 Manufactured after 18th August 2005 The card I own is it the same as the the KR versions? Which family of models does it come from Gaming, SC, and or XC?

So, I'll try to explain as clearly as I can, also it's my first time writing here so please be kind. (Also, English isn't exactly my first language) Okay so, my monitor shows "No signal detected" after the windows logo, this only happens if I have my graphics drivers installed and it works normally if I boot into safe mode and delete the drivers. It worked perfectly before, also I use an adaptor (VGA to HDMI) so couple days ago I tried fixing this by removing the graphics card and reinstalling it and it started working again but yeah I had to like move the cable a bit. Then after restarting it once it stopped working again. I thought it was an adaptor issue so I bought a new adaptor and nothing, still having the same issue. Any help? Ah I have tried couple of things so far, reinstalled windows too. So at this point I am starting to think that it's a cable or a port issue, my monitor does support DVI so maybe I can try buying that? Anyway I am here to see if I can try doing something else.

As many reviewers mentioned in their reviews of the new NVidia RTX 4070 (NON-Ti) yesterday, each generation of new graphics cards has seen a rise in pricing and I have an idea as to why that is, and unfortunately if this theory pans out, things will only get worse from here... You see, Moore's law has been steadily dying off for quite some time now. Its not completely dead yet, but the advancements in computing horsepower per dollar have significantly slowed down. Yes, the last few generations of Graphics Cards and CPUs have shown rather large jumps in performance, but they have also seen quite large jumps in pricing. There is a reason for this that is unavoidable, allow me to explain: You see, Moore's law is slowing down because we are approaching the absolute physical limits of what traditional silicon transistors can actually do. Years ago when TSMC began work on the 5nm process node now found in current generation hardware, an article came about that stated there could be major delays with manufacturing functional 5nm cores because of some serious engineering challenges. The article stated that TSMC engineers had even spoken about the possibility of 5nm not being feasible at all. This was because they were concerned that the transistors were so small and so dense that it would create a "nuclear core" thermal run-away scenario where cores are producing so much heat so fast with this density that a standard IHS and standard cooling solution just wouldn't be able to remove the heat as fast as the cores create the heat. This would create a situation in which exotic cooling would be required just to cool consumer desktop PC cores. To be honest, in a way, their prediction and concerns are starting to show some truth to them. We have Ryzen 7000 and Intel 13th Gen that both run so hot at full load that they essentially throttle by default, "targeting" their TJMax and "boosting until they hit the limit". Obviously we enthusiasts aren't dumb enough to fall for the marketing. We know well that the cores have simply gotten too hot and maximum performance now has to be throttled to suit each end-users cooling solution. Thats what it really is. Even a 360mm AIO CANNOT properly cool an Intel i9-13900K without some sort of modification or advantage such as direct-die cooling. Properly cool meaning keeping the CPU under 90-95°C during all-core 100% load workloads at default settings. Besides the thermal issues with continuing to shrink transistors ever more, there is the sheer amounts of incredibly expensive technology and engineering involved to even get transistors this small, and the yields aren't great either. Some people across internet forums have begun to speculate that for 3nm and smaller transistors to actually function correctly and not run into nuclear run-away thermal issues, we will be forced to step away from traditional silicon transistors and switch to a more exotic technology - something like carbon nanotubes for example. And I believe they may just be correct in that assumption seeing how many issues there are already - 600 watt GPUs and Flagship CPUs so hot, modifications are required just to keep them under TJMax. So seeing as were are knocking on the door of what is physically possible with standard silicon no matter how advanced the manufacturing process gets, it only makes sense that prices would rise as we reach the very edge and the R&D + engineering required to make further improvements just gets exponentially more expensive and more difficult to pursue. It also explains how the last few generations of hardware just keep getting more power hungry and run hotter. Now a lot of you might point out the fact that the absolute monster that is the RTX 4090 doesn't actually run that hot. Well guys, why do you think this generation of flagship GPUs have coolers the size of actual bricks you use to build houses? Meanwhile, CPU coolers can't really get any bigger because of case compatibility. Imagine Noctua trying to make a 250mm tall version of the D-15, that would be absolutely rediculous. How about an 800mm (200mm x 4) radiator for liquid cooling? It just wouldn't make any sense. So in conclusion, I feel that MSRPs will continue to spiral out of control until the majority of people simply stop upgrading. Then some time can be taken to focus on pure efficentcy and lowering prices. I don't even want to think about how expensive the first carbon nanotube-based CPUs will be. Can you imagine the equivalent of a modern 8-Core typical CPU that costs over $1000? A Ryzen 7700X that costs as much as a flagship GPU? That is the day that PC advancement for the general populace essentially comes to an end for a while. So what do you guys think about the situation on pricing and what do you think will happen in the future? Makes for some pretty interesting thoughts, that much is for sure...

Hi, this is my first time posting here, and i am in need of a few opinions. So basically, i currently have the ASUS ROG Strix GTX 1080 8GB, and my friend is selling his Zotac RTX 3060 12GB. I was thinking about buying it. I would like to know if its worth switching to the RTX 3060. I saw a few benchmarks on youtube comparing it and it seems there is a slight FPS difference. I mainly play F1 22, CSGO, Warzone, and occasionally RUST with a 27" 1080p monitor. Your opinions would be very much appreciated, should i switch to the RTX 3060 or stay on the GTX 1080?

Budget (including currency): Roughly $3800 USD in total. Country: Australia Games, programs or workloads that it will be used for: KVM, Compiling, Programming, Video Editing, Gaming, VR, 3D Modeling, etc. Other details (existing parts lists, whether any peripherals are needed, what you're upgrading from, when you're going to buy, what resolution and refresh rate you want to play at, etc): Wanting to run a KVM virtual machine with PCI passthrough via OVMF. I plan on running Arch Linux as the host operating system, and running Windows 11 as the guest system. Only wanting to run the AM5 platform, and I plan on using a Ryzen 9 7900. (non X) My problem is finding a motherboard that can support 2x GPUs. Preferably 2x RTX 4070 Ti's. Every motherboard that I have come across has a full PCIe 5.0 x16 slot that can run in x16 mode, but finding one that can run another graphics card at x16 seems impossible. Maybe I'm missing something. Just can't seem to find a motherboard that can run both GPUs at x16 speeds/lanes. Most seem to throttle down to x4 speeds/lanes. The PCIe version doesn't bother me. I don't care if it's PCIe 4.0. I just need to find a decent motherboard that can support 2x GPUs that run at full speed, but won't break the bank. (I don't exactly want to spend $1000 on a motherboard.) I mostly know all the other parts I plan on using in my system, but am just struggling to find a motherboard that can support my requirements. Thanks for any recommendations/help!

Should I get the XFX Speedster QICK RX 6650XT or the Sapphire Pulse RX 6650XT? Which one will be better for longevity?

I am getting a 4080 for my birthday however..... i cant really figure out with board partner i should go with. Any ideas?

My PC isn't running too hot, however the fan on my 2060 is super loud. As a way to increase airflow without getting a new case, could I flip my PSU to use as a way to extract air?

I live in a developing country and i want to build a gaming pc on a tight budget. I've chosen i3-12100F with ASUS Prime H610M - E D4. My GPU options are GTX 1650, RX 6500 XT and RX 580. All are under my budget but I'm confused if i should go with a PCI Express Gen 4 card with 4GB VRAM since my CPU is 12th gen or go with the RX 580 which has 8GB VRAM but PCI Express Gen 3. I will be gaming on 1080p only. Pls do not recommend buying a used GTX 1080 or similar because there are almost no options on the used market here and its very risky. Also I'd like to know the required wattage for my PSU. Newbie PC builder here so not really sure how to choose wattage for PSU. Help is appreciated :)

Hello all! System specs GB Aorus Elite X570 Wifi Ryzen 5800x3D RTX2080S (old) RTX4070TI (new) 16GB DDR4 3600 750W PSU TLDR; GB Aorus Elite X570 Wifi + 4070ti crashing on PCIEx16 Gen 4, appearing more stable on Gen 3(?) I recently purchased an ASUS Tuf RTX 4070ti. I installed this into my system right away. I did some diligence before the purchase, making sure I had the wattage and parts that weren't too too bottlenecked for 1440 future-proofing. This GPU was truly the last upgrade I could make without having to scrap the system and move on to next gen mobo, higher PSU, ram, etc. I installed the card, reinstalled NVIDIA drivers and was on my way. 2 days later, I noticed some hitching in game; COD and FIFA 23. Something was off. I did another driver install thru geforce xp, and a driver reset (w+ctrl+shift+b)and it seemed to relieve the issue. Today, I am noticing massive hitching, slow performance/ instability in windows desktop, automatic restarts, black screens on power up and restart, and monitors disconnecting on their own. A panic set in, and to forums I went. I first attacked my BIOS version, by updating to the latest software as my BIOS was approaching 1 year old. This had no effect, however I did have to reapply all of my BIOS settings. I read much about DDU and performing a true clean graphics install via safe boot, etc. I did this, and my system was performing well. Shortly after, I restarted the computer as I had just downloaded multiple things and re-established the graphics drivers. The computer black screened, and I was forced to power down and start back up. The hitching and low performance continued. I re-entered the BIOS and then change my PCIEx16 gen from auto (which would be 4.0 as it is a 4070ti) to 3.0. Now the system appears stable and responsive, with appropriate performance in game as well. Now I have only powered down twice, so I am hoping I don't re enter the same loop. Has anyone else encountered this issue? Or can anyone confirm that my reasoning/solution is correct??? I am really trying to avoid these crashes and instability. Any other tips to remedy this, and make sure it doesn't start happening again? Thank you!

I have an i7 7700k with an RTX 2070 right now. I built my pc back in late 2017. I think my pc is still pretty good right now but I've been thinking about upgrading my graphics card to a 40 series card, probably a 4070. I was wondering if I should upgrade my i7 7700k to a newer one like the 12th or 13th gen i7, or wait for 14th gen. If I do get a 4070 will my GPU be bottlenecked by my i7 7th gen ?

Good day, I have a new rx 7900 xtx from sapphire the vapor x model Today i got a game crash telling me the driver has timed out and the application got closed. Before the 7900 xtx, i had a 6950 xt wich i sent back cause the coil whine was unbearable. The same issue occured once there too Is there a problem with my graphics card or as i read, a problem with the drivers? I played apex on the 7900 xtx My specs: Cpu: r7 2700x Ram: 16 gb g skill trident z Mobo: b450 aorus pro Power supply: seasonic prime 1000w 80 plus platinum

Greetings, I will try and make this short; I was recently gifted a machine that is currently running Windows 7. The motherboard on this machine is a Hewlett-Packard 158B and it has a NVIDIA Quadro 6000. I was wondering if this motherboard is even capable of running a 3090. The reason why I have a 3090 is because it was on my main machine before I upgraded to the ti model. I was wondering if that motherboard is even capable of a card like that. And if not, what good graphics card is capable? since the Quadro 6000 is starting to show it's age lol. Thank you.

Hello, I am wrapping up my first PC build and the last factor is a GPU. I was initially eyeing the 7900xtx, but after more deliberation, am looking more at the 4090. I have adopted the ideal of buy once, cry once (for better or worse) as I plan to run this setup for awhile before upgrading. I have been thinking of the MSI Suprim X or the Suprim Liquid X, but was wondering if anyone had experiences with these cards, or would suggest another. I personally do not care for the aesthetic of the Strix card, though I do understand it is one of the highest rated for power/performance. It will be housed I'm a Lian Li 011 dynamic evo and I do plan to vertically mount the card. I currently have a 360 radiator for my CPU AIO mounted on the top, but if I went with a liquid cooled card, that would likely be moved to the vertical mounted position and the GPU AIO mounted on top. Any insight or experiences with these, or other comparable cards would be lovely and I would appreciate it very much, thank you! Side note, I am already running a 1000w PSU as well as an i7-13700k with 64gb ddr5.

Is it possible to boot steam os from the micro SD card and then use an external GPU on steam os? I can only find videos of people trying it with windows.

so im going to buy a gpu after waiiting for over 2 years. long story short, recently theres been a problem with Polaris architecture (DX12.0 feature level)- the forespoken game, that thing. so im thinking if its good idea to buy an rx570 right now ?? i play at 900p and i dont mind medium settings at games.... or should i pay double of rx570 and go for gtxx1070/1660 super/rx 5600xt? will DX feature level 12.1 really matter in coming years? heard of any upcoming games that wont work over fl12.1? thankyou!

In 2017, a rat entered my gaming pc and it pissed on the graphics card. When I turned on my pc, it fried my poor Asus Strix GTX 970 GPU. Today i decided to take a look it and saw burnt pci-e contact pins. There is no actual damage to the vram, capacitors or GPU die itself. But the contact pins are burnt and something probably decomposed on the card. So, these are my questions- Q.1 Is there any possible way to restore my graphics card? Q.2 Should I clean it up with isopropyl alcohol and try turning it on? Q.3 Will it damage my pc?

i need some help with buying the right thickness of thermal pads for the vram on msi 1070ti gaming x. Thanks for any help.

My 1060 gpu Is overheating at 96c while running heaven at medium at running a anywhere between 10-30 frames. Anyone know a fix?