Search the Community
Showing results for tags 'thermals'.
-
As described in the title, I'm running a Ryzen 5 3600 in NZXT's H1 case with the built-in 140mm AIO and PSU (detailed specs below). This has been a long-running issue, but has slowly gotten worse over time. My CPU temp stays locked at 75C+ while at idle. While running a game--for example Destiny 2 @ medium settings--it's pegged at 96C and won't cool down. Have tried a couple of different solutions (listed below) but with no success. This is my first PC I built myself, and although I'm pretty technical, I'm struggling to figure out what I could try next for solutions. Any help/suggestions are appreciated! I've attached a screencap of HWMonitor, programs currently open are Discord, Chrome, and HWMonitor. Specs: AMD Ryzen 5 3600 AMD Radeon RX 5500XT ASRock B450 Gaming-ITX/ac 16GB DDR4 RAM NZXT H1 Case + 140mm AIO + 750W PSU Windows 10 v22H2, build 19045.3930 1920x1200 display Can provide additional details if necessary Attempted solutions: Adjusting fan curve Cleaning dust filters Removing top/sides of case Unscrewing AIO+Fan and letting run outside case Tightening screws on CPU block Replacing thermal paste on CPU Updating drivers Additional Details: Unsure if these details are related, but here they are just in case. 1. I get a lot of instability, PC crashes/BSODs somewhat frequently (once every couple weeks or so). Many times, I will need to reinstall my graphics drivers for some reason, not sure why. Again, could be unrelated. 2. Was subject to a lot of power cuts ~1 year ago, over the summer running a portable AC on the same circuit would trip the breaker in the apartment, causing immediate shutdown.
-
Hi all, I recently switched from a Ryzen 5 3600 to a Ryzen 7 5800X3D. It may be important to note that I initially upgraded from a 3600 & RTX 3070 to a 5800X3D and 7900 XT, and was running a stress test with Prime95 & Furmark at once on a 750W PSU. The computer immediately shut down, so I upgraded to a 1000W PSU and don't have this issue anymore. They both run fine at the same time (could this have damaged by CPU/motherboard causing this issue?). I was idling fine on the 3600 around 35 C, but this 5800 idles around 65-70 C. I've asked around on Reddit and other forums and everyone seems to agree that these temps are way too high. From what other people post, I should be seeing around 30 to 40 C idling (when I say "idle", I mean with Discord, Steam, and HWiNFO open at once). I've tried the following: Disabled A-XMP in my BIOS (made no difference, so I turned it back on) Updated my BIOS Reset CMOS Swapped out the kryosheet I re-used from my 3600 for a new one Changing the cooler from a be quiet! Dark Rock Pro 3 to a DeepCool Assassin IV Changing from a kryosheet to thermal paste Nothing I've done has changed the temps and I'm completely lost at this point. Here are my specs: AMD Ryzen 7 5800X3D 32 GB (4x8 GB) DDR4 3200 RAM (3200 w/ A-XMP on) (if the brand matters it's TEAM Group TFORCE) MSI B450 Tomahawk (latest BIOS revision as of this post, not Tomahawk MAX) AMD 7900 XT Corsair RM1000E 1000W PSU I'm using HWiNFO to read my temps, and reading the Tctl/Tdie temp. Screenshot If anyone can point me in the right direction here as to what I'm doing wrong, I would GREATLY appreciate it. The only thing I haven't tried is reinstalling Windows, but I'm not sure how/why that would help.
-
I just recently upgraded from AM4 to AM5 CPU (thanks Newegg BF sales), which required a whole new mobo and CPU. Got a Ryzen 7 7800X3D based on GN reviews of it and capitalized on a pretty good deal. One thing I'm running into is keeping this hot li'l guy cool along with my EVGA 3080. I moved to a Gigabyte B650 Gaming X AX board and I'm noticing that it is not keeping the tower cool enough, particularly because the temperature in the case does not kick on the case fans due to the "System1" temperature. Which I can only imagine is a temperature based on sensors on the board, not CPU or GPU. With that, my ambient temp in chassis only stays about 25-26 C, even under load, so the intake fans & exhaust fans on the case hardly ever kick on because of barely any change in ambient temps. Also, I tested some fan curves in the BIOS based on the CPU temps, but then that doesn't account for when my GPU is under load. I mostly game at 1440P, so there's a heavy load on the GPU when gaming and, even if the CPU is not getting too hot, I want the case fans to cool off the whole machine. What do you recommend I do to accommodate for controlling case fan speed (intake and exhaust) when the ambient temperature sensors aren't reporting much change? Here's my PCPartpicker list: https://pcpartpicker.com/list/NYZzFs Thanks for your thoughts! -Grey
-
This is my first post, so please excuse any formatting mistakes As the title says my 7900xt is hitting 105-110c on the hotspot when at full load. I'm using Starfield for testing as it is the most intensive real world us I can throw at it. The high hot spot temp wouldn't concern me, but its the delta between the GPU temp (Edge) and the hot spot, its about 40c which seems higher than other reports I've seen I have been able to get temps slightly under control with an undervolt without losing much performance if any. This still hasn't fully solved my issue. GPU Info Temps with Default Profile Temps with Undervolt Undervolt Settings I have seen similar complaints on other forums and websites, anything y'all would suggest to fix it? Ive seen mention of using a thermal pad as the TIM for the GPU I'm not sure i trust this. Is anyone else having this problem? Do y'all think its just a case of RMA? Thanks - LBB
-
The Hyte Y60 is a proper good case, but the side panel adjacent to the vertical GPU mount is very close to the graphics card itself. For larger GPUs, it makes more sense to take off the side panel for thermals. I was wondering if anyone knows of any services that would be able to make a replacement side panel with vents for better GPU temps? Acrylic would work here, so it doesn't have to be tempered glass.
-
Grettings, recently I run out of thermal paste and I was wondering what was the best thermal paste at the moment, but most important i'd like to see some bad/good experiences and what thermal paste do you recommend for each situation. I was using the artict mx-4, it works fine, but I listen that it doesn't work very well over time on direct die, like gpu or laptop chips. I'll apreciate your comments and feedback, ty. (and sorry about my bad english) -
New to LTT forums so I hope this is the right place to ask. In HWInfo, my normal "GPU Temperature" is usually fine, maxing out at 70C, but my hotspot temp consistently gets to 90-95 under moderate load, and playing intensive games like Cyberpunk can bring it up to 105 or so. I've tried opening up my case and pointing a fan on full-blast at it to see if it was just an airflow issue, but that only decreased temps by 2-5C. I've also tried some of the basics like updating drivers and dusting my card with compressed air. The only thing left I can think of is to repaste, but I'm not an expert at this stuff and I don't want to ruin my graphics card in the middle of an already busy semester. In case it matters, my card is the MSI Ventus XS OC 2080 Super.
-
I've noticed recently that my gpu has been running a little hot. I'm planning to take the heatsink off to do some cleaning and repasting, however in fear that I may ruin the factory thermal pads on the vrms and vram chips, I want to buy thermal pads just in case. I can't find any documentation about how thick the pads should be on the internet, can anyone help?
-
I built a new gaming/coding PC for myself a few days ago and as I was stress testing to see if everything is good, I see that when I run the Aida 64 stress test my CPU shoots to 98°C immedietly. However, the clock speeds tends to stay always between 5 and 5.1 GHz. The CPU usage is consistently at a 100% regardless. What's concering me the most is that I read that AMD CPUs tend to thermal trottle when it reached 95°C. I updated my BIOS and all drivers and now I can see that the CPU maxes at 93-94°C while maintaining the same clock speeds with 100% usage as before. Do I need to updgrade my CPU cooler of something? I'm currently using a Cooler Master Hyper H412R in an Corsair 4000D airflow case. I've only 1 intake and 1 exhaust fans but I've orderd more fans for the case. I'm also running a 4060 which does not go above 60°C. Also, as I'm here. Can you suggest me a Linux distro which could be stable for NVIDIA GPUs? I've installed ubuntu but I'm unable to boot into it if my monitor is connected to the GPU (DisplayPort and HDMI) but I am able to boot into it if its connected to my motherboard. It's annoying to switch the cables everytime as I play games on Windows and Code in Ubuntu. The WSL2 in windows isn't detecting my GPU and I'm left with no other option other than dual booting.
-
Got an Asus TUF OC 7900XTX It's kinda beasty and can easily reach 2750MHz on a 1.1v undervolt, at 420W TBP But I don't want noisy fans and limit them to 2000rpm... So in gaming GPU temp goes up to 70C, which is fine (?), but junction/max temp reaches 95-99C, I kinda wonder why, don't recall my 3080 having such difference between average temp and max (maybe cause the chip is different..) And is this a real issue or just harmlessly hot ?
-
I recently got a Topdon TC001 thermal imaging camera to use as a trouble shooting tool and it's been awesome for testing PC thermals. I had to make a wooden frame side panel with a cling wrap window to allow IR band light to pass through as the regular glass is thermally opaque and highly reflective. The cling wrap emits weak thermal radiation which is nice as it shows the heat pattern that would be spread across the glass while still being transparent for the most part. I plan to test several different flow regimes as it is useful to understand how stagnation and local pressure effects heat in the case. This is the first experiment I ran while playing Diablo 4, trying to see if my old idea of back to front case flow was working how I imagined it. Turns out yes, it's working as planned with most of the hot air exiting into the dead space at the top front of the case. I know this will cause higher dust ingestion but the goal is to provide the best possible cooling potential to the CPU and GPU. No air gets used twice when setup this way and the motherboard stays cooler over all. The idea behind the bottom fans is to push cool air under the GPU cooler preventing recirculation of the hot exhaust air back down between it and the case panel. The F9 mounted on the empty case slots is crucial in this role as the back corner of a case is historically bad for stagnation and recirculation. There's a divider between the bottom front fan and the top two which prevents mixing of the flows inside the case from which has space for a full rad and fans, without this the case rapidly heats up due to the same hot air being passed through the GPU over and over. I've had a lot of push back about running opposed flow top fans and feel vindicated now that I can see the air flow across the ram from the top rear 140mm to the top front 140mm. I will try flipping the top front fan later today and see what kind of difference it makes. Top front exhaust small.mp4
-
Hello, Temps of my WD BLACK SN770 2TB are really high, not sure if its a sensor issue. It's installed in an Asrock X670e Taichi, slot M2_3. I have a Silicon Power 2TB UD90 installed in the slot below and its nowhere that hot.
-
I have a PC which had a apu 2400g, I ran it for 4 years and no issue came, recently I bought a 3060 which ofcourse bottlenecked my cpu but I have to take the risk, I planned to upgrade the cpu too. Now the problem came with thermal temps, I noticed that my cpu was getting too hot after pairing with 3060, like 50-65 idle and 100 crossed (103 max I saw) with cinebench, now I know cinebench runs the cpu 100% but that was concering. I had an ambient temp of 40°C (I live in India) so I decided to upragde my cooler, I upgraded it to Coolermaster ML240 AIO cooler, after it had installed the idle was still somewhere 40-55 but I considered it normal as the ambient was also close to 39-40, and cinebench it was going 75-80, and games was 50-60. But after enabling game oc mode of my MSI B450 gaming plus and x-amp profile 1, It rose to 50-60 idle and 55-75 gaming, cinebench it came with 85 max. Should I be concerned or not. Please its been eating up since I saw the temps of gaming. Much love Kaushal
-
-
I am having a problem with thermal throttling on my Palit Gamrock RTX 3070, last time I changed my thermal paste I saw the thermal pads are crumbling. But I didn't pay much attention to them. Now the issue is this my GPU thermal throttle on the temp limit way before reaching the actual limit (Games stutter also). Maybe the memory modules are overheating? See attached pictures. If this is really a thermal pad issue, which is probably nonexistent now. Should I buy 2mm pads to replace all of them? Another problem is that my RGB strip is malfunctioning, halfway through one strip on one side. Video of my Card RGB Malfunctioning. As you can see half of it is of the same color, and it sometimes turns completely off. Which one should I buy to replace it? I can't find anything on it on the Internet, there are RGB, RGW, different pins, etc. Any help with any of the issues is appreciated. I am lost, good gelid thermal pads are expensive here I don't want to waste money.
-
I've recently built my first pc and am occasionally having some unusual thermals . I am unsure if this is normal so first my build: Asus b550, ryzen 7 5700x, radeon 6750xt 3xfan, 750W 80+ gold PSU, samsung 990 pro SSD, 1 TB hdd, 3 128mm front fans, 1 128mm back fan, some small tower-style cpu cooler, some sort of antec case that doesn't have good front airflow. The weird thermals i've run into include: GPU hotspot reaching about 95C max under extreme load (the average is about 88C), i don't know if this is okay for the hotspot, the average GPU temp is super low like 60C. My CPU would run at about 80C when gaming but it does sometimes spikes to like 90C, and this is without PBO. All of these temps i've evaluated using Hwinfo. My main question is what's going on with the GPU hotspot? My old builds hotspot was never above 85C. Also removing the front panel to increase airflow didn't really change the hotspot temp, but it did lower the cpu temps by about 5C. Is the CPU temps randomly spiking normal during gameplay? Should i reapply thermal paste? Is this possibly a software error (i have newest bios and drivers and everything)? Sorry if i seem really confused, i have never used a AMD gpu nor cpu so i'm unaware if this is okay for AMD specifically.
-
Hi! Sorry for the vague title but I didn't know how to specify the issue I'm having. I've recently bought an Asus Vivobook Pro 14 (M6400RC) and have a problem with FPS drops while gaming. My first thought was that it's probably because of the thermal throttling but after checking the temperatures I'm not sure what's causing those drops. I recorded a few minutes of gameplay with Riva Tuner overlay, here's the link: https://www.youtube.com/watch?v=lhf-vrc3jR0. As you can see, the first minute or so is fine, I'm getting 50 - 60 FPS but then it suddenly drops to 10 - 20 FPS, stays that way for around 5 minutes and after that the performance goes back to normal. You can see the CPU temperature rising before the drop but it goes up to around 85 degrees, which for a laptop (especially such a thin and light one) is not a bad result and it seems weird to me if it started to throttle at such a temperature. The GPU temperature is also fine, it sits comfortably at 75 degrees, VRAM is also not utilized completely, the same as RAM. Everything seems to be in order, yet I still get those drops. Am I missing something? If anyone has an idea what's going on here, please share it with me. Any help is much appreciated. Specs: CPU - AMD Ryzen 7 6800H GPU - NVIDIA GeForce RTX 3050 4 GB DDR6 RAM - 16 GB LPDDR5 4800 MHz Storage - 512 GB SSD M.2 PCIe OS: Windows 11 -
Hello everyone! I recently got myself a Ryzen 9 7950X3D in order to really max out my gaming performance, especially for VRChat. However, I have noticed that on my system, the CPU gets hot rather quickly. Sometimes I get spikes to roughly 80°C whenever something happens in the background. Updating to the newest Nvidia Game Ready drivers also made the CPU hit 90°C for a few seconds which seems very odd. I am using the Be Quiet! Silent Loop 2 240mm for cooling the processor. It's top mounted inside my Fractal Meshify 2 Compact RGB case. Maybe the cooler is not working properly? Fans do spin and the pump is at 100% according to Fan Control. I also made sure to benchmark my CPU with Cinebench to test its power, and that made it hit 90°C, max temp at 90,3°C. Reviewers don't reach 90°C usually and max out at like 87°C-ish most of the time. My AIO pump fans are spinning at 100% as well. I score 35652 pts for multicore and 2006 pts for single core performance, not exactly what I should be getting as well with tests usually reaching 36500 pts and 2065 pts online. So if any of you just recently bought the same CPU, do you guys get similar temps to me? Most of what I see online is not as high as my temps are. I am getting worried since the TJMax for this processor is 89°C which I am clearly above. My setup consists of: Seasonic Vertex GX 80 PLUS Gold 1200 Watts, Ryzen 9 7950X3D, Asus ROG Strix X670E-E Gaming WiFi, Be Quiet! Silent Loop 2 240mm with ARCTIC MX-6 paste, G.SKILL Trident Z5 NEO 32GB DDR5 6000Mhz CL30 RAM and Windows 10 64 bit. I am using the up to date BIOS and chipset drivers from AMD. Eventhough it's in German, I have attached some monitoring results from HWInfo64 during my benchmark with Cinebench. Thanks in advance!
-
I'm interested in your thoughts on this. Cases with restrictive front panels are like the Thermaltake Versa H17 and Deepcube Macube 110 shown in the images below. How do thermals in these compare with Open Airflow front panels such as the DeepCool MATREXX 40, Versa H18, etc.? Do you think it really matters? If anyone has their own data or Youtube reference videos on this topic, feel free to share.
-
i need some help with buying the right thickness of thermal pads for the vram on msi 1070ti gaming x. Thanks for any help.
-
I have an rtx 2060 super FE model grpahics card, and in my opinion, the temps have been.... not great, especially because I enjoy overclocking and the temps don't really allow it too much. I am considering replacing the stock paste with something aftermarket. Currently I have half a tube worth of noctua nt-h1 What would be the temp difference between stock and ht-h1? would it be worth it? or should I go a bit higher with something like conductonaut liquid metal? If neither of these sound like a good idea, can someone recommend a good paste/liquid metal or other way to decrease temps?(I don't have the money to go with water cooling)
-
I'm planning to repaste my laptop (specs below) as it runs a little too hot for my liking. Running at 90C and above even at idle. However I'm unsure of what to do on a particular aspect of the process. From videos and images (also below) my laptop uses thermal paste on components around the CPU and GPU. I've got a pretty good grasp of applying CPU and GPU thermal paste so that isn't an issue for me. To the point, for the components around both chips would it be sensible to replace the paste with a suitable thermal pad? If not what thermal paste would be appropriate for those components? Laptop Specs: ASUS ROG Zephryus M GA502DU Ryzen 7 3750H GTX 1660 TI Max-Q 16GB RAM
-
Hi everyone, I'm in some sort of an impasse, my XPS 13 9370 (i7 8550u, 16GB, 4k display) hasn't seen any maintenance in about 2 years (it was my dad's and I bought it from him at half the price) and the temps have been getting higher, idleing at 50° and under light load at 60°. Now, of course I'm going to get rid of every possible bit of accumulated dust I see, but my main question is should I get the Thermal Grizzly Kryonaut for repasting the CPU or should I get the Thermal Grizzly Conductonaut liquid metal and give it a shot? (Also, I'll be switching the thermal pads from stock to Thermal Grizzly Minus pad 8's for better results)
-
I am wondering if my intagrated GPU should remain off for slightly better thermals. My main goal is keep my laptop fanspeed as LOW as possible most of the time, I've done literally zero gaming on the machine because the graphics are about the same as my PlayStation 4 anyway. -
Hello everyone, Most posts I see discussing Ryzen 5000's series include problems concerning idle temps / thermals in general/ boost problems etc... After a deep-dive into my 5950x this is what I learned in regards to getting the most out of my system and solving those issue. System spec: CPU: 5950X Cooler: NZXT 280mm AIO Ram: 32GB G-Skill 3600 Mhz GPU: RTX 3080 founders PSU: 850W Gold Case: Fractal Meshify S2 General purpose: Programming/engineering and some gaming. Beware of motherboard software packages! As an example, as a part of my Asus X570E software package, it installed an app called "Game First" which other than handicapping my WIFI also kept pushing for a boost frequency of 4.35Ghz preventing the CPU from ever truly idling. Uninstalling this piece of junk alone has reduced my idle temps from 49 degrees (Celsius) to 34c with the CPU being able to automatically under-volt to 0.95 with 3.6Ghz clock speeds. Old (non-updated) service (background) applications tend to cause the same problem. It is best to set your system to its default properties and use procMon/task manager to track and kill apps which unnecessarily cause an all core boost. Apples iCloud backup has also been marked as a possible problematic sw. Bios - other than updating to the most recent BIOS, some manufacturers set the default v-core to 1.4V which might be considered to be a bit too much. I did the following: 1. Enabled PBO with curve optimized set with a negative (all core) value of 12. 2. Undervolted the CPU by 0.1. 3. Set FCLK and RAM speeds to 1:1 ratio (in my case 1800Mhz). Results Idle: 34c (completely quiet) Load: 70c (Cinebench R23 30 minute stress test) All core boost: 4.6 Ghz consistent Single core boost: 5 Ghz R23 Score: 25512 Without using an under volt: Idle: 38c Load: 74-75c All core: 4.65-4.7 Ghz Single Core: no change 5 Ghz R23: 26110 + Random BSODs Note: I'm aware I can probably remove the under-volt and reduce my PBO to prevent the BSODs and maybe even gain a little bit of performance but I prefer a cooler quieter system than an increase of 2-3%. Hope this helps!
.png)
