Avva

It's an asus prime x470 pro they have a dirt cheap heatsink and the heat sink has little mass and the contacarea is poorly designed i would be running an 3700x on them i believe it to be an 8+2 phase power delivery setup with 2 phases doubled. Its not like i would run into issues with overheating i believe they are ir vrms so they work up to 105°C but im just curious.

Hi everyone, I'm currently building my PC inside a Phanteks Eclipse p400a digital case. I have a front mounted deepcool gammaxx v2 l360, with the fans installed for intake. I installed 1 120 mm rearexhaust fan. I was wondering whether to set the 2 120mm top mounted fans to intake or to exhaust. Im primarily focused on VRM heat dissipation. I have not done any testing since I'm still awaiting the arrival of my gpu. If any one has done some testing on this subject feel free to share it.

Allright so the TLP window was like this by default i unchecked 1 clamp on the long turbo side of things. https://prnt.sc/me8pp7 As for undervolting cache and core frequenzy. https://prnt.sc/me8hkt https://prnt.sc/me8hbr After setting Speedshift to 0 i regained a permanent clock of 3.4 over all cores. Previously it would turbo under a Stress Test like Prime95 but during light workloads or playing Games it would decrease to random clocks after i got it fixed sitting at 0,79 Mhz all the time. I did uncheck BD PROCHOT as it actually caused throtteling to 0.79 during testing. https://prnt.sc/me8zvt So all my Tests regardless of TSbench prime95 or whatsoever the same limitreasons did pop off but nothing throtteled. Moreover its boosting to 3.4-3.5 (i adjusted boost on 3 or 4 cores active to a multiplier of 36 http://prntscr.com/me95up) Since the temperatures stayed at 85 max which also showed the Result of Repasting the dies which i did yesterday. (arround 14-10 degrees drop in Temps) which is a neat Sidenote. Only time will tell if the issue has been solved permanently if something were to show up again i will definetly take Note of it and publish it here, Anyway im extremely grateful for ur help, as i didnt have much to do with Laptops in the past my knowledge was rather limited especially since the Bios options u can actually tweak are such a bad joke. For the rest of u who also put effort in to help me solve my issue im also very grateful. This being said i wish u guys a good rest week take care

Voltage usually shows the difference when thermal throttling is actually caused by overheating, which is not my case. It throttles at 50° the same as it does at 80 to 90. To be more precise i had undervolting tested from - (150, 125,100,75,50,40,20) For overvolting i tried with 0.1 steps to 1.1. I had turbooing set to a really low watt i believe 25 to 30 and duration on 28 secs.

So im experiencing quite some trouble with my current Asus Rog Strix Scar. The CPU (7700 HQ) throttles without any Reason to the point were its stuck at a universal Clock of 800 Mhz on all Cores. Regardless of having: 1) Power Management Cpu state minimum & maximum set to 100% 2) regardless of voltage regulation i used ITU to undervolt it to -150, while having boost time set to zero, -increasing voltage to 1.1V with turboing enabled aswell as disabled 3) I tried using Throttlestop to disabled BDProchot & Speedstep as i cant disable them in the BIOS where i cant really do anything to begin with. It didnt really matter which loads or temps the CPU was faced with, the same Story happened in AIDA64 Prime95, Benching in Deus Ex Mankind divided, i tried playing other Titles which differed in CPU demmand with the same Result. I had actually changed the Thermal Compound to some MX-4 . So its relatively clear that either the Sensor failed or the Software is some piece of cake. I did notice that the Fans actually stopped spinning at their previous RPM because i was missing the sound of vacuuming. Anyway the funny part is, whenever i press the ROG-Gamekey or whatever they named it the clock actually skyrockets to 3.4-3.6 where it should usually be. For full specs im using a Win 10 Ultimate Version Intel 7700 HQ Geforce GTX 1070 desktop grade 16 gb 2400 Mhz (prolly not running at 2400 cuz u cant select XMP in bios) 256 Nvme m.2 i believe both ram and ssd are from samsung & 1 2 TB HDD If any1 has a decent solution which doesnt require to swap out the sensor for a new one, maybe disabling it as a whole it would be really appreciated cheers.