When your PC feels laggy with Nvidia GPU drivers that keep hitting 100% under idle situations, there are several chances to cause that problem.
Either its virus, trojan on mining crypto, the NVIDIA drivers mismatch, misconfiguration on NVidia settings, Windows scaling issues and many more
After tweaking several stuff, I gave up and decided to solve this by revert back to Windows 10 with NVCleanStall. There result is amazing, now my GPU load RTX 3060 Ti is back to normal
Ideally, when your Windows is idle, your GPU load utilization should be like this
Fyi, this is running under 240hz of monitor refresh rate and using Windows 11 22H2
To solve the problem:
Reduce performance by go to Windows start menu “Adjust the appearance and performance of windows”.
Try to remove unnecessary animation and others stuff.
3. Install the latest Nvidia drivers and do not change the 4. Go to Task Manager, find Desktop Window Manager, right click and choose “Go to Details”. Find dwm.exe and right click. Set priority to Normal.
If the GPU load still high, Please check your Nvidia settings and change Power Management from “Prefer Maximum Performance” into “Normal” then reboot.
If this High GPU load still exists, the last resort is to “uninstall your GPU Nvidia Driver” and use the default driver from Windows (early version)
In my case, the 2021 drivers works like a charm for me!
This tutorial will help you to run Tensorflow in Windows 11 to use GPU for modelling, with the latest Visual Studio, CUDA and latest drivers. In this case, I’m using RTX 3060Ti Lite Hashes and Python 3.7 – 3.10 for Windows. (I use Python 3.10.9)
TLDR;
We need to install
Clean Windows Installation and GPU drivers card
the Microsoft Visual C++ (MSVC) compiler 2022
the CUDA Toolkit 11.8
the cuDNN libraries
Install tensorflow
Pre-requisites
1. Windows fresh installation to remove the complexity of issues or bugs is recommended here. I did re-install my Windows 11 in this step
When your Chia not working, the plots not detected and you facing this issue as below:
2021-11-14T01:14:03.189 farmer chia.farmer.farmer : ERROR Harvester did not respond. You might need to update harvester to the latest version
2021-11-14T01:15:34.203 farmer chia.farmer.farmer : ERROR Harvester did not respond. You might need to update harvester to the latest version
2021-11-14T01:16:44.015 farmer chia.farmer.farmer : WARNING Respond plots came too late
2021-11-14T01:16:45.076 farmer chia.farmer.farmer : WARNING Respond plots came too late
2021-11-14T01:25:50.229 farmer chia.farmer.farmer : ERROR Harvester did not respond. You might need to update harvester to the latest version
2021-11-14T01:25:53.323 harvester chia.harvester.harvester: WARNING Looking up qualities on C:\Chia\A2-1\plot-k32-2021-06-01-12-28-439ad908425b679f5b2461355304db7dfa9e2d5d43f15144f17f581de068bb9a.plot took: 166.60197138786316. This should be below 5 seconds to minimize risk of losing rewards.
2021-11-14T01:25:54.230 farmer chia.farmer.farmer : WARNING Respond plots came too late
2021-11-14T01:28:55.762 farmer chia.farmer.farmer : ERROR Harvester did not respond. You might need to update harvester to the latest version
2021-11-14T01:30:26.760 farmer chia.farmer.farmer : ERROR Harvester did not respond. You might need to update harvester to the latest version
2021-11-14T01:31:00.841 harvester chia.harvester.harvester: WARNING Looking up qualities on C:\Chia\A1-4\plot-k32-2021-07-07-10-53-1551a1b70e7d90261a19047a186a4839bac323d6e2f581e74456d9b4e2ee7e91.plot took: 236.3222620487213. This should be below 5 seconds to minimize risk of losing rewards.
2021-11-14T01:31:02.674 farmer chia.farmer.farmer : WARNING Respond plots came too late
2021-11-14T01:31:04.106 farmer chia.farmer.farmer : WARNING Respond plots came too late
Quick way to import password in Google Chrome, where its disabled by default or not found in settings. If you are using windows, you can go right-click on the Chrome desktop icon and choose Properties
Then, change the shortcut path and add “–enable-features=PasswordImport” in Target icon shortcut
In this case, I’m using Asus RT-AC59U V2 connected to main router from Internet provider. Before i’m explaining how to do it, the reason why I need to have the same IP address class across network, is to make sharing or ping between them easily.
First, I just realized any computer in the network able to ping computer A that connected to main router via LAN. Meanwhile, computer B that connected from another router can’t received ping from computer A.
After checking on “ipconfig”, I found both machine A and B have different IP address class eg: 192.168.50.xxx and 192.168.4.xxx. This is why cant ping between computer.
Fyi, I’m using both Ubuntu Server and Windows 10 in this network. Where for Windows 10, the firewall already turn off and sharing mode in LAN settings already on.
So, my router by default have operation mode as “Wireless router mode”. in WAN settings, it’s received IP from main router. But, when other PC connect to this Asus router, it will have 192.168.50.xx.
Then, I change my ASUS operation mode into Access Point mode and problem solved! All the machine that connected to my router will following the same IP address as main router!