Install Furmark

0 views
Skip to first unread message

Adriana Gowen

unread,
Aug 5, 2024, 2:15:39 PM8/5/24
to anpytipan
Ilike the screenshot built-in function. Like under Windows, just hit the print screen key or ATL+PrtScn (for the active window) to capture the screen. But Ubuntu goes further because it automatically opens a screenshot utility with the screen capture in. Then you can quickly screen-capture and save the image.

I also play a bit with compiz-fusion, the 3D desktop manager (see HERE and HERE for some madnesses with compiz). The core of compiz is installed by default on Ubuntu but to get cool effects you have to install some extra plugins. Not a problem: sudo apt-get install compiz-fusion-plugins-extra can fix it ?


For packaging I suggest you to provide at least a .deb and a .tar.gz. Quite a few people use .rpm distros as well.

Tip: a program called alien is useful to convert packages to other formats, like .deb to .rpm or .rpm to .deb.


Thanks! If I read the information correctly, this tool is designed to verify GPU memory stability. I am not sure if this task is GPU intensive. OTOH, graphic renderer such as Furmark has pushed TDP to the limit.


fwiw i keep a windows install around for new PC build testing. Mostly i run my workstations as headless Ubuntu boxes, but for new builds i find the Windows stress testing and benchmarking tools are a bit easier for me to use and understand the results, as well as compare with community results.


It seems that there are many users who have difficulty uninstalling programs like FurMark from their systems. Some experience issues during uninstallation, whereas other encounter problems after the program is removed.


I have replaced my ATI HD4850 graphics card with a Nvidia GTX460. I uninstalled all previous drivers and ran Driver Sweeper in safe mode. I then installed the latest Nvidia driver for my card. Playing a 3-D game I was only getting 2fps so I dtried testing the graphics with Furmark which I have used many times. It would not load saying I needed OpenGL compliant graphics controller. I then checked my Pc with OpenGL extensions Viewer 4.0 s/w and it confirmed that I was on version 1.1.


Even when viewing the 2d menu screen within the flight sim the mouse pointer movement is delayed, like a half-second after I move the mouse each time. This it did when set to OpenGL as well. As soon as I exit the game and am back in Windows environment the mouse pointer is fine.


I also encountered this same issue and updating device drivers (tried several version numbers) did not resolve the issue.

I resolved the issue by removing ALL Nvidia installed software under add remove programs/Apps having to reboot 3 times between uninstalling the Nvidia programs. After all of the Nvidia programs were uninstalled and the resolution defaulted to the generic driver I re-installed the driver and that resolved my error.


I got my GTX 1080 and I would like to stress test it as I don't have use of it with DAZ Studio as I only have the General Release. The problem is that it is installed as the second video card. I tried FurMark and Kombustor and I didn't see any way to select individual cards or even get them to stress test the second video card.


I tried GPU-Z but no matter what card I have selected the test is done on the primary card, my GTX 960. It is the same behavior I am seeing with the other stress test tools. The card is visible to the OS, DAZ Studio, etc. Even the stress test applications can see it and report on it. I am baffled.


But you're going to have to install something, and testing it with HWMonitor and Daz Beta gives you real world data; it will be what you're using it for so relevant to you, not a collection of generic tests, and you can uninstall beta after you're done.


Can you not simply disable your primary card temporarily in Control Panel so that the system defaults to the new card? That is easier and quicker than physically removing anything, and when the testing is done you can re-enable the primary card.


No need to actually disable 960, just set 1080 to primary display temporary and plug monitor into it. As long as you don't hit the 8GB limit (minus running Windows desktop) when rendering you can even leave it primary and it will add the 960 automatically for rendering as long as the scene file fits into the 4GB limit too - vice versa it probably hits the 4GB limit far sooner if that card runs Windows desktop too, not using the additional cuda cores from 960 anymore then. If your scenes come closer to the 8GB limit often it would be recommended to keep the 1080 free from running Windows desktop, but I guess you more often reach the 4GB limit than the 8GB (once Pascal support is added to release version, and until then you free some extra memory from 960 for rendering when using 1080 for desktop) ..


Note that if you are specifically told to use HWInfo for game logging, ignore the steps pertaining to Furmark and Prime95 (i.e. follow steps 1, 4, 5 and 6). If told you are doing a stress test, you must download all 3 programs and follow all steps accordingly. None of the programs are optional during a stress test.


All 3 programs need to be running for at least 10 minutes, or until your PC crashes. Start the timer after you get furmark running for the best results. Ensure the Furmark window is on top, and do not use your computer during the 10 minutes.


I recently recycled a few RX580's I bought during the height of 2017-2018 coin mining craziness. Back then RX580's sold for around C$600, turning out at aronud 30-ishMH/s. At that rate, we figured that we'd break even with ETH price above $800. But maintaining many GPUs and have them run with high-enough stability is troublesome. We even got fire alarms, then there was a few instances that the PCIe riser or MOLEX power cable/connector just melted. The best for personal safety we stopped the mining process.


Since GCP (Google Cloud) is kind enough to be out of resource frequent enough, plus it is just impossible to get a hand on RTX 30xx card at non-ridiculous price at the moment, we figured it'd be a good idea to look at AMD's ROCm offering. Deep learning is a lot of fun, but it is not fun if we can't get GPUs. Hopefully this will do for now.


As of March 2021, ROCm support for PyTorch is pretty good. PyTorch even has added a beta-build for ROCm. But AMD magically decided to cut off support for ROCm at Vega. That means non of the pre-Polaris (Polaris, Ellesmere, Baffin, etc.) do not work for sure post ROCm 4.0. I also remember seeing posts about support being iffy at ROCm 3.8+. BUT ROCm is also NOT supported for the latest GPUs (big Navi, at the time of writing). AMD, please spent more time and money on supporting the deep learning community, so researchers are not subject to the mercy of nVIDIA.


If you follow the official AMD guide for ROCm installation, you will see issues installing PyTorch. Along the lines of "some binary cannot be found for current device", referring to the discountinued support. One such post is here. Here, gfx803 is RX580.


For some reason in the last couple months Photoshop and Bridge (mostly photoshop) both regularly cause my computer to crash and auto restart the system. The program isn't freezing or crashing that I can tell, the whole computer just restarts. I don't think its my system since this is the only software that causes it. I did stress tests with Furmark, Prime95, CPUID and Cinebench. Temps are all fine even during those tests. I've used PowerDirector and blender which are resource intensive without issue.


This current project I'm working on has 70 photos in the folder. I am only opening 10-12 photos at a time (25MB each). I use bridge to preview, open in camera raw, adjust, then open in photoshop. Sometimes is restarts while opening, sometimes while working on photos - no consistency I can tell aside from just using photoshop.


I'm not familiar with the AMD platform, but it has come to my attention that they very aggressively install gaming components/modifications to everything from BIOS to device drivers. This may explain why there seems to be a disproportionate amount of issues with AMD systems compared to Intel.


I'd start by going over all installed software and remove everything you don't need. Update the drivers, in particular the GPU driver. Go to Nvidia's site and download the latest Studio driver, not the Game Ready driver. When you install it, check "clean install" to remove all traces of the old driver. Uncheck the extra components in the driver package (GeForce Experience and PhysX). You don't need them. If already installed, uninstall them first.


Ok I updated my nvidia driver (was 532 current is 552). I also updated my chipset and bios based on what I think you were saying about AMD. Its definitely not heat - I have a custom water loop with 2 radiators and 8 fans. Like I said even in the stress tests the temps were fine (below 60 C cpu and 80 C gpu).


Hey so weird event today, I installed a fresh version of atlasos on this laptop about 2 weeks ago and today I hop on it to play around with it after not touching the laptop in idk must be at least a week (was powered on the whole time) and the thing is freaking out stuck at max disk usage across multiple normal processes (nvida/windows processes etc) and at a 100% ram usage as well.


thought at first it may have been some malware attack or something though idk how thats really possible when it was literally just a fresh install of windows with atlasos installed only other applications I have on here was a ninite install package of opera/winrar/notepad++/paint.net/furmark and cpuid.


Many tools are offering CPU and GPU benchmarking in windows 10 and windows 11. These software results are very reliable and can really push your system hardware harder. If your PC is overclocked, we recommend using software like 3DMark to see if the temperature is pushing out of the limit you want and frame rates are dropping or stable.


focus on the temperature section, it will slowly increase the power usage of your GPU. You can also see the GPU temperature will be slowly increasing. This test normally lasts for 30 minutes, if your computer or laptop manages to run this test for 30 minutes, without giving you any warning sign, that means your graphic card is fine and performing as it should.

3a8082e126
Reply all
Reply to author
Forward
0 new messages