Intel Burn Test
Intel Burn Test
I'm testing my AMD FX 8350 with eight cores and it's still delivering around 32-34 gflops at maximum load for about 1500 seconds. I did a standard test earlier with about 32 gflops per 3 minutes. When I set the overclocking to Auto in the BIOS, should I have done that?
I no longer see the point of synthetic benchmarks on modern CPUs because they evaluate in ways no application will ever face, many people use versions that won't harm the CPU by removing modern instruction sets, and they don't simulate a multitasking environment where even stable 24-hour tests can fail. Using RoG Real Bench as a stability check, I discovered I can keep more operating conditions at lower temperatures but with slightly higher voltages. If you're looking for improvement, you might want to give it a try.
why do you think it can do more? i'm not sure exactly, but i'm just curious about whether my current readings are accurate. it's not like i'm planning to push it to 5ghz—it won't happen. i'm mainly wondering if those numbers i'm seeing right now are correct. around 32-34 gflops per 1500 seconds?
At a specific velocity it operates at a defined rate and efficiency, making measurement through GHz within a model specification appropriate. When evaluating across various processors, a Gflops value becomes relevant, though it may not be fully applicable since it depends on the processor in use.
I no longer see the point of synthetic benchmarks on modern CPUs because they evaluate in ways no application will ever face, many people use versions that won't harm the CPU by removing modern instruction sets, and they don't simulate a multitasking environment where even stable 24-hour tests can fail. Using RoG Real Bench as a stability check, I discovered I can keep more operating conditions at lower temperatures but with slightly higher voltages. If you're looking for improvement, you might want to give it a try.
At a specific pace it operates at a defined rate and ability, so measurement should focus on GHz in relation to model numbers. When comparing across various processors, a Gflops reading becomes helpful, though it may not be fully useful since you're limited by the processor you own. How would I proceed with testing the processor using the RoG Realbench suggested by JackNaylorPE?
I didn't say he wasn't. I'll let this information move forward and observe the outcome.
Arrowsight :
how would I proceed to test the processor using that RoG Realbench suggestion from JackNaylorPE?
1. Download it... these are my main tools for CPU testing
http://rog.asus.com/forum/showthread.php...load-Links
http://www.hwinfo.com/download.php
2. Install it
3. Launch the program.
When you open Real Bench, shift both windows to the left side of the screen. Open HWiNFO64, run "sensors only", if prompted, click "Disable this sensor". Position the HWiNFO64 window in the upper right corner of the screen. Adjust the bottom to fill the full screen height. Make the following adjustments:
-Right-click on "System" at the top and choose hide.
-In the next part, hide the last four lines starting with "Core CPU Thermal Throttling" (if you're monitoring temperatures this won't help)
-Skip the following section and hide the section after that (covers CPU Package through DRAM Power)
Now all these steps are in place so you can view everything at once. You should be able to see all core temperatures at the bottom of the window. If not, hide a few more lines. Save and Quit will save your changes.