Unfortunately, that's something more up Boris's alley. There might be a way to expose that information through openhwmonitor library but I have no idea how.
Unfortunately, that's something more up Boris's alley. There might be a way to expose that information through openhwmonitor library but I have no idea how.If we could get the same data for average frequency of P and E cores we'd have it all during the run we'd have it all with no chance for human error!
I'm not sure the compiler version will affect the scores dramatically. I expect most critical paths are already using assembly code or intrinsics. It would still remain interesting to know for sure 😀Still on GCC 13.2.0 unfortunately.
If you are feeling bored, you just need to modify the script "scripts/mingw-w64-build" to use the latest Clang and then compile HandbrakeCLI.exe for us on one of your Linux machinesIt would still remain interesting to know for sure 😀
I know, that sounds like something a boss would say, with absolutely no idea of the effort involvedyou just need to modify the script "scripts/mingw-w64-build" to use the latest Clang
My easy way is not for running Geekbench. Only for HandBrakeCLI. That's why I'm showing only the Begin Encode option.The easy way doesn't mention the Geekbench 5 prerequisite. When I go to download 5, I am taken to the download page for 6. Found the old version here: https://www.geekbench.com/legacy/
It seems to require Geekbench to run if I enter the number 3. Or maybe not, I was just hitting enter and though that meant 3?My easy way is not for running Geekbench. Only for HandBrakeCLI. That's why I'm showing only the Begin Encode option.
Didn't remove the other options from the Powershell script because I don't want to deal with things not working due to removing the wrong block of code. So while you can enter some number other than 3, it's not supported in my mangled PES
The original PES thread is this: http://www.portvapes.co.uk/?id=Latest-exam-1Z0-876-Dumps&exid=thread...-efficiency-of-x86-cpu-architectures.2597905/
I think the 9800X3D is just slightly more efficient than the 9950X. 171.5Wmin for 9800X3D and 192.9Wmin for the 9950X3D.So 9800X3D is performing at more than twice higher rate than 9950X. Could this be due to V-cache?
@lightmanek , if you get the time, can you disable one CCD and see what score you get?
Question is, is it due to V-cache (fewer power intensive trips to IOD)? Hoping for that answer.I think the 9800X3D is just slightly more efficient than the 9950X.
Sure I am away for few days but I would love to do it once I get backMeteor Lake and Panther Lake support added to the DLL.
@511 , would you like to give the benchmark a go, with High Performance power profile, HT on and off? It's really easy and quick to run (download/instructions in the first post of this thread). And you could try the Jellyfish encode too if you want to contribute more.
@Schmide , interested in contributing the score for your Ryzen laptop?
I've tried today, but my PC just doesn't boot when I try to disable CCD1. Tried 3 times and I give up as I have limited time this weekend to spare.So 9800X3D is performing at more than twice higher rate than 9950X. Could this be due to V-cache?
@lightmanek , if you get the time, can you disable one CCD and see what score you get?
Thanks for tryingTried 3 times and I give up as I have limited time this weekend to spare.
Could be a bug in the latest LibreOffice? I'm using version 24.8.4.2Any idea what causes this? It only happened to me once so far