Hello! I’m going mad trying to figure why my friend’s computer (which is older than mine) is exporting a lot faster than mine… Here’s some facts:
We are doing the same thing, combining two video sources into one, spliting the image in center. The combined video is 41 min total. We checked to have the same parameters (in the advanced tab).
When we export it, it takes 1h06 for him and 1h35 for me…
We tried with Parallel processing or not… Same difference, but longer for both of us.
Here’s some spec of our PCs :
Mine :
Intel i7 12700KF 3,61 ghz
Processor is at 30%
His:
Intel i7 7700K 4,20 ghz
Processor at 90%
We have the same GPU (which is nearly not used anyway).
How is that possible? Why my processor is only at 30%… I don’t get it.
Intel® Turbo Boost Max Technology 3.0 Frequency ‡5.00 GHz
Performance-core Max Turbo Frequency 4.90 GHz
Efficient-core Max Turbo Frequency 3.80 GHz
Performance-core Base Frequency 3.60 GHz
Efficient-core Base Frequency 2.70 GHz
If the export is only using 4 cores then you are using only 30% of your CPUs and your friend is using 100%. I assume you are using Windows 10, which does not differentiate well between E-processors and P-processors (Windows 11 does much better), so you are running at around 3GHz while he running at around 4GHz, hence you will take around 30% longer than he, which matches what you said.
it is the Windows thread director that schedules whether to use E-cores or P-cores. this is what one benchmarker had to say:
Windows 10 doesn’t interface with Thread Director at all and Windows 11 does, this would suggest that the issue lies with Intel’s Thread Director, not Windows 11 itself. We also tested with the E-cores disabled completely, and Windows 10 is slightly faster than Windows 11.
Cinebench R23 sees absolutely no difference between the two operating systems regardless of whether we toggle the E-cores on or off. The benchmark simply uses all cores available and scales nicely. Ditto with Blender and V-Ray. Corona renderer seems to have a big problem with Windows 10, where the scheduler is sending traffic to the E-cores exclusively and the P-cores sit idle. The E-Core disabled performance of both operating systems is the same. With Windows 11 and E-cores enabled, Thread Director is working its magic, propelling the i9-12900K into the same league as the Ryzen 9 5950X. Windows 10 seems to play well with KeyShot, although Windows 11 is slightly faster.
So some renderers work fine, others don’t. The suggestion to shut down the E-threads on Windows 10 kind of defeats the object of buying a system with this processor installed. But updating to Windows 11 doesn’t necessarily improve all applications.
Gotta say, an i7-7700K with hardware encoding is still one screamer of a box, even in 2022. One of the places I work is still running one. If Shotcut ever drops to eight threads or less due to code that doesn’t thread much, then those fast cores start to make a big difference in export time. A 16-core box that’s sitting half-idle with slower cores can’t touch it.