compressor slower than handbrake by a lot with my project?

I was unfamiliar with NVENC... it seems to refer to encoding using NVIDIA GPUs.

Obviously that won't be possible in a Mac.

It is understandable that using dedicated hardware in the GPU would yield a quicker encoding than just the CPU, so that 24 vs 42 minutes is easily explained.

The Activity Monitor is really kind of meaningless, especially with Apple Silicon. The software to hardware interface may be completely different for the two applications, so comparing AM behavior doesn't really tell you anything. Some Macs have hardware decoding/encoding for specific codecs on the chip. Others don't. Some have H.264, some have HEVC and H.264. Decode is a factor in performance speed as well as encode. Drives are a factor also.

You just need to run actual tests for time and quality on the media and see if it works for you. You have both applications you can test them against each other.

There's another point to consider - Handbrake utilises 100% of CPU, and almost no profile can engage GPU. So with 100% CPU load modern M1-M2-M3 Macs can shoot up to 108C temperatures on CPU, then fans kick in, and it becomes more or less manageable but still Handbrake stresses CPU to max. While COmpressor is never using 100% of resources perhaps with idea to keep overall responsiveness of OS high for other concurrent tasks. So Compressor is more gentle on you Mac and less likely to fry it with a 2 hour long export!

You showed us the settings but apparently did not say anything about either the source video. You also do not seem to have actually run this through HandBrake, just relying on an estimate, if I understood correctly. Is that right?

The HandBrake images are not from the same computer, so there also that extra variable to consider. You didn’t say what that pc was.

You should run some “fair” tests by compressing the same videos in the same mac - HandBrake runs on the Mac and is free to download.

Actually I am curious as to how HandBrake on the mac compares to the pc, too.

Any compression where the built-in dedicated video engine of your Mac can be used is likely to be much faster in Compressor. If done on the CPU and GPU it may be close.

What is the frame size of your video? I am thinking that if it is not a standard one like 1080p (1920x1080) or 4K (3840x2160) or something, it may bod down in the CPU.

I just did a quick test exporting a two and a half minute video (4K ProRes) to 4K HEVC and it took only 42 seconds, never going above 60% of a core in CPU and about 30% GPU, according to Activity Monitor.

I will try a different pixel size and aspect ratio to compare, but only later, as I have to leave for a few hours.

EDIT: The test above was on an M1 Max Mac Studio.

There's a trick that can work wonders on export speed especially if 3rd party plugins like BeautyBox are involved: you do export from Final Cut Pro itself using compressor settings (google how to set compressor settings as export destination in Final Cut Pro). Sometimes this cuts export times in half.

Here are two tests done on M1 Max Mac Studio

1) 15 minute 1080p, 25fps video compressed to 1080p 8bit HEVC 1000Kbps

Compressor: 1 minute 8 seconds - using around 60% CPU and 30%GPU (*)

HandBrake: around 2 minutes (sorry I did not time it exactly) - using 900%+ CPU and no GPU (*)

2) 15 minute 1080p, 60fps video compressed to 1080p 8bit HEVC 1000Kbps

Compressor: 2 minutes 48 seconds - using 140%+ CPU and 67%GPU (*)

HandBrake: 4 minutes 25 seconds - using around 900% CPU and no GPU (*)

(*) As reported by Activity Monitor

TGGE wrote:

Well I need it compressed more. my recording stuff should be fine though so I don't get how it can be compressed that much.

If you need it compressed more, you probably do not need to make it 60fps, do you?

You can make it 30fps, and significantly reduce the size.

What versions and OS are you using?

NVEnc

I won’t be able to do more testing until later today or tomorrow, but recently I tried converting to HEVC and it was extremely fast, while using very little CPU and GPU, which suggests that it took advantage of the video engine. I would expect the same may happen with H264 as well.

Is that example: compressing a 1080p video to 1080p at 1000Kbps bit rate representative of what you need?

Also: what are the codec of the original video, in case that makes a difference?

I could try a few examples on both Compressor and HandBrake.

If you have a PC with an Nvidia card that’s what you should use if you use Handbrake as it has hardware encoding options for that system.