FCP 10.6.2 - Export significantly faster on M1 Mac Mini, compared to M1 Max MBP 16”

Question

Level 1

4 points

FCP 10.6.2 - Export significantly faster on M1 Mac Mini, compared to M1 Max MBP 16”

This befuddling question just came up: It appears the latest update to FCP 10.6.2 has either super charged my M1 Mac Mini, or it has severely hampered my M1 Pro and M1 Max machines?

Using a project that used to bring my iMac Pro 18 core Xeon to its knees, I tried an export comparison between three machines:

M1 Mac Mini, 16GB, 1TB, 10Gb Ethernet

14” MBP M1-Pro, 32GB, 4TB

16” MPB M1-Max, 64GB, 4TB

All machines are running Monterey 12.3.1 and FCP 10.6.2

In every single export test, the Mac Mini way outperformed the other machines.

Project 1 is all ProRes clips, but a lot of them. Was120 separate chorus singer videos, this version had sections in chorus box effects, many tracks. Pretty difficult project to handle. 3:55 run time.

In this case, I left it un-rendered on each machine. Project on same thunderbolt, OWC NvME drive, directly plugged into thunderbolt port on each machine. All three machines played back with no dropped frames. Scrubbed like butter.

That was amazing enough.

Next, I tried exporting to H.264, to the same hard drive. I got:

M1 - 1:33 / M1-Pro - 2:04 / M1 Max - 2:01

All are exceptional times, but what is up with that mighty M1?! Tried a few times, same result. Result was identical when I put the project on each machine’s internal drive, and exported to internal drive.

Tried another project with 14:30 minutes of mostly video recorded from Zoom, so mp4 of some kind at 25p. Exported to H.264 30p, the results (same results, internal or external):

M1 - 2:45 / M1-Pro - 7:19 / M1-Max - 7:24

What does this mean? How is the lowly M1 Mini crushing these Pro laptops? There is no real differentiation between Pro and Max, but the base M1 is clearly working better. I couldn’t believe how fast it was speeding through the exports, just chewing through the footage.

All machines were silent during export, fans never reached above 1600 rpm on any machine.

Is the performance the M1 is experiencing supposed to be distributed to the Pro and Max, and it’s just not happening? Is there a setting I’m supposed to enable? Should I be buying up all of the M1 Mac Minis?!

Posted on Apr 17, 2022 8:00 AM

Reply

Answer 1

Top-ranking reply

trodenborn Author

Level 1

4 points

Apr 17, 2022 9:26 AM in response to Tom Wolsky

M1 Mac mini EtreCheck

Reply

Answer 2

Tom Wolsky

Level 10

152,780 points

Apr 17, 2022 8:11 AM in response to trodenborn

Please post an EtreCheck report https://etrecheck.com giving the application Full Disk Access in Privacy System Preferences. Copy the report an d use the Add Text button to paste the result into reply box.

One for the Mac mini and one for one of the other two machines.

Reply

Answer 3

Apr 19, 2022 6:59 AM in response to trodenborn

This is utterly surprising.

Could you run the famous bruceX test by Alex4D:

https://blog.alex4d.com/2013/10/30/brucex-a-new-fcpx-benchmark/

on these macs and tell us what you get?

This will provide something that others with similar machines could compare to.

Reply

Answer 4

trodenborn Author

Level 1

4 points

Apr 19, 2022 7:47 AM in response to trodenborn

Actually, now that I look deeper into it. It appears that the mini is maybe better at handling actual video footage, where the Pro and Max are better at handling graphic elements?

In a test with my original footage (bunch of ProRes clips, stacked), I noticed that adding a text scroll at the end of the project, changed things. Each machine hit a bottleneck at the titles, but the Pro and Max chewed through the titles faster than the Mini. In that test, the results were:

Mini - 6:23 / M1 Pro - 5:40 / M1 Max - 5:31

I will clearly test further.

Reply

Answer 5

Tom Wolsky

Level 10

152,780 points

Apr 17, 2022 9:34 AM in response to trodenborn

I'd do a complete uninstall of CleanMyMac and Chrome on all the machines and reinstall FCP. (Really I would wipe and rebuild any system that has been touched by CleanMyMac.)

Reply

Answer 6

Apr 18, 2022 12:12 PM in response to trodenborn

Have you tried completely different test projects with different types of media and effects?

It could be that the projects you are using are purely by coincidence optimised for the M1 mini.

Reply

Answer 7

trodenborn Author

Level 1

4 points

Apr 17, 2022 8:47 AM in response to Ian R. Brown

Background rendering turned off, on all machines

Reply

Answer 8

trodenborn Author

Level 1

4 points

Apr 18, 2022 10:57 AM in response to Tom Wolsky

Completely wiped the 14" M1-Pro, deleted the volume group, re-installed Monterey. After that, only installed FCP, Compressor, and Motion, and their update. Same result. Same export times, within a margin of error. couple of seconds slow on one, couple of seconds faster on the other. No real change.

Reply

Answer 9

Cofe

Level 1

39 points

Apr 18, 2022 2:28 PM in response to trodenborn

Very interesting. Only have the MBP 16 M1Max so can’t test.

I wonder though if there is more to the story. When they released 10.6.2 it says ‘optimised for the Mac Studio’. It doesn’t say that it is for the M1 in general which I did find odd. So maybe the M1Mini’s architecture is similar/same as the Mac Studio and therefor benefits but the MBP does not regardless of what version of the M1 chip?

Reply

Answer 10

trodenborn Author

Level 1

4 points

Apr 19, 2022 6:11 AM in response to Ian R. Brown

Just created a new FCP library on the MBP 14 M1-Pro. Added three mixed clips, a mov file 2 mp4's. Basic web clips. No rendering. Trimmed heads and tails to a 17:44 clip. Copied that library to all three computers. Exported to H.264 on internal drives:

M1 Mini - 3:21 / M1 Pro - 8:55 / M1 Max 8:56

I'm still not sure I believe what I'm seeing here. The M1 Pro was blanked and re-set twice, for good measure. The Mini just hammers through exports. Does anyone else have a supercharged Mini like this?

Reply

Answer 11

Apr 19, 2022 7:03 AM in response to trodenborn

Is this a standard 1080p project with no effects, transitions or titles?

Just out of curiosity I bunged several 1080p50 clips straight out of my Panasonic SD800 into a timeline lasting 6 minutes exactly.

No titles, effects or transitions.

I then exported the test project on my ancient 2012 i5 Mac mini (10GB RAM) and also on my ageing 2017 i5 iMac (8GB RAM) using Computer>Faster Encode.

The Mac mini unsurprisingly crawled along and took dead on 8 minutes whilst the faster iMac took 2 minutes 50 seconds or just under half the running time of the project.

The M1 Mac minis are supposed to be around 3 times faster than my iMac which would suggest they would take around 1 minute for the 6 minute project.

This is pretty much the same speed proportionally as your M1 mini is producing.

So my conclusion is that your M1 Mac mini is performing as expected but the other 2 computers are only achieving a third or even a quarter of the speed they should.

Reply

Answer 12

trodenborn Author

Level 1

4 points

Apr 19, 2022 10:04 AM in response to Ian R. Brown

So now, I have narrowed it down to a kind of export (H.264), and a timeline setting (1080p 30). That combination, seems to either limit the Pro/Max chips, or boost the M1. Can't tell which.

Other codecs and timeline settings I have tried so far, have more predictable results, with the Pro/Max being faster, but almost no differentiation between the Pro and Max chips.

Reply

Answer 13

Apr 17, 2022 8:06 AM in response to trodenborn

Has the mini got Background Rendering turned on . . . clutching at straws?

Reply

Answer 14

trodenborn Author

Level 1

4 points

Apr 19, 2022 7:35 AM in response to Luis Sequeira1

Now that ran more like I expected it should. I tried the bruceX test, and the results I got were:

to H.264:

Mini - 0:20 / M1 Pro - 0:13 / M1 Max - 0:12

to ProRes 422:

Mini - 0:17 / Mi Pro - 0:11 / M1 Max - 0:10

So I guess only the media my clients send to me is affected.

Reply

Answer 15

Apr 19, 2022 7:40 AM in response to Ian R. Brown

In my haste I overlooked the fact that the library on my iMac was stored on an external USB 3.0 HDD so I transferred it to the iMac's internal drive and re-ran the test.

It was slightly faster taking only 2m 40s for the 6 minute project so there was no significant discrepancy in my findings.

Reply

FCP 10.6.2 - Export significantly faster on M1 Mac Mini, compared to M1 Max MBP 16”

Similar questions