Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

waloshin

macrumors 68040
Original poster
Oct 9, 2008
3,505
270
IMG_1955.png

Review after review it seems like the H264 export speeds seem to be the same no matter if you pay the double the price for the M4Pro interesting…
 

840quadra

Moderator
Staff member
Feb 1, 2005
9,490
6,393
Twin Cities Minnesota
Right.

Well Apple usually releases updates that fully support new hardware not long after the devices hit the Public. Would not be shocked if later releases tap into new processor capabilities a bit better.

Time will tell if this is the case here.
 

waloshin

macrumors 68040
Original poster
Oct 9, 2008
3,505
270
Right.

Well Apple usually releases updates that fully support new hardware not long after the devices hit the Public. Would not be shocked if later releases tap into new processor capabilities a bit better.

Time will tell if this is the case here.
Though you would think that a M4 Pro with more cpu cores, more gpu cores should export significantly faster than a M4.
 

840quadra

Moderator
Staff member
Feb 1, 2005
9,490
6,393
Twin Cities Minnesota
Though you would think that a M4 Pro with more cpu cores, more gpu cores should export significantly faster than a M4.
You would, but it isn't always the case.

Been watching similar videos, comparisons / etc since the Power PC days. It is amazing what under the hood tweaks can do to turn the tables.

There were issues where Core Duo systems were faster than Core 2 Duo systems in the early days. Similar had happened between M1 and M2 systems not that long ago in some strange edge cases before software was updated.

I don't code to Metal or Apple Silicon, so I don't know what tweaks are available when new software is compiled for new processors.

Wish I could compare an M4 to an M4 Pro at my home. Sadly I only have the M4 Pro, and the only other M4 in my household is in an iPad Pro.
 
  • Like
Reactions: G5isAlive

Chancha

macrumors 68020
Mar 19, 2014
2,338
2,157
Screenshot 2024-11-14 at 12.29.13.png


But wasn't it totally expected? For the 4th year in a row, FCPX H264 export times are directly related to the number of encoders on chip. The rest of the system makes no difference, the encoders do all the work and there is no other bottleneck on the rest of the system.
 

EugW

macrumors G5
Jun 18, 2017
14,968
12,935
I don't edit video but yeah... This shouldn't come as a huge surprise for straight exports to h.264, as the media engine is going to be doing most of the work.
 
  • Like
Reactions: Parowdy

840quadra

Moderator
Staff member
Feb 1, 2005
9,490
6,393
Twin Cities Minnesota
I didn't realize I hadn't installed FCPX on my Mini but was able to do some testing on my M3 Pro Macbook Pro. Appears that XI has some optimized export processing as compared to FCP X

Both times are average after 3 attempts.

SystemExport 4K HVEC 10-bit HLG Dolby Vision videoTime
Macbook Pro M3 Pro (14”) 12 Core CPU 18 core GPUExport 9:14 4K video (FCP X)5:25
Macbook Pro M3 Pro (14”) 12 Core CPU 18 core GPUExport 9:14 4K video (FCP XI)5:18
 
  • Like
Reactions: EugW

DownUnderDan

macrumors 6502
Apr 19, 2018
461
680
Hobart Australia
At this point I think the single encoder engine on the M4 and M4 pro is basically acting as a handbrake on export performance. Put the dual setup the Max ships with and I would bet the times would be identical between Pro and Max on exports.
 

EugW

macrumors G5
Jun 18, 2017
14,968
12,935
At this point I think the single encoder engine on the M4 and M4 pro is basically acting as a handbrake on export performance. Put the dual setup the Max ships with and I would bet the times would be identical between Pro and Max on exports.
? That's like saying if you put the extra CPU cores from the Max into the Pro, the Pro will behave just like the Max for CPU workloads. ie. The point of the Max is that it has extra performance features over the Pro.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.