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

ColdCase

macrumors 68040
Original poster
Feb 10, 2008
3,365
276
NH
I thought I'd ask here in case someone may shed some light on how well FCPX stabilization runs on a MacPro in comparison to my current machine. I'm not sure I'm reading the published performance benchmarks correctly (I haven't found any that specifically include stabilization, dunno how it differs from other effects).

I am currently editing on a 2014 rMBP 15 inch 2.6GHz Intel Core i7 with 16GB RAM. I have been happy with its performance, but am now getting into stabilization of relatively challenging video clips. A 10 minute clip takes 60 minutes to process using almost of all 8 so called cores (780% CPU). There are times, I think, I've noticed perhaps CPU related temperature throttling as the CPU utilization drops to 70%. I know just enough to be dangerous, however. I've tried "CoreMelt Lock and Load" without noticeable time improvement (perhaps more efficient code but then it seems to max out at 380% CPU use) and has undesirable artifacts.

I've read a number of posts here, some reviews, and benchmarks and it looks like an 8 core MacPro with dual D700s and at least 64 GB of RAM may be the sweet spot for my application. If I understand the benchmarks and comparative performance sites like Bare Feats reports on, I should expect at least 50% better performance. i.e. that 10 minute clip would take about 30 minutes on this config Mac Pro. Is that in the ball park?

Worth the money and buying timing are subject of another thread :)
 
Yeah, it will cut your rendering times quite a bit.
Achooley your rMBP has 4 cores with 8 threads.
The 8 core MP has 16 threads.
A really big help is going from 16GB to 64 GB of memory.
I'd add 2 SSDs in raid 0 for another speed boost.
Good luck.
 
If you can post a short video here for us to download and what (and how) exactly you want to do. May be some people can tell you how the nMP perform, you can then compare the result to your current Mac, and decide which one is the best for you.
 
  • Like
Reactions: ActionableMango
Thanks for the suggestion. I tried to create a short clip that would be easily posted and would take much time as it is an imposition, but once I get less than 20 seconds or so (HD video which is over 200 MB), FCPX only does one pass analysis, longer clips seem to be three passes. I wonder if the apple store would let me test the performance on a store MacPro.
 
Thanks for the suggestion. I tried to create a short clip that would be easily posted and would take much time as it is an imposition, but once I get less than 20 seconds or so (HD video which is over 200 MB), FCPX only does one pass analysis, longer clips seem to be three passes. I wonder if the apple store would let me test the performance on a store MacPro.

Long time ago, I went to the Apple store and ask them to let me test test the performance of the Mac mini (just play back a high bit rate Full HD movie), they allowed me to do it.
 
As a follow up, I got access to a 8 core (3.0 GHz) with dual D700s, 32GB RAM, 1TB internal drive Mac Pro (2014 vintage) and tried just a couple things. It was not my machine so I didn't want to push or adjust it and there were not any performance indicators available beside activity monitor. All I did was turn off FCPX background render. In many situations, the Mac Pro activity monitor was reporting ~1500% CPU usage where the rMBP activity monitor reported ~ 780%

Working files were all on the internal drive or an Samsung 850 Pro SSD in a TB enclosure.

Handbrake close to doubled its performance (Master to AppleTv3). An hour rMBP transcode took about 30 minutes on the MacPro.

The delta to stabilized performance was not as impressive, 10 maybe 20% depending. These test clips take three passes, A clip that took 60 minutes to process on my rMBP took 50-55 minutes on the MacPro. Perhaps I missed something in my haste, but I was a bit disappointed.

Not scientific, just observational and an overall estimate, perhaps not useful. The Mac Pro did feel a snappier when editing with no background processing going on. The Mac Pro fan was a lot less noisy, but it seemed to throw more heat :)
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.