- Maximizing your color board
- Mastering speed effects
- Working with Compressor
- Learning helpful keyboard shortcuts
- Uploading videos to the web
- Setting up workspaces
Skill Level Intermediate
- [Jeff] Welcome to Final Cut Pro 10 Weekly. This is Jeff. Nick's not here this week. And I spent some money on a brand new MacBook Pro and I thought the results were really interesting enough that they were worth sharing as a tip, especially if you have old hardware and yeah, you know the message at the end, I'm gonna suggest that you buy new hardware but let's see what sort of difference it actually makes. So I'm going to be comparing my 2014 MacBook Pro, my current one.
I'm gonna look at two common benchmark tools and then I wanna take a look at some real world, two common exports, some ProRes material and some h264. I did some HEVC exports and something interesting happened. Let's find out. About four years ago, I picked up this MacBook Pro. It's a 15-inch i7 with 16 gigs of RAM. Recently, I picked up a brand new system and it too is a 15-inch machine but it's a hexacore. That's six i9 cores rather than an i7 and I've got a lot more RAM on it.
The comparison looks like this. It's a fourth gen i7 versus an eight gen i9. The GPU has been upgraded, it's a Radeon, that's an AMD not an nVidia card and it's new RAM Apple. Nowadays no one uses AMD cards. So there's a real sizeable change here and I spent similar amounts on both systems. The first tool I used was Geekbench. It's one of those industry standards for systems that can test your CPU. I also went ahead and tested its GPU under its Compute tab.
When you run these sort of tests, it goes ahead and it shows you its results in a web browser. And the comparison looks something like this. And my final results look like this. They do a test of a single-core processor versus a multi-core processor. It gives you the idea of what happens when it's just brute force versus good programming. And then I went and did the built-in GPU against the discrete GPU. My speed ranges from 120% to 180% faster. That felt pretty good for the purchase. Another major tool out there is a tool called Cinebench.
It's made by Maxon who makes Cinema 4D, and while it's initially meant for just an open GL, just a GPU test, if you go and you choose the advanced benchmarks, you'll get CPU testing as well. Cinebench yielded similar results, between about 127% to 180%. And they actually rate the open GL, the video card, as their major feature. And you can see here the frames per second, I was getting about almost twice the frames per second with the newer video card.
To be a little bit more realistic, I built a five minute sequence, and I built it in a very specific way. I included different types of footage type, from Red to XDCam, to ProRes, some h264, and a flavor of h264 called AVCHD. I did this specifically because Final Cut 10 gets some benefit from the video card for a format such as Red. The timeline looked like this, where I had each of the five different flavors in their own compound clip. And then an adjustment layer running across the top that I built with motion.
Each of the layers had a color preset and the adjustment layer happened to have on it a vignette along with time code. I wanted to give it something to actually render and actually have to process. From there, I did two of the major exports using the share feature, the Share Master File, and the Share to Apple Devices, which happens to be an h264 file. And I chose the best quality, the slower export. I also went over to Compressor and built a custom HEVC, that's H-vac, that happens to be h265, and I manually set it to five megabits per second.
In between every export, I made sure to clear all my renders, that way, Final Cut 10 would get no benefit. If you're not aware of it, under a lot of situations, when you go ahead and you do a share, every further share gets to use the actual render files. And the results were almost a little bit unbelievable. Sharing a ProRes file from that timeline on my old system took almost 10 minutes, where it took only about three and a half for me to actually go ahead and export it from my new system.
That's a 208% increase. I feel some of that is because of the GPU and some of that is because of the CPU. In both cases, everything was done on the local drives, and the drives here aren't a huge factor for the export. When I did it for Apple devices set to the best quality, you can see it really started to see the difference of the age of that older system. It was taking me almost three times as long. Again, those percentages, though, they're pretty close on this small representative sample. The one that caused me starting to dig was the H-vac one, that's the h265 file.
When I exported this on Final Cut 10 on my 2014 system, it took over an hour, where it only took five minutes under the 2018 system. And I was a little bit in shock. I ran it a second time, and all these were running Compressor directly. It caused me to do a little bit of digging, and I discovered that this system happens to have a hardware-encoding chip for h265 content, and specifically it works at 8-bit. This is a over 1,000% difference and if I was doing this sort of work, this right here would be the clear reason to buy this system.
What I ended up doing was adding a new setting, telling it to be an MPEG4, going to the video tab, telling it to go to H-vac and making sure the profile was set to 8-bit color. If I had set it for 10-bit, I'd have lost this acceleration. I then went down here and chose Custom and made this 5,000 kilobits, or five megabits, per second.
That's all I did, and you can certainly test this yourself on your own system. In conclusion, the benchmarking test felt really good, but the actual practical Final Cut 10 results for export, my biggest bottleneck, felt amazing. I generally tell people, every about 24 months or so, you start looking at new hardware. Here's what happens when you wait 48 months, and I wish I had made this jump earlier. And the last result, the H-vac result, was really eye-opening. I wish Apple did a better job of publishing which systems had this chip versus which ones don't.
If they ever came out with, say, a Mac mini with this, it would be really revolutionary, just for putting out these sort of files. h265 is definitely the future, and the question is, when do mobile devices have them in enough mass that people choose that content as an advantage? I hope this has opened your eyes a little bit. It certainly has opened mine. This is Final Cut Pro 10 Weekly, I'm Jeff, thanks for watching.
Q: Why can't I earn a Certificate of Completion for this course?
A: We publish a new tutorial or tutorials for this course on a regular basis. We are unable to offer a Certificate of Completion because it is an ever-evolving course that is not designed to be completed. Check back often for new movies.