I have been incredibly impressed with the Apple M series chips. They were introduced back in 2020, and while the initial numbers were impressive, I stuck with my iMac Pro for quite a while. It was almost the perfect editor machine. Fast and powerful with a beautiful display, and tons of pro connectivity in a simple form factor. Pull it out of the box, plug it in and go.
But then the Mac Studio came along. While it didn’t have a built-in display, it was very fast, with even more connectivity and a good bit of portability.
It took some in-depth reviews on my part to fully convince me that the move to Apple silicon from Intel-based chips was a good one. I was fortunate to do a lot of testing for things I do in post-production, which I shared in my reviews.
First, there was part one and part two of a review of the 16-inch Apple MacBook Pro M1 Max. Part 2 was a fight against my iMac Pro.
Then, it was part one and part two of a review of the Mac Studio and the M1 Ultra.
This “review” is going to be a little bit different
Once I was finally convinced of the power of the Apple M series chips, I decided to trade my next-to-last generation Intel Macbook Pro for a 14-inch M2 Macbook Pro. While slow compared to the M chips, my beloved iMac Pro is still a workhorse, so I kept it in production and went with the M2 laptop due to its versatility.
But then the M2 desktops came along with the M2 Mac Mini and the M2 Mac Studio. So I requested to lay my hands on an M2 Mac Mini and an M2 Mac Studio to have a full line of M chips.
- M1 Ultra Mac Studio
- M2 Max 14-inch Macbook Pro
- M2 Mac Mini
- M2 Max Mac Studio
This was the comparison I wanted to test with a number of editing and post-production tasks. But then I realized just how difficult it is to do comparison tests when you’re working out in the real world like we at PVC do vs. having a dedicated testing lab with dedicated writers like you might find at The Verge or a Mac rumor site.
My best intentions have put me a few months behind in getting this done. It has also brought up an interesting problem thing when looking at comparing multiple systems like this when most of them are working in real-world production. It’s almost impossible to keep them on the same OS and the same software versions. Some software requires the latest OS. But you might not want to update a production system with the latest OS. So then you’re not using the latest software.
In my case, I usually reserve my laptop (which makes up my home system as it doesn’t travel a lot) for early updates before I upgrade office systems. And while I’m usually pretty fast to update, I’ve been doing a good bit of collaboration, so I didn’t always want to update as my collaborators couldn’t update. It’s a bit of a cat-and-mouse game.
So that made for an interesting bit of testing as some of these renders and exports might have been in identical projects opened in different editing application versions on different operating systems. Would that make much difference? I don’t think the OS would make much of a difference but NLEs do speed up with new versions.
It’s also worth noting that Apple’s Apple Silicon roadmap is moving at lightning speed. An Apple Event on November 30 is rumored to bring us M3 chips in some Mac form factor did bring M3 Macs including a new MacBook Pro. We never saw Intel Macs move at this pace.
The transition to Apple Silicon brought up the Mac vs. PC debate again
I’ve long thought the Mac vs. PC debate was over, but I was wrong. Between some of the discussions I’ve had online and in-person since I’ve been trumpeting the performance of the Apple Silicon chips, I still get the “well, PCs are faster for less money” argument thrown out semi-regularly.
That may or may not be true, as I’ve never researched that argument. I have no desire to have a PC in my personal workflow. While editing is mostly the same once you’re inside of Avid, Resolve or Premiere Pro (No Final Cut Pro, though), they aren’t exactly the same. Nor do you spend all your time inside of the NLE.
But I can only offer this personal anecdote on the PC vs Mac debate. While finishing a 1080 broadcast series earlier in the year, I was working on the Mac Studio, and a fellow editor in the building was working on his PC. I don’t know the exact specs, but it was well-outfitted. We were rendering hundreds of interview clips from Resolve and finishing the show in Avid Media Composer. He was having render issues on the PC, so I was able to export the interviews on the Mac Studio. And even when timing exports, he could render on the PC; the Mac Studio won.
In the end, he brought his own Mac Studio to the edit suite to complete some of the work faster and more reliably than his custom-built PC.
What are in those chips again?
The different versions of the M chips can get kinda confusing with Max vs. Ultra vs. Pro. Only the MacStudio comes with the Ultra. But basically, it goes like this Ultra > Max > Pro > Chip. It’s that line between the Max and the Pro that gets tricky.
For the video editors, all these chips include a “media engine” with Hardware-accelerated H.264, HEVC, ProRes, and ProRes RAW. A Max vs. a Pro will get double the video encode, and ProRes encode engines.
It gets confusing, so I made this little composite chart for myself when looking at the different specs so it helps to see them side-by-side. For video editing, I would never recommend buying a base M-chip, which you can do in a Mac Mini. When you add “pro” to the product line, chips begin with the Pro chip, like in the MacBook Pro.
The systems I was comparing:
MacStudio M2: M2 Max (12 core, 64 GB ram, 38 core GPU)
MacStudio M1: M1 Ultra (20 core, 128 GB ram, 64 core GPU)
Macbook Pro: M2 Max (12 core, 32 GB ram, 30 core GPU)
Mac Mini: M2 Pro (12 core, 32 GB ram, 19 core GPU)
Your eyes could quickly glaze over 🥺 when trying to decide which spec to buy, especially if you’re on a budget. If you have to have the portability of a laptop then the choice is easy, go with the Max. And while the MacMini is a nice bang for the buck, I think the longevity and connectivity of the MacStudio make the extra cost worth it.
The biggest question on the MacStudio line is Ultra vs. Max when it comes to buying the chip. Max Tech does a good job of breaking that down in the video above. The $2,000 premium for the Ultra might not be worth it as $2,000 is a lot of money.
When the MacStudio came out, I was almost convinced that the two extra Thunderbolt 4 ports on the front of the Ultra were a worthy consideration. But as I’ve used the machine, I’m finding that most of my plugging into the front are USB-C SSD drives, so I’ve changed my tune on that a bit. And you get 10 gigabit Ethernet as a default so if you’re working in a network environment, that’s important.
The tests
These are mostly all export tests dealing with 1080 and 4K media. There is a lot of “real world” stuff here that is going out to broadcast television. If you want realtime-playback RED 8K denoise tests, you’ll have to look elsewhere. I still find that those asks aren’t the norm in video post-production unless you’re just doing it because you can. And even if you are working with such high-resolution media, if you’re doing it right, you’re using a proxy workflow.
17-minute 4K H264 source media in a 1080 timeline mulitcam. Exported to H264 YouTube preset (faster encode option where applicable)
| Software | MacStudio M2 | MacStudio M1 | MacMini M2 | 14-inch MacBook Pro M2 |
|---|---|---|---|---|
| DaVinci Resolve | 1:46 | 1:50 | 2:40 | 1:49 |
| Adobe Premiere Pro 2023 v23.5 | 2:52 | 2:02 | 5:47 | 4:20 |
| Adobe Premiere Pro (Beta) v24.0.0 | 2:47 | 1:54 | 5:29 | 4:20 |
| Final Cut Pro | 1:46 | 1:58 | 2:16 | 2:09 |
I think these results are what I would expect. You can really see those extra encoding chips at work on the M1 Ultra, but then the M2 with less dedicated ProRes and video encoding chips seems to have narrowed the gap.
Scene edit detection on an 8:44 file
| Software | MacStudio M2 | MacStudio M1 | MacMini M2 | 14-inch MacBook Pro M2 |
|---|---|---|---|---|
| DaVinci Resolve | 22 sec | 38 | 50 | Forgot that one 🙄 |
| Adobe Premiere Pro 2023 v23.5 | 19 sec | 38 | 21 | 21 |
| Adobe Premiere Pro (Beta) v24.0.0 | 19 sec | 43 | 20 | 23 |
Transcribe an 8:44 file (in seconds)
| Software | MacStudio M2 | MacStudio M1 | MacMini M2 | 14-inch MacBook Pro M2 |
|---|---|---|---|---|
| DaVinci Resolve | 43 sec | 43 | 52 | 52 |
| Adobe Premiere Pro 2023 v23.5 | 22 sec | 25 | 21 | 21 |
Topaz Video AI – upscaling 3:13 DV SD video to 1080 HD
| Software | MacStudio M2 | MacStudio M1 | MacMini M2 | 14-inch MacBook Pro M2 |
|---|---|---|---|---|
| Topaz Video AI stand-alone | 17 min 17 sec | 11 min 17 sec | 22 min 47 sec | 21 min 35 sec |
This test was done using Topaz’s preset to Deinterlace footage and upscale to HD, generating a ProRes .mov. At first, I was puzzled by the 6-minute difference in the M1 and M2 Mac Studio’s but I guess Topaz is really taking advantage of 4 ProRes engines on the M1 Ultra vs 2 ProRes engines on the M2 Max. What this does tell me is that the more a vendor can optimize their software to the specific architecture the better.
Neat Video noise reduction – 3:13 SD in a 1080 timeline with Neat applied per clip, generic profile
| Software | MacStudio M2 | MacStudio M1 | MacMini M2 | 14-inch MacBook Pro M2 |
|---|---|---|---|---|
| DaVinci Resolve | 3:49 | 5:51 | 4:36 | 4:51 |
| Premiere Pro Render ProRes timeline | 1:47 | 2:52 | 1:49 | 1:57 |
| Premiere Pro Export YouTube Preset | 1:52 | 3:02 | 1:55 | 2:00 |
One thing I noticed on the M2 MacBook Pro vs the Mac Mini is the fans in the laptop kicked on pretty quickly once the renders began. But comparing these results, it appears that Neat is doing work optimizing for the new chip generations.
Export a 22-minute 1080 23.98 broadcast program with a broadcast-safe filter
| Software | MacStudio M2 | MacStudio M1 (FCP 10.6.5) | MacMini M2 | 14-inch MacBook Pro M2 |
|---|---|---|---|---|
| Final Cut Pro ProRes | 1:48 | 59 sec | 1:12 | 47 sec |
| Final Cut Pro H264 YouTube | 1:49 | 1:58 | 2:56 | 1:42 |
Observation: Final Cut Pro is fast. This is not new news!
Export full 1 hour 22 minutes 1080 23.98 program with titles, graphics and credits
| Software | MacStudio M2 | MacStudio M1 (Resolve 18.5) | MacMini M2 | 14-inch MacBook Pro M2 |
|---|---|---|---|---|
| Resolve ProRes | 3:53 | 3:53 | 4:42 | 4:19 |
| Resolve H264 YouTube | 6:38 | 7:11 | 10:18 | 6:13 |
Export 5-minute RED R3D 8K with LUT and vignette applied from a 1080 29.97 timeline
| Software | MacStudio M2 | MacStudio M1 | MacMini M2 | 14-inch MacBook Pro M2 |
|---|---|---|---|---|
| Adobe Premiere Pro 24 ProRes | 3:18 | Didn’t test | 2:34 | 3:27 |
| Adobe Premiere Pro 24 H264 YouTube | 3:20 | Didn’t test | 3:54 | 3:25 |
One thing I kept an eye on with this RED edit was how the machines were able to keep up with realtime playback. I was quite surprised how well the Mac Mini and MacBook Pro were able to churn out full frame rate on this edit. There weren’t a lot of effects or mulitcam or anything like that but 8K scaled to 1080 isn’t anything to sneeze at. I might still choose a proxy workflow on most occasions, but it’s nice to know you might not have to on a small job or with a quick turnaround.
Do you have an M-series Mac yet?
I think we are at an exciting place in the development cycle of the Apple Silicon Macs regarding the heavy usage required for video post-production. Gone are the days when editors needed to feverishly reference something like the MacRumors Buyer’s Guide and wait for just that perfect moment to buy a new Mac so you could maximize your investment.
It feels like the M Macs will have such good longevity and be refreshed often enough that it’s no longer necessary. If you need a new machine, buy one and enjoy the benefits. This is especially true if you’re still working on an old Intel Mac. The benefits are tremendous.
On a side note, I was sitting in on a troubleshooting call with some editors and one of the NLE makers. The editor was describing their upgrade troubles, and the question inevitably came up about what hardware they were using. The answer was an Intel cheese-grader Mac Pro. The point was made that, at some point, the software makers have to cut off support for old systems. They can afford to support previous-generation architecture for only so long. Just be aware of this if you resist moving to Apple silicon. While it is an expense, you must weigh cost vs performance versus a mandatory push to upgrade if you want to work on the latest software. That isn’t always an easy decision, but it must be considered.
Now, I wonder what the results would be when running Avid Media Composer through some of these same tests. 🤔✅