Hollywood thinks new Mac mini 'could be huge' for video editors

Posted:
in General Discussion edited November 2020
With the claimed performance of the Apple Silicon M1-based Mac mini, coupled to increased working from home, movie editors say Apple is catering for creative professionals.

Apple's new Mac mini
Apple's new Mac mini


While the new Mac mini has yet to arrive in users' hands, Apple's claims about its performance are beginning to make certain Mac Pro users consider switching to the Apple Silicon machine. Reportedly, video editors are looking to see whether the M1 speeds up key parts of their work.

"Most of the editing industry is either on old cheese-grater or trashcan Macs," editor Harry B. Miller III told The Hollywood Reporter. "It seems to me the Mac mini with this new chip could easily and fairly cheaply replace all those old units."

"Fox Studios, for example, had been refurbishing cheese-graters to keep them up to date with CPUs and memory for more complex workflows," he continued. "It would now be cheaper to replace them with the new [Mac] mini."

Miller, whose almost 40 years of editing most recently includes the thriller series "Treadstone," says that he's buying one for himself. "I've used a [Mac] mini for the past 18 months... Because of the pandemic, I know there have been a lot of equipment investments into iMacs for remote work. I could see the [Mac] mini replacing them as well."

The Hollywood Reporter also spoke to Michael Cioni, a member of the Academy of Motion Picture Arts and Sciences, and senior vice president of innovation cloud-based video editor platform Frame.io. Cioni is similarly positive about the new Mac mini, but also says that it redresses the previous perception that Apple had abandoned pro users.

"[The Mac mini signals that Apple is] continuing its commitment to meet needs of today's creative professionals," he said. Cioni also noted that Apple specifically spoke about video editing in its November 10 event and said, "that's a positive narrative for creative professionals we haven't received from Apple in nearly a decade."

Apple didn't refer to video editing in the new Mac mini, but it did emphasise the M1's capabilities for it with the MacBook Air and MacBook Pro
Apple didn't refer to video editing in the new Mac mini, but it did emphasise the M1's capabilities for it with the MacBook Air and MacBook Pro


"Professionals in M&E [Media and Entertainment] now require a dozen apps working at the same time -- many of them high-end tools," he said. "The ability for M1 to smoothly run simultaneous professional tools concurrently is a response to how today's creatives prefer working; all in one kit, all at the same time."

"Having [DaVinci] Resolve, Final Cut [Pro X], Frame.io, Adobe and Nuke open at the same time is certainly the way I work and [M1 could] dramatically improve that experience," he concluded.

Miller believes that the Mac mini "could be huge" for the industry. He says the main issues for all editors, regardless of what software they use, are rendering times and preparing viewable videos for clients to review -- and that the M1 should help with both.

"They are computer-processing-intensive actions," he said. "We have to see if this new M1 chip speeds that up. It should. Dramatically. Then Avid, Adobe and DaVinci... have to use those hardware and OS advances in their software development."

While each of these video editing apps will run on Apple Silicon, Apple's own Final Cut Pro X has already been optimized for it. Adobe has said that it plans to update its major apps, presumably including Premier Pro, early in 2021.

Comments

  • Reply 1 of 19
    I agree that the new M1 chip looks like it could run any of those apps and get great performance. My question, however, is whether or not it can run Resolve, Final Cut [Pro X], Frame.io, Adobe and Nuke all at the same time... given the 16GB limitation on RAM.
    williamlondonbsbeamerpulseimagesFidonet127
  • Reply 2 of 19
    Rayz2016Rayz2016 Posts: 6,957member
    hmlongco said:
    I agree that the new M1 chip looks like it could run any of those apps and get great performance. My question, however, is whether or not it can run Resolve, Final Cut [Pro X], Frame.io, Adobe and Nuke all at the same time... given the 16GB limitation on RAM.

    That's the question.

    Having said that, a lot of it depends on what kind of optimisations they've done with the OS.

    But conventional thinking would reckon 16GB is quite tight for high end video work.
    watto_cobra
  • Reply 3 of 19
    Rayz2016 said:
    hmlongco said:
    I agree that the new M1 chip looks like it could run any of those apps and get great performance. My question, however, is whether or not it can run Resolve, Final Cut [Pro X], Frame.io, Adobe and Nuke all at the same time... given the 16GB limitation on RAM.

    That's the question.

    Having said that, a lot of it depends on what kind of optimisations they've done with the OS.

    But conventional thinking would reckon 16GB is quite tight for high end video work.
    16GB is tight for high end web browsing. But the real constraint is the gpu. Especially since it's sharing RAM with everything else. A slow GPU means far fewer real time renders of transitions and effects. This as a massive impact on speed of workflow. I don't see this Mac Mini being used for any serious editing work. But perhaps there will be a future model that has expandable memory, or at least more memory options, dedicated gpu or egpu support, etc. Has anyone seen any gpu benchmarks posted?
    williamlondon
  • Reply 4 of 19
    Why the report sounds like a paid endorsement?

    Sure a new apple silicone iMac Pro or Mac Pro will easily do the job. But a 2 port thunderbolt Mac mini for professional video editing?
    williamlondonpulseimagesBeats
  • Reply 5 of 19
    beeble42 said:
    Rayz2016 said:
    hmlongco said:
    I agree that the new M1 chip looks like it could run any of those apps and get great performance. My question, however, is whether or not it can run Resolve, Final Cut [Pro X], Frame.io, Adobe and Nuke all at the same time... given the 16GB limitation on RAM.

    That's the question.

    Having said that, a lot of it depends on what kind of optimisations they've done with the OS.

    But conventional thinking would reckon 16GB is quite tight for high end video work.
    16GB is tight for high end web browsing. But the real constraint is the gpu. Especially since it's sharing RAM with everything else. A slow GPU means far fewer real time renders of transitions and effects. This as a massive impact on speed of workflow. I don't see this Mac Mini being used for any serious editing work. But perhaps there will be a future model that has expandable memory, or at least more memory options, dedicated gpu or egpu support, etc. Has anyone seen any gpu benchmarks posted?
    I agree - with 2.6 teraflops, although unclear how that translates to real world use, seems like a modest GPU and underpowered for high end work. That said, it depends on WHEN it is too slow. If you work with proxy footage that still looks great and only the rendering to a final format is slow, that is much more acceptable since it won’t hurt the author during the creation process. If the GPU however falls flat on its face during the creation process, that’ll be an unacceptable bottleneck, especially when eGPUs are not supported.
    williamlondon
  • Reply 6 of 19
    As many points it is too early to judge. It will for sure not replace PRO machines and Apple would not shot into own leg. So Even powerful, CPU/GPU shares memory. No one knows I/O performance, limited expandability so it may be useful for some tasks as addition to PRO workflow. It can be used for video editing but question is in what resolution/codec/bitrate.  In past was done PRO video editing on much worse machines.
    viclauyyc
  • Reply 7 of 19
    I’ll just throw in my €0,02: what if the Unified Memory Architecture is the key here?

    No, really, hear me out: what do conventional PCs (even PowerPC Macs back in the day) do when they run out of memory? They do paging of stuff from memory into storage, A.K.A. virtual memory.

    The thing is, I’m guessing the on-chip memory of these things is not only more power efficient, but fast. Really fast. Maybe even much faster than conventional memory in DIMM and SO-DIMM modules connected to the CPU via its memory controller, as it resides on the SoC itself and benefits from direct, fast lanes connecting it to the CPU cores themselves.

    What about SSDs? Well, those are also really, really fast, even for such a paltry, apparently non-Pro machine like this. While they’re not exactly melded into the CPU as well, it seems Apple has done some advances in this area as well.

    So… maybe paging to virtual memory isn’t as much of an issue as it was before? And maybe said UMA is better at managing the whole process as well?

    If Apple adds a “Mac mini Pro”, or “Mac Pro mini”, or whatever they call it to the lineup, in space gray to match the later Intel models and the DTK, with an M1X/M2 chip, four TB4/USB4 ports, and, say, 32 GB of this newfangled UMA memory, expect smaller studios to pick this thing up. It’s the trascan-redux-as-lunchbox, minus the melting and inefficient dual GPUs and at A THIRD of the price, and it might very well sell like hotcakes.

    For now, this first generation is more of a proof of concept than anything else (at least as far as some less demanding segments of the professional market are concerned), but keep an open mind and your eyes peeled for benchmarks and real-world testing.
    edited November 2020 williamlondonroundaboutnowSamsonikkraulcristianwatto_cobra
  • Reply 8 of 19
    How Apple can call the 13 inch MBP a Pro machine while lacking a discrete GPU is beyond me. 
    williamlondoncia
  • Reply 9 of 19
    Eh. The media - news and entertainment - have had a strong preference for Apple products from day one. They have stuck with Apple through all their transitions - Motorola 68x, PowerPC, Intel, iPhone, iPad - and really don't consider other platforms unless they are forced to. As the software providers go where their users are, these folks won't have any problem getting the software that they need or have any reason to switch. Quite the contrary tighter integration with iPad Pro and other devices that use the mobile versions of the same tools will only benefit them, as will the significant increases in performance and battery life. The same is going to be true of developers who primarily write software for Apple platforms.

    The question is whether people who work outside the longtime Mac-dominated industries like media and more recently iOS/iPadOS/macOS development will be able to get the software they need and how long it will take them to do so. Keep in mind: this was already a real problem for Intel-based Macs, which is why homebrew and virtualization were so popular.

    Take developers - whom Apple is specifically targeting and hoping to swipe in larger numbers - while we can presume that the BSD/GNU in macOS will have C/C++ covered and client based Javascript is fine due to it running in the browser, when will Java (especially OpenJDK) and Scala support come? Python? (And the other data science languages like R and Julia?) PHP? Ruby? GoLang? Server-based Javascript like node.js? And what about databases like MySQL, MongoDB and PostgreSQL, or tools like SQL Developer for them? And then there are IDEs: IntelliJ, Eclipse, VSCode, Android Studio (yes lots of Android developers use macOS) etc.

    Keep in mind: Apple is targeting developers because they are the ones most able to rely heavily on open-source tools instead of proprietary ones. For example 35% of developers primarily write code on Linux. While Dell, Lenovo and HP have released (Intel-based) Ubuntu and Fedora ultrabooks targeting that market, Apple is working to convince them that M1 Macs are the best option for Linux developers, and that Windows developers should join the party too. But right now even basic stuff like ant, git and maven aren't available on Apple Silicon yet. Now nearly all of these are available on Linux for generic ARM (think the Qualcomm/MediaTek/Samsung Exynos/Ampere/Calxeda/Raspberry Pi) Linux, which Apple says can be virtualized on macOS to meet developer needs. But using what virtualization tools? (Developers strongly prefer VMWare Fusion and VirtualBox to Parallels, though some of the most technical ones are capable of using KVM/QEMU, which also should will be available as part of macOS.) 

    And again, developers are the best case scenario as they are the ones most likely to be able to handle stuff like open source tools - whether downloading the binaries or building new binaries for the Apple M1 from source - and virtualization. Other users? Not a chance. That is why people outside the creative/media industries are going to have to take a "wait and see" attitude.

  • Reply 10 of 19
    How Apple can call the 13 inch MBP a Pro machine while lacking a discrete GPU is beyond me. 
    Yeah, yeah, we all know you think very highly of yourself.
    roundaboutnowwatto_cobra
  • Reply 11 of 19
    mainyehc said:
    I’ll just throw in my €0,02: what if the Unified Memory Architecture is the key here?

    No, really, hear me out: what do conventional PCs (even PowerPC Macs back in the day) do when they run out of memory? They do paging of stuff from memory into storage, A.K.A. virtual memory.

    The thing is, I’m guessing the on-chip memory of these things is not only more power efficient, but fast. Really fast. Maybe even much faster than conventional memory in DIMM and SO-DIMM modules connected to the CPU via its memory controller, as it resides on the SoC itself and benefits from direct, fast lanes connecting it to the CPU cores themselves.

    What about SSDs? Well, those are also really, really fast, even for such a paltry, apparently non-Pro machine like this. While they’re not exactly melded into the CPU as well, it seems Apple has done some advances in this area as well.

    So… maybe paging to virtual memory isn’t as much of an issue as it was before? And maybe said UMA is better at managing the whole process as well?

    If Apple adds a “Mac mini Pro”, or “Mac Pro mini”, or whatever they call it to the lineup, in space gray to match the later Intel models and the DTK, with an M1X/M2 chip, four TB4/USB4 ports, and, say, 32 GB of this newfangled UMA memory, expect smaller studios to pick this thing up. It’s the trascan-redux-as-lunchbox, minus the melting and inefficient dual GPUs and at A THIRD of the price, and it might very well sell like hotcakes.

    For now, this first generation is more of a proof of concept than anything else (at least as far as some less demanding segments of the professional market are concerned), but keep an open mind and your eyes peeled for benchmarks and real-world testing.
    Someone did a great job of debunking this earlier. UMA isn't magic. 16 GB is 16 GB, flat out. UMA's benefit is that it is faster, allowing you to crunch through one task and get to the next more rapidly. But ultimately it is the same amount of resources. Meanwhile the traditional CPU/GPU setup isn't just 16 GB of RAM. It is actually 20-24 GB of RAM because the GPU will have its own 4-8 GB of memory in pro machines. And the separation keeps the GPU tasks from taking resources from the CPU during operations that need heavy CPU and GPU resources simultaneously, like creating a graphical representation of a data science set. 

    Long story short, UMA and integrated graphics will be great for 32/64/128 GB machines. It will even be an improvement for non-pro users on 16 GB machines. Those Wordpress plugins will never be faster! But pro users should wait until faster memory configurations are available. And considering this great article on the I/O limitations of the M1 chip - and remember I mentioned precisely such things back when I was an Apple Silicon skeptic as much more goes into legit "pro" workstations than mere clock speeds and benchmark results - pro users might need to wait on the M2 chip also. Looks like it will be a 2 year transition with the casual users taken care of first after all.

    https://www.zdnet.com/article/the-new-m1-mac-mini-is-intriguing-but-its-trade-offs-may-hint-at-unspoken-internal-chipset-limitations/
  • Reply 12 of 19
    ciacia Posts: 248member
    Unless there's some super secret sauce in the new MacOS that makes programs no longer need RAM, editing on 16GB is a non-starter for real work.
  • Reply 13 of 19
    How Apple can call the 13 inch MBP a Pro machine while lacking a discrete GPU is beyond me. 
    Yeah, yeah, we all know you think very highly of yourself.
    Except that he isn't the only one talking about this. Take a read and in the process remember that he is a longtime Mac Mini guy.

    https://www.zdnet.com/article/the-new-m1-mac-mini-is-intriguing-but-its-trade-offs-may-hint-at-unspoken-internal-chipset-limitations/

    The real issues that he raisers aren't so much about memory as it is I/O. Smartphone chips don't have to deal with serious I/O, vigorous multithreading - if I remember my Android programming stuff a lot of it is event driven and asynchronous - or true multitasking. It looks like Apple handled the multithreading and multitasking at this stage but they still have a bit of work to do with the I/O. The way to deal with it at this stage is sell devices to consumers that won't need it in the first place. 

    I believe that Apple will have to start being more like Intel, Qualcomm, Samsung, MediaTek, AMD and everyone else by offering chip tiers.
    Ax: mobile family
    M1x: iPad, MacBook Air and entry level MacBook Pro and Mac Mini. Equivalent to Core i3 and Core i5 but a lot faster.
    M2x: iMac, "pro" MacBook Pro and Mac Mini. Equivalent to Core i57 and Core i9 but - again - a faster. 
    M3x: Mac Pro and top end iMac. Equivalent to Xeon and maybe faster. Look, don't get angry. Remember, unlike the Core i series, which top out at 8 cores for the i5 and i7 and 10 cores in the i9, the Xeon in Macs go up to 28 cores and the Xeon line goes up to 56. Unlike the Qualcomm family of CPU makers, who reached diminishing returns after surpassing 8 cores over 5 years ago and have been stuck on 8 cores ever since, Intel actually knows how to get proportional performance from scaling cores. Apple meanwhile is just now getting to 8 cores for the first time. They are not going to beat the Xeon performance with 8 cores. To put it another way, beating Intel's 10 cores with 8 - which the M1 accomplishes and is similar to how the 4 and 6 core Ax chips have beaten 6 and 8 core Qualcomm Snapdragon chips - is totally different from beating Intel's 28 cores with 8. 

    So yeah, if you have real performance needs, wait till the next batch of Apple Silicon Macs roll out. Clock speed isn't everything by a long shot.
  • Reply 14 of 19
    BeatsBeats Posts: 3,073member
    Wait until they see Apple Silicon iMacs and Mac Pros!

    cloudguy said:
    Eh. The media - news and entertainment - have had a strong preference for Apple products from day one. They have stuck with Apple through all their transitions - Motorola 68x, PowerPC, Intel, iPhone, iPad - and really don't consider other platforms unless they are forced to. 

    BECAUSE THEY ARE BETTER. All the rationalizations in your post are ridiculous and show how bad you want to discredit Apple. You wanna believe pros choose Apple for no reason.

    As someone who worked in the professional industry we cannot buy crappy Windows machines that often crash and delete data and are slow as a turtle. We also don't use knockoff iPads for work.

    Imagine working on a movie and you spent 12 hours "in the zone" and your windows machine shuts down and deletes the hard drive? You didn't back up your work that day and lost 12 hours and the "magic" you created. A typical person who does not work in the industry would say "just do it again", except it doesn't work that way. Your mood has changed and you can't get the work done exactly the same.
    Samsonikkwatto_cobra
  • Reply 15 of 19
    I’ve been creating motion graphics and 3D animation for over 25 years, and I’ve seen quite a few paradigm shifts, technology advances, OS rebuilds, and companies come-n-go. I’ve also forever been caught between to camps: ProSumer and Professional. It’s irritated me to no end that people claim this device or that software isn’t “really professional grade,” then by extension I am not a professional. The tech does not define your “professional” label—your product defines “professional.” Are you a paid wedding videographer? Are you a TD at a boutique VFX house? Are you an indi-filmmaker? Do you develop games for The Switch... If it’s your primary career income, by definition you are a professional. I’ve used everything from an Intel 486SX-25 to PowerPC thru Intel Xeon & AMD Ryzen to create my work. I’ve done entire productions (3D, compositing, & edit) on my 2012 i7 MBPr w/16GB RAM. Of course the more RAM the faster the process, but just because a wee’lill Mac Mini has only 16GB RAM and is the size of a few decks of cards doesn’t make it any less “Professional” than the maxed-out Mac Pro. It’ll just be more weighted down by it’s limitations. However, I find in my world, limitations lead to innovation and revelations. I for one am really excited about the M1 Mac Mini for my indi-filmmaking and visual design work — be it Pro or hobby. And the price-point is crazy good.
    Samsonikkwatto_cobra
  • Reply 16 of 19
    How Apple can call the 13 inch MBP a Pro machine while lacking a discrete GPU is beyond me. 
    The same way they always did with the Intel 13" MBP (no change).
    edited November 2020 watto_cobra
  • Reply 17 of 19
    mainyehc said:
    I’ll just throw in my €0,02: what if the Unified Memory Architecture is the key here?

    No, really, hear me out: what do conventional PCs (even PowerPC Macs back in the day) do when they run out of memory? They do paging of stuff from memory into storage, A.K.A. virtual memory.

    The thing is, I’m guessing the on-chip memory of these things is not only more power efficient, but fast. Really fast. Maybe even much faster than conventional memory in DIMM and SO-DIMM modules connected to the CPU via its memory controller, as it resides on the SoC itself and benefits from direct, fast lanes connecting it to the CPU cores themselves.

    What about SSDs? Well, those are also really, really fast, even for such a paltry, apparently non-Pro machine like this. While they’re not exactly melded into the CPU as well, it seems Apple has done some advances in this area as well.

    So… maybe paging to virtual memory isn’t as much of an issue as it was before? And maybe said UMA is better at managing the whole process as well?

    If Apple adds a “Mac mini Pro”, or “Mac Pro mini”, or whatever they call it to the lineup, in space gray to match the later Intel models and the DTK, with an M1X/M2 chip, four TB4/USB4 ports, and, say, 32 GB of this newfangled UMA memory, expect smaller studios to pick this thing up. It’s the trascan-redux-as-lunchbox, minus the melting and inefficient dual GPUs and at A THIRD of the price, and it might very well sell like hotcakes.

    For now, this first generation is more of a proof of concept than anything else (at least as far as some less demanding segments of the professional market are concerned), but keep an open mind and your eyes peeled for benchmarks and real-world testing.

    You definitely have a point.

    The thing is that SSDs are in the same league as RAM lately. DDR4 transfer rate is about 15-20 GB/s, while SSDs are rapidly approaching (or even surpassing) the 4GB/s speed range. So in the worst case we are talking just five times slower (my old spinning hard disk had 20MB/s on a good day, about 1000 times slower than RAM).

    This should DEFINITELY make the paging in-out of RAM (into and out of virtual memory, id est, the SSD) worlds faster.

    Still, not the same, and there are many other considerations (page size, latency, you name it) but we are getting in the ballpark. This is something that we will discover when the actual Macs arrive. It could be a pleasant surprise.


    watto_cobra
  • Reply 18 of 19
    How Apple can call the 13 inch MBP a Pro machine while lacking a discrete GPU is beyond me. 
    Maybe that particular machine should just be called MacBook not MacBook Pro.  But get over the naming and get over the "specs".

    Why can't a high performance GPU be integrated? The RAM is "integrated" now. That means it somehow performs worse? "Discrete" or "integrated" is irrelevant. It's $1300 and eats its competition for lunch.

    It's beyond you? That says more about you than Apple.
  • Reply 19 of 19
    cloudguy said:
    How Apple can call the 13 inch MBP a Pro machine while lacking a discrete GPU is beyond me. 
    Yeah, yeah, we all know you think very highly of yourself.
    Except that he isn't the only one talking about this. Take a read and in the process remember that he is a longtime Mac Mini guy.

    https://www.zdnet.com/article/the-new-m1-mac-mini-is-intriguing-but-its-trade-offs-may-hint-at-unspoken-internal-chipset-limitations/

    The real issues that he raisers aren't so much about memory as it is I/O. Smartphone chips don't have to deal with serious I/O, vigorous multithreading - if I remember my Android programming stuff a lot of it is event driven and asynchronous - or true multitasking. It looks like Apple handled the multithreading and multitasking at this stage but they still have a bit of work to do with the I/O. The way to deal with it at this stage is sell devices to consumers that won't need it in the first place. 

    I believe that Apple will have to start being more like Intel, Qualcomm, Samsung, MediaTek, AMD and everyone else by offering chip tiers.
    Ax: mobile family
    M1x: iPad, MacBook Air and entry level MacBook Pro and Mac Mini. Equivalent to Core i3 and Core i5 but a lot faster.
    M2x: iMac, "pro" MacBook Pro and Mac Mini. Equivalent to Core i57 and Core i9 but - again - a faster. 
    M3x: Mac Pro and top end iMac. Equivalent to Xeon and maybe faster. Look, don't get angry. Remember, unlike the Core i series, which top out at 8 cores for the i5 and i7 and 10 cores in the i9, the Xeon in Macs go up to 28 cores and the Xeon line goes up to 56. Unlike the Qualcomm family of CPU makers, who reached diminishing returns after surpassing 8 cores over 5 years ago and have been stuck on 8 cores ever since, Intel actually knows how to get proportional performance from scaling cores. Apple meanwhile is just now getting to 8 cores for the first time. They are not going to beat the Xeon performance with 8 cores. To put it another way, beating Intel's 10 cores with 8 - which the M1 accomplishes and is similar to how the 4 and 6 core Ax chips have beaten 6 and 8 core Qualcomm Snapdragon chips - is totally different from beating Intel's 28 cores with 8. 

    So yeah, if you have real performance needs, wait till the next batch of Apple Silicon Macs roll out. Clock speed isn't everything by a long shot.
    I'd wager more something like M1 for the entry level, M1X for the mid level, M1Z for the high end level, all first generation.  Then M2, M2X, M2Z second generation a year later, M3… after that, etc. Just as with the A chips for the last 10 years.  Though my X and Z suggestions are arbitrary.  The point is the number will be the generation, with some additional designation for variants.

    And I doubt anyone seriously thinks Apple will try to beat Intel's 28 cores with 8 Apple ones.  There's no reason these M chips should ever be limited to 8.  More cores - many more - must be coming, and from what we've seen so far, I'd wager Apple knows how to make them work better together much more than Intel or anyone else does.

    I'll be very surprised if Apple's Pro chips in their labs aren't already eating Xeons for lunch.
Sign In or Register to comment.