Why would you want to do that if you can just run Windows or Linux?
There's nothing anymore that a Mac can do that a PC can't.
You COULD build a hackintosh, but performance will be subpar.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Hard Truths About The Apple Silicon Replacement to the Mac Pro (2019)
- Thread starter Yebubbleman
- Start date
- Sort by reaction score
Why would you want to do that if you can just run Windows or Linux?
There's nothing anymore that a Mac can do that a PC can't.
You COULD build a hackintosh, but performance will be subpar.
Whole point of a Mac is that it's a turn key solution.
It would be easier to buy a Window 11 PC or even a Linux box than to do a Hackintosh.
Many Hackintosh owners retired their Intel PCs because Apple Silicon Macs are superior value.
Hard rumors? TOO EARLY TO KNOT UP THEM KNICKERS. the rumors may prove to be accurate, but until then, they are just that
Performance i7-12700K ona macOS is exactly this same as on Windows 11. And so You lose nothing in performance when run macOS. This same is in graphic performance. Many people prefer to work on macos but not necessarily on M1/M2 chips. It is not important what is inside the computer Intel or M1/M2 chips. Performance and price matter .
It actually is important.
Maybe not to the user who wants to just browse the Internet and do some light tasks, but what's inside changes the architecture. And this has implications on software compatibility.
Now, sometimes it is possible to virtualize software. You may consider the performance penalty is not significant in your use case. But it IS there.
Thank you, MacPro. It has been a trend on the Internet that just because some people like a particular celebrity, or product or brand, that they feel they should defend the product / brand / celebrity to death, even when they show a weakness or have a weak point. But I feel that approach prevents people from improving.
I like Apple too, and I have some of their products. But some people don't understand that Apple is not perfect, and there are points they can and SHOULD improve. Because change is only carried by imperfection. If something is perfect, why change?
I personally would like to see Apple more on the bold side again. Different product shapes, formats and colors. For example, on the iPhone line, it bothers me to no end that they leave the more interesting colors only to the intermediate phones, and only silver / muted colors to the pro line – as if a pro product automatically meant "dull".
If we're all open, we can definitely reach to different solutions and ideas together!
And what if we all wait to the day when Apple announces the Mac Pro? Would that be a possibility? And wouldn’t it be nice when Gurmann and all the other idiot leakers shut the **** up?
Am I the only one who likes to be surprised at Apple keynotes?
Am I the only one who likes to be surprised at Apple keynotes?
Apple wants to spread sales as evenly as possible over a devices' cycle, and if that means delaying the non dull colors then thats what they are gonna do...
Hi! I have Resolve Studio on my 14" MacBook Pro and a super-fast (well, for 2019) PC workstation with 18-core i9 and 3 Nvidia 2080TI. Perhaps I can provide some color on the benefits of both platforms. The only reason I have the PC is because of a certain 16mm motion picture film scanner software only working on Windows... this scanner captures uncompressed .BMP (!) image sequences, where each .BMP is around 1.5MB. To handle these files in any useful manner, in either Resolve or Final Cut Pro, I have to transcode the image sequences to video files, using Adobe Media Encoder (subscription$$$, bad) to a common intraframe video codec appropriate for editing, such as Apple ProRes 422 HQ or Avid DNxHR HQX. Once the image sequences are converted to video files, I could work with them on Mac or PC. So, why Windows? Well, the transcoding process of .BMP image sequences to video files is really, incredibly CPU intensive. And, the digital noise reduction process, using NeatVideo, and the frame interpolation process, to smooth out 16fps 8mm and Super8mm films to smoother 24fps, is GPU intensive! In spring 2019 when I bought this system, I ran some tests on a friend's iMac with the fastest GPU Apple offered at the time, and I was only getting 4-5fps of NeatVideo noise reduction. On the PC system, with 3 Nvidia 2080TI, I was getting 27fps, which was just barely enough to see the results in real-time, played back at 24fps. So, the PC system was a no brainer. Initially, it was a 10-core system. But, I swapped out the 10-core CPU for the 18-core CPU when I realized the .BMP image sequence to video file transcoding was too slow for my taste and I wanted to speed it up. I had a really neat workflow that enabled me to, in combination with a 24TB software SSD RAID, to scan a film in the scanner software, which really only used a single core of CPU, while simultaneously transcoding the previously scanned film from .BMP image sequences to video files, in the background, with Adobe Media Encoder, which used a dozen or more CPU cores, while simultaneously rendering NeatVideo noise reduction in DaVinci Resolve Studio, which used my 3 Nvidia 2080TI, and all of this was possible simultaneously due to the incredible speed of the PC.
Nowadays, however, now that Mac's have caught up (and, I really mean that in the nicest way possible... Apple targets a different market than the super-max-performance-at-any-power-level PC market), what I would do now, in 2023, is this: have a cheap, inexpensive Windows PC just for the scanner. Then, copy the scanned image sequences to a portable SSD, then use a 14" M2 Max MacBook Pro maxed out on 12C and 38C, with lots of unified memory (NeatVideo noise reduction actually benefits from having lots of memory), for rendering the image sequences to video files. And, I'd also be using Resolve and NeatVideo on the MacBook. Times have changed. Having the crazy fast PC is no longer necessary.
Sorry that was longwinded. But, it ultimately leads up to this: the primary issue with Resolve on Windows was that Resolve on Windows DOES NOT OFFER export to any flavor of Apple ProRes. It can, however, export to an equivalent intraframe codec called Avid DNxHR. But, in reality, I'd prefer to keep everything in one codec family throughout the entire post-process. Now, with the new MacBook's, or a Mac Studio for that matter, I can achieve this goal of staying within the same video codec, on a machine that is plenty fast.
TL;DR:
Resolve on Windows was useful to me 4 years ago, in SPRING 2019, when Apple didn't have sufficiently fast GPUs in their computers, for rendering noise reduction on scanned 16mm film footage, nor did Apple have sufficiently fast CPUs in their computers for rendering uncompressed .BMP image sequences to ProRes video files. But, times have changed. Mac's have those things now.
So, I'm not quite sure who Resolve on Windows would be for nowadays.
P.S. - example of 16mm film before and after digital restoration : https://vimeo.com/342584642/f818c99ec9
Yes and no. I was rendering multiple passes of NeatVideo on the PC, and that would involve bringing in the DNx file, applying NeatVideo, rendering out to DNx, then re-importing, rendering again to DNx, several times. Ultimately, I would end up with a DNx file that would still have to be converted to ProRes, because I wanted to archive the "masters" as ProRes files. So, now, on Mac, I can avoid that last final step. But, yes, for the most part, as you said, it's merely a preference. It's worth mentioning, however, that Adobe Premiere, on Windows, offers ProRes as an export codec. But, Resolve on Windows doesn't.
Many people wish Resolve on Windows would offer ProRes as an export codec.
That's one of the key differences, as far as I am concerned.
Thanks so much for sharing your workflow. I teach video production at a university and knowing how other professionals work is super important.
Agreed. There was a twitter post a year or so ago from someone working on the memory team at Apple stating they were working on the "slower secondary RAM" option. We will see when it gets released. But there are no facts here until the system is released. If what is stated is true, I still believe there is NO NEED for a Mac Pro if it will just be a Mac Studio.There are no facts here just speculation. Until Apple releases or announces the new Mac Pro there are no hard truths.
If you want to get wrapped up in rumours, go ahead, I heard they will offer a pink colour version for gamers with LED light strip options.
Agreed. It is quite common for me to work on 8-13 and sometimes even 16 hour videos. Yes....HOURS. My modern Macs, even my M1 Mac mini is SO MUCH FASTER at editing and exporting these large videos compared to my Windows PC with a 10th gen i9 and a 3080 Ti. Final Cut Pro is faster than Premiere Pro on Windows. Davinci Resolve is faster on macOS than on my Windows system.
THIS is what specialized parts of an SOC does. This is why comparing generic CPUs like Intel to M1/M2 etc is not very good. You are not using the Neural Engine, or the Media Encoders. Just testing cores for cores and frequency for frequency.
At my level, these processes were necessary because I could only afford a cheap $6,000 scanner that captured image sequences. Real industry professionals who scan films for a living would actually use an $85,000 scanner called a LaserGraphics ScanStation, which has a couple of nifty features that speed up the process. For example: the ScanStation has an infrared scanning layer that detects dirt and scratches and uses that infrared layer as a matte to cancel them out, and it does this in real-time before making a ProRes file itself, in the scanner, along with other improvements, such as scanning speed… many times faster than what I had. But, glad to hear my synopsis was useful to you in some way.
Agree 100%. I'm curious if you prefer FCP over Resolve and why? btw, I have not edited anything for money in 15 years. I moved into a different segment of the industry. But, I am definitely curious if and why you like FCP, as someone who is using it in the most complex way imaginable... long sequences. I don't know anyone else who's doing that to ask them about it.
For more complex edits, I have a great workflow going in Final Cut Pro. So I still prefer that. Could be I just need to migrate that over to Davinci and do a proper test at that point. But for the simpler edits, I prefer Davinci. I do like how Davinci segments the work with the tabs though. I wish FCP would do something similar so you can get a narrow view of your edits based on what you want to accomplish.
Edit: Plus I use things like the new Voice Isolation feature in FCP which is just a one click fix to take out some breaths and background stuff. I am sure Davinci might have something similar, but not sure.
Haha. You just proved to me that I know so little about editing that I cannot even understand or appreciate the nuance of these differences. But, FCP works better for you, helps you be more productive, and that's great!
I just started using Davinci so I could very well switch over at some point. However, it has immediately replaced Premiere Pro for me on Windows (since there is no FCP on Windows).
In general, I think you'll find more little things that could be perceived as bugs in Resolve on Mac, compared to FCP X, because Blackmagic has made some different editorial decisions than Apple when it comes to using processes on multiple threads.. things like an unopened sequence thumbnail not refreshing until you click it, things of that nature. But, overall, I like it. It feels closer to FCP 7 to me, in terms of the interface. I never used FCP X long enough to get used to it. I never really even gave it a try. So, don't put any weight into my opinion; it's worthless, cause I never really used FCP X much. It's great to have so many options. There isn't a right or wrong choice. There's only money and getting paid, IMO. I don't edit for money. I'm merely a prosumer who dabbles. Thanks for sharing your experiences.
My kids end up all over the place. Some work on high level animations with place like Sony Imageworks (one of them’s done the new Spider-Man but won’t say what scenes till it’s out) while others work in the non-profit sector and make instagram reels. At the end of the day, I’m happy as long as they’re pleased with their careers, so while it’s important to know what Hollywood is doing, it’s just as (or maybe more) important to know what others do too.
Hence, unlikely
Scientific computing and AI workloads would be somewhat easy to scale to this setup, but it's so uncommon to see mac used for those, that ultimately I don't think we would see this happen.I have a nice career in the film industry, too! We all work for the same five companies, no matter what your skill is in this industry. lol. My dabbling with 16mm film scanning, as a summer hobby, during summer hiatus (while not working on a 40-week a year TV show as a QtakeHD video assist operator...) was to scan old 16mm films I found at the State Library in my home state of Indiana. They had a HUGE collection of old documentaries filmed in my home state and I wanted to see what was on them. So, I offered to scan them in. For free. Anyway. Nice chatting! I would LOVE to have legit professional equipment for film scanning, but the upfront cost of a ScanStation is too much for me, for a summer hobby. That's awesome your students end up doing what they love. So few people can say that. Cheers! : )
Apple deprecated OpenGL and OpenCL in the early prelude to Apple Silicon back at WWDC 2018
"...
Apple has confirmed that it is deprecating OpenGL (Open Graphics Library) and OpenCL (Open Computing Language) in favor of Metal.
..."
Game Devs Express New Fears Over Future of Mac Gaming as Apple Deprecates OpenGL and OpenCL in macOS Mojave
Yesterday at WWDC 2018, Apple revealed macOS Mojave, which is set to bring users a Dark Mode, redesigned Mac App Store, organizable Stacks,...
www.macrumors.com
OpenCL is 'stuck' because Apple gave up. I understand that Apple was 'fed up' but that was a bit premature. OpenCL 3.0 came in 2020.
" ..
OpenCL 3.0 makes the OpenCL ecosystem significantly more flexible by enabling hardware vendors to focus their resources on functionality that their customers need. This is achieved by slicing all functionality beyond OpenCL 1.2 into optional features that can be queried in the API, with macros to indicate whether optional OpenCL C language features are present. .."
OpenCL 3.0 Specification Finalized and Initial Khronos Open Source OpenCL SDK Released
Today, the Khronos OpenCL Working Group is happy to announce the release of the finalized OpenCL 3.0 specifications, including a new unified OpenCL C 3.0 language specification, together with an early initial release of a Khronos OpenCL SDK to enable developers to quickly get up to speed using...
v3.0 would have allowed Apple to 'cherry pick' parts past 1.2 that they wanted to do and not do. The could have kept that alongside Metal but chose not to. Similar factors they choose now to "not allow distractions from Metal" and simplify the stack.
Apple ignored Vulkan also, so it is relatively consistent behavior. Not sure why folks are pushing the "maybe Apple isn't going down the proprietary rabbit hole here" notion. Minimally, Apple has taken on an anit-Kronos standards stance. Apple has skipped Sycl also.
" ..
SYCL typically requires fewer lines of code to implement kernels and also fewer calls to essential API functions and methods. It enables creation of OpenCL programs by embedding the device source code in line with the host source code.
... Using SYCL, it is possible to develop a clean, modern C++ based application without most of the setup associated with OpenCL C code
.."
Migrating from OpenCL Code to SYCL*
Programming oneAPI projects to maximize hardware abilities.
The push from others to weave OpenCL with C++ is another factor to why Apple is walking away. There is a Swift (and previously ObjectiveC ) agenda pushes regardless of where the rest o the HPC market is going.
Worse case Apple has taken the "we're big enough people have to eat whatever dogfood we provide" stance. A Metal (and no alteratives) stances isn't going to help in attracting 3rd party GPU vendors to come to the platform at all. Combined with Apple hyper focused on throwing dGPUs out of as many Mac systems as possible as the default GPU. It is just a very small market for which even less code is portable too.
When the Intel branch of macOS gets turned off OpenGL and OpenCL would likely go with it. ( So several more years to go). They could easily go before then because Apple put those two API on deprecation watch even before nuking the 32-bit apps. Those APIs are far , far more entangled in Apple that 32-bit fragments were. So the 'reprieve' is lasting much longer, but Apple has already put them on 'death row'.
This all likely contributes greatly to why Apple is leaning very hard into herding as many developers as they can into adopting Metal. Until a ton of OpenGL/OpenCL/open, portable interfaces infrastructure has been completely unwound they can't lower the 'kill switch' pain of turning them completely off.