Contribute
Register

Hackintosh 2022 and beyond - Is there any point now?

Status
Not open for further replies.
Resolve Studio Also
SuperScale 3X on 4K media in 2,5 faster with Neural Engine (MacStudio) than CPU (iMac 2019 top of the line)
apple_neural_engine_davinci.webp
To be clear media engine and neural engines are different units, yes?

One is a codec typical for every GPU to offload AV transcoding and the other is the NN ML optimized for Apple's AI apps. AMD and Nvidia have their own forms of these ML units.

Presumably Apple tailored both to boost according to their architecture / product biases, so expect to see very effective, or they wouldn't have bothered.

Everybody else thinks this same way, but for example Intel/AMD... can only release developers kits and app notes, not literally orchestrate features from design through deployment.

Microsoft can't add and brag about Windows features per same because they're so backwards compatible, only a small slice of their pie has compatible HW for ML optimatizations.

When AR / VR goes big, expect Apple to stand wayyy out ahead with clear products and upgrade paths while Wintel will be wandering around in the field trying to brag that electric typewriters plus Ribbon/Clippy 11 are still very relevant. The PC industry is a such a herd of cats, the only way it can join is by the force of another Apple. But why would today's Wozniak and Jobs want to base on Wintel?! The tide already turned but nobody noticed. Meta has been firing a machine gun at its feet, too bad for Oculus. That leaves Sony.
 
To be clear media engine and neural engines are different units, yes?
Yes I think so

Fun thing, is that Hardware from Apple (Apple Neural engine), works for heavy functions in Resolve Studio (Davinci neural Engine)
Everything is "Neural Engine" :lol:
 
Last edited:
I thought after the introduction of the Mac Studio there would be no need for another hack. But Apple just doesn't respect the needs of the user to grow. As a producer I need PCIe cards for audio accelerators, at least 4TB of NVMe storage for virtual instruments, more USB ports for controllers and the ability to add new innovations at my leisure. Thunderbolt chassis cost $400 and up, cables $50 and up, adapters $35 and up, etc. I dont want to have to spend $6,000 for a workstation to accommodate all that.

My latest machine (built by impulse) provides everything I’d want for $1,000. l’m not a fan of upgrading OSs, some software is optimized to be with certain hardware, so I’m now happy with Big Sur and still using High Sierra and El Capitan on heavy used machines. Now that I know Raptor Lake is a viable option, I don’t know if I’ll ever buy aMac again.
 
My latest machine (built by impulse) provides everything I’d want for $1,000. l’m not a fan of upgrading OSs, some software is optimized to be with certain hardware, so I’m now happy with Big Sur and still using High Sierra and El Capitan on heavy used machines. Now that I know Raptor Lake is a viable option, I don’t know if I’ll ever buy aMac again.
I just posted this into GPT4
Buy an apple Mac with M1 chip or build a Hackintosh for video production?
Below is the answer.
As an AI language model, I cannot provide a personal preference or opinion. However, I can provide you with some information that may help you make a decision.

Buying an Apple Mac with M1 chip would be a better choice if you're looking for a computer that can handle video production with ease. With the M1 chip, the latest Macs can handle heavy-duty tasks such as video rendering, editing and processing with significantly better performance than the previous Mac models. Moreover, the M1 chip also provides better energy efficiency, making it more cost-effective in the long run.

Building a Hackintosh, on the other hand, may be comparatively less expensive, but it also comes with several drawbacks. Building a Hackintosh would require a good amount of technical knowledge and skills, and it may also be challenging to find compatible hardware components. Additionally, updates and compatibility issues may arise in the future, which may require additional technical expertise to fix.

Taking all these factors into account, if you prioritize reliability and ease of use, buying an Apple Mac with M1 chip is a better option for video production. However, if you have technical expertise and want to explore customization options, building a Hackintosh could be a fun and exciting project.
 
I predict that the Mac Pro as it was previously known, a HEDT form-factor, will never come back.

There are no major market workflows for which it is a suitable design.

AppleSi cannot be on a GPU upgrade treadmill, which is the major thrust of the Intel market both for gaming and workstations.

All models offer options with so much RAM that only only very narrow specialized workloads required further options, and these represent an exponentially decreasing market segment. Today's customers can just buy a fully loaded option. If they need to upgrade, they an buy another one! To justify this analysis, I'll point out the enormous price/performance gains over the last decade. The kit has advanced by orders of magnitude. Yet tyros continue to lament that they didn't buy a higher end option, etc., without any rigorous analysis of what limits their work beyond the seats of their pants. IOW, the user's critique of the upgrade potentials is the same today as it was in 1993, even though the kit is thousands of times more powerful. We can extrapolate that since the PC market has exceeded customer expectations by orders of magnitude, Apple no longer cares about Mac user-groups upgrade gripes when making product cuts. You could see the beginning of a sea change with the outcry over FCPX, in which users for the first time became to complain that Apple was improving their stuff too much.

Thunderbolt/USB4 effectively extends PCIe storage via a handy form-factor. It's limits don't stand in the way of most local workloads.

The idea of a "server" in personal computing has been replaced by various incarnations of very-low-cost appliances and the "cloud". Server, at a personal level, is now a euphemism for a economical headless box that serves a narrow purpose. At an industrial level, server is defined by the rack, be it remote or local. Whether you choose to administer your "server" via Mac, Linux, or Windows, is purely a matter of taste. None does anything the others can't with respect to server administration. If you look at where Intel has taken Xeon, it has nothing to do with personal appliances. It's about cloud provisions.

The last vestige of "workstation" form-factors was in prestige of scientific computing. The 2019 Mac Pro was a glamorous throwback to a corporate workstation market-segment has retired as overall compute and network power has risen. The part of the system that sits next to the user has become so enormously powerful and energy efficient it fits in a mobile phone. As Steve Jobs famously proclaimed in 2007 introduction of the iPhone:
• It's a great mobile phone...
• It's the best iPod ever...
• And... It's an internet communications device!

Well, 15 years later, that device is a hundred times more powerful on every performance axis.

The power of HW is not a limit for personal computing any more.

So what is anyone waiting for in a reboot of what was formerly known as the Mac Pro? Wheels?! It's an antique form-factor.

As to macOS Intel support staying around longer, I see no reason to expect this. Apple retires HW on a steady schedule. Why should this change? There's no more reason to expect longer-term Intel support than there was PPC support. All of this has been seen before. Sure, it could happen, but I think the 10 year progression from Trashcan to Mac Studio provides great clarity of Apple's direction, without them needed to make any speeches. The product says it all.
 
2 hr battery life for the ski goggles!?.
 
a macbook is very very very expensive.I would not be able to afford one for at least 5 years of saving,Hackintosh is the only soluton.As a noob with hacks to produce a mac on pc i only need usb passthrough.the only reason why i am building on is for iphone ,ipad related work.A vm does not give me usb passthrough 100 percent.SO i did the cheap way...lol
 
I'm reading the point about not being able to afford a new Macbook, and an very sympathetic to problems of cost.

But you must agree there's a major distinction between the high cost of a good in market terms, and the personal situation un-affordability to buy a thing.

By every measure Mac laptops and personal computers are insanely less expensive today than any previous time, and that's not counting performance, which has advanced by a factor of a million or so since the first Apple laptops.

Apple has numerous 1000ish offerings if you want to be on the bleeding edge, which in terns of literal dollar counts is less than half the cost of the 1984 Mac. Considering inflation the new kit is about 1/5 of the unit cost.

Measured as performance-per-dollar over history, today's PCs are free. They just happen to have a cost step to entry, but this has become arbitrarily low, even for new gear.

Case in point, a month ago I bought a new Beelink mini PC (NUC) with 4 core Celeron compute equal to 2008 Mac Pro, 16G, 512G SSD, 4K x3, USB3 gen2, and Windows 11 Pro pre-installed for $180 US retail, at a sale price. So the idea that there's any cost factor prohibiting access to a capable new PC is absurd.

As to fine points about Thunderbolt, GPU, neural chips, ProRes, etc, well as Rosanne Rosannadana's father used to say in the '70s: "It's always something!" But we can generalize that at some degree of featurefulness means we're longer considering a commodity, but a specialized object. Historically, such artifacts are always subjectively priced, e.g., the price of a unit cryptographic key on the Bitcoin blockchain is currently hovering around $30K. Is that affordable?

I am not trying to be hyperbolic.
As to needing a Mac, today that is nothing but a pure personal preference. A preference I respect, but it's totally subjective in any rhetoric about value.
As to catalog of cheap kit that can be hacintoshed, I'm afraid I could find no path from Beelink to macOS, but I'm not an expert. I chose to dual boot Windows and Ubuntu, both of which work exactly as you would expect for the spec. And it's only going to get harder.

But good news: MSFT is effectively giving away Windows 11, and Linux communities welcome tech-minded enthusiasts!

If your time is worth anything, you can run both those stacks on almost any HW in a tiny fraction of the hours required for hackintosh...

And... if you chose Linux, the effort you expend learning arcane details has value in the future, whereas for hackintosh it's appears to be At World's End— To quote Capt Barbosa: "For sure, you have to be lost to find a place that can't be found, elseways everyone would know where it was!"
The future of hackintosh is at best Thar be dragons!

To repeat: a measure of hack value in terms of "not being able to afford a Macbook before 5 years of saving" is not a comment about the value of Apple Macs.

Comments about why you can't live without a Mac would be illuminating to forum: In my case, the preference to Mac is that I like the layout and the Macs I've purchased have turned out be well-made and decently long-lived. Apple has made a lot of junk, but I somehow avoided it.

My foray into hackintosh was due to free PC kit right before AppleSi. It took on a life of its own and I ended up with a stable powerful hackintosh. But I am left only with reservations. M1 kit has reset my expectations.
 
Status
Not open for further replies.
Back
Top