Mac Studio GPU performance

Apr 12, 2021
Asus z590 ROG Maximus XIII Hero
RX 6600 XT
  1. MacBook Pro
  2. Mac mini
  3. Mac Pro
Classic Mac
  1. Centris
  2. Power Mac
Mobile Phone
  1. iOS
On these tonymac forums, it's common for posters to share benchmarks and use them to suggest that kit is/is-not suitable for some purpose.

OTOH it's basically unheard of for posters to explain how particular kit actually gets used to get work done and how that performance matters to job results.

It is a mantra in the PC industry that its magic is empowering consumers to build new worlds, reach previously unattainable new goals. It's like old-timey religion.

But the fact is that good use of existing tools usually counts for more than improved tools, where there's a sway as expert craftspeople gain access to new powers that change how we see the world.

At a level of pure compute, video is at least 10x harder than audio. But even tho PCs can do video with ease today, there's an eternal audio nerd who for some reason can't get a PC that others are using well for vid to meet the performance demands of his music project. This is about how the tool is used.

I'm not making a point about audio users doing it wrong—although that might be a good point—I'm making a point about learning to use tools with an experts comprehension of their power and therefore their limits.

When anyone talks about how some Apple devices are always behind some PC kit in raw power, like GPU cores, there's a part of the story that's usually missing: the question of how the device is intended to be used.

What I mean by "intended" is that a tool maker by definition has a strong idea about what the tool is good for; he's a domain expert of its uses. If he doesn't have a good idea, that tool is by definition not useful and gets abandoned.

Apple designs its devices with its key tools in mind: say FCP or Logic, and ensures the device is configured to be usable. If you think these tools suck, the easiest out is to blame the HW for also sucking. But Apple has a track record of decently matching it's devices and apps and that's part of how they changed the content industry!

Today, there's not much meaning in comparing Apple and PC partly because what we expect PCs to be was so heavily influenced by Apple, and because Apple has always been a good student of modern computing.

Apple today does something at the consumer level no other personal computer company does: they fully vertically integrate their offering to specific markets with a deep understanding of what work means to those segments.

Hackintoshers instinctively avoid cognition of this basic fact: that while an Apple device (e.g., GPU core count) may not be up to some PC reference point, it's likely to be irrelevant to how Apple understands the device and intends it to be used.

If Apple is really bad and doesn't understand, it's products will fail.

Apple products are not failing. They have become standards by which everyone expects guidance in new value propositions. Even people who typically hate Apple can see this is true.

The hackintosher is now a quite peculiar breed of enthusiast. I mean this in the best sense, I don't want to put anyone down and I don't hate myself — at least not for hackintoshing. He lives on the edge of Apples influence on PC design.

Regrettably it's become a knifes edge.

I'm not really stoked to see people set aside their hacks for Apple kit, but we are stuck: they did it, they changed the world again, even if its ultimate meaning is just the narrow consolidation of Apple to their own interests.

There's another world out there, where the spirit of the hacker lives on, a brave old world...

Called Unix