I started googling and I'm going to just say what I find may not be comprehensive. The first video is from someone I'm familiar with, he's largely a PC person who loves building ITX computers and clearly a 6,000 dollar desktop build should beat out a laptop, so take this one with a grain of salt.
Even the kernel is mostly architecture independent, xnu has just 11% of C/ASM code in the four architecture-specific sub-directories arm, arm64, i386 and x86_64. I'm not saying it's the best idea, I'm not saying it's the only idea, I'm not saying they must keep Intel support. What I'm doubting is that we on the outside can be sure that one solution is clearly better than the other. As you said yourself, there are always tradeoffs. I guess we are all a bit antsy to see what Apple will unveil, hopefully not too far in the future.
Agree. Even the Watch used the same E-cores as the M1. Sure, it's "just" the E-cores, but it's also just a Watch!
Where did you get iPhone 11? In Geekbench a single 3.7 GHz Zen 4 core about matches a single 3.2 GHz Firestorm. And while Geekbench is not my favourite benchmark at least it's cross-platform.
I'm not sure about this. I think it might be the other way around actually.
No, but they already exist. What is more work, maintaining that code or developing a workstation CPU? Rhetorical question, we don't know. Sure, there's more to it, but there's also more to an ARM Mac Pro than just the CPU.
More often than not things are copied once to the GPU and then used many times. Think textures in a game. Of course that's not always the case, so the long answer is, as so often, "it depends".
My mistake, the 4.5 GHz in the GB listings are probably base clock, not actual frequency during ST tests.
This. Plus full in-house control of all components, leverage when ordering chips, cross benefits with their whole product-line, logistics etc. No way Apple will go back to relying on development of key components by external parties. Apple must be quite frustrated with the slow development of their own modem chips…First, yes, Apple should believe that an M2/M3 Extreme would be better than any CPU AMD can make for the Mac Pro.
Using a CPU with many slow cores designed for data center use will not beat any Apple Silicon that is 2x Ultra for macOS users. It'll look good in Cinebench though.
Second, Apple does not want to support AMD and Nvidia products in addition to Apple Silicon. It's not as simple as slapping together PC parts from Newegg and call it a day. You need to write drivers, support them in first-party software such as Finalcut, and provide years of support. All of these things cost money, time, and resources. It also makes the codebase for macOS much more complicated in the long run.
Third, AMD + Nvidia does not satisfy the unified memory model. This is the model Apple is going for. It has huge advantages in performance and applications that support it.
Lastly, AMD CPUs and Nvidia GPUs simply don't support a lot of features that Apple Silicon can. IE. Using the neural engine for extracting objects from photos. Using ProRes encoders. Using the secure enclave. Using image processing acceleration. Even if these parts can be replicated with AMD + Nvidia, Apple would not want to. Not for the Mac Pro which is a niche.
Apple employees are probably itching to get rid of support for x86 CPUs and AMD GPUs in macOS. Speaking as a software engineer, there is probably a ton of code in macOS and first-party software that is something like "If Intel CPU, do this. If Apple Silicon, do this.". I would want to delete this code asap.
I have been wanting to see numbers on this for a long time. Like how much software really needs more than a recompile? We all know that "optimised for" is often just a marketing slogan.
In my experience compiling the same software on multiple platforms helps expose and fix bugs which improves stability.
I don't consider those benchmarks to be fair because Anandtech used WSL instead of Linux. The performance penalty for using WSL is unclear.
A benchmark that only tests integer performance. But does it check Apple relevant things like audio/video accelerators? ML cores? Etc...? There is more to Mx than pure integer pushing power.....a lot more!
Apple could still remove Rosetta 2 in a couple of years. That's a separate decision.
It is not could, it is will! They will give people a few years to transition and then it is game over for x86!
