Please don't (implicitly) misattribute quotes. Not that it was terribly important in this case.
I dunno, for instance when CP 2077 became Mac native there were certainly resolutions and quality settings where the native and Xover versions had practically identical performance, but for most of the actually playable settings, there was quite a difference.
www.notebookcheck.net
Not sure what you're trying to say. Total bandwidth is known. What's available to each core, and to each core cluster, is not yet known for the M5 Pro/Max (as far as I know, anyway); it's extra interesting in this generation because they are presumably all across the fusion bridge from the memory controllers.
But that also has to go through gpu translation of the shaders. WoA is directx and hlsl all the way.I dunno, for instance when CP 2077 became Mac native there were certainly resolutions and quality settings where the native and Xover versions had practically identical performance, but for most of the actually playable settings, there was quite a difference.
Cyberpunk & AC Shadows on Apple's M4 Max - Gaming performance comparable to the RTX 5060 Laptop
Gaming on macOS is gaining momentum with more releases of AAA-titles. Cyberpunk 2077 and Assassin's Creed Shadows are two of these titles and they feature an integrated benchmark. This gives us the chance to compare the gaming performance as well as efficiency of the M4 Max with Windows devices.
Some games may show less of a difference, others may show more, but there is a reason people generally prefer a (good) native port to Xover when available (the good being important) beyond convenience - Xover/Wine is great, don't get me wrong, but a full native port is going to be much more performant.
And even when the translation layer was just Rosetta 2, you could see performance differences, though admittedly smaller. Early translation layers for WoA were likewise blamed for poor gaming performance as well, though I believe Prism is thought to be much better. Yes, the GPU matters much more, but if the CPU is being hamstrung ... that can still matter.
Yes, I had an entire second paragraph that addressed that even just WoA translation can cause performance issues. To expand on what I wrote, when it was just WoA pre-Prism translation was poor enough that Macs running games through Xover did better. Prism has improved things of course, but even when all that was required was Rosetta 2, you can still see the impact. CPU translation takes a hit. How much is game dependent and how much translation is required, this latter bit also being my main point about the comparison between the platforms to begin with. To sum up:
www.notebookcheck.net
Yes, I had an entire second paragraph that addressed that even just WoA translation can cause performance issues. To expand on what I wrote, when it was just WoA pre-Prism translation was poor enough that Macs running games through Xover did better. Prism has improved things of course, but even when all that was required was Rosetta 2, you can still see the impact. CPU translation takes a hit. How much is game dependent and how much translation is required, this latter bit also being my main point about the comparison between the platforms to begin with. To sum up:
1. Yes, Asus is of course going to choose examples that benefit it. It's advertising. They aren't required to make it fair, though particular egregious examples abound which I do believe cross ethical lines (some of the Intel crap when the M1 was first released comes to mind). Reporting Diablo 3 scores doesn't cross that line because of #3.
2. macOS likely has many more native games than WoA
a) native games are going to, on average, be more performant than translated games, even when it is just translating x86 to ARM and nothing else. While I always look in askance of CPU makers advertising a 5% improvement in gaming over their competitor as some great win, taking a 25% or more hit in ST performance is going to matter for frame rates and the quality of graphics you can set to get those acceptable frame rates, some games more than others of course.
3. the majority of games will require translation layers for both and WoA will have an advantage requiring fewer translation layers, but I think it is fair to also point out that there is a difference in workload even when the average consumer doesn't care.
For instance, I do not include Qualcomm chips on my analysis of GPUs because I'm pretty sure they're not as inefficient as is shown because I'm pretty sure the NBC data comes from release pre-Prism:
Analysis of the Apple M5 SoC: Apple silicon extends its lead over AMD, Intel and Qualcomm
Notebookcheck analyzes the new Apple M5 SoC in the MacBook Pro 14 M5 in comparison with contemporary offerings from AMD, Intel, and Qualcomm.
And even post-Prism for Elite 2 GPUs I still might not include them because they'll be at a disadvantage that no other GPU is at (unless they are really good despite that disadvantage, then I'll admit that's worth noting). Now, I'm attempting to do hardware analysis, not advertising nor even "here's what the average consumer should expect" analysis. So I fully recognize that's different, but that's where I'm coming from.
Likely.
www.servethehome.com
Really? Because it sounded to me like they didn't want to work too hard at SMT so they did a minimal version of it.
Yes, that's pretty new. And Shilov is at TH. It's funny, he started out really clueless. Now he's the gold standard and head and shoulders above the other reporters there.
No, it was just a few months ago.