is not like these are 2 identicals devices...so different display tech and brightness...true tone is on and the lightning is not in the middle of the two..who cares about the lightning at the end of the day
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.
M1 Pro/Max's GPU performance for gaming is quite poor.
- Thread starter sunny5
- Start date
- Sort by reaction score
Eh, the 3080 could have it’s place in this game, turning everything to Ultra and enabling RT Shadows just because you can is nice I guess. At the end it showed neither system was 99% GPU bound so (the Mac was more GPU bound than the PC) yeah turn up the quality till you are in my opinion.
It was just weird, like HDR was enabled on the Windows version but not the Mac version. I just ran the bench on my desktop and it looked like the Mac version (color wise).
SoTR is said to be an unoptimised port for macOS and should be optimised for Windows? Imagine if it’s optimised for Metal and TBDR.Apple is putting to shame the whole industry with just M1 , M1 pro and M1 max that are in 5-6 macs
Glad we have windows still...because otherwise with this kind of scaling, in 2-3 years Apple could have faced anti-trust lawsuit
Trying to figure out why he's getting such low FPS on a 3080 when a Lenovo Legion 3060 which is a step below Thinkpad is higher at similar TGP.
3060 100W
3060 100W with DLSS quaiity
Hope he's not pulling something like this guy who has VSYNC on.
The screenshot shows a resolution of 1920x1080. The video shows the benchmark running at 1920x1200.
Don’t think it’ll matter? The GTX3080 clearly is running on at least a 120Hz screen.
Agree. My real concern would be lack of m1 native programs in general. Though once rosetta is gone this might pick up.
For that reason while i might grab a base mini or air to screw around with, I’m not willing to go beyond that in spending on macs. Currently it cannot get things done for me for work since I require windows programs. Gaming is even worse and not a consideration. That leaves typical tasks. And for that an mba is pretty good factoring in battery. An updated mba with these new screens would be even better.
Nice performance by the M1 Max, but let's not forget that on the very same game, anandtech showed that the M1 Max was much slower than a radeon 6800M, which is itself slower than the nVidia card. I'm not sure why it's not the case here, but I suspect overheating on the PC, a downclocked GPU or faulty testing methodology.
What Apple is lacking:
1. They did not work with major game studios to develop games that can take advantage of M1 Pro/Max.
2. They did not purchase any studio to make first-party games
3. They did not create a gaming division at Apple
4. They did not create MacOS as a gaming platform. This is a more complex thing to fully answer in a few sentences, but you can read this thread for more details:
Vulkan and Metal (some observation… | Apple Developer Forums
^ in contrast, iOS, which is based on the exact same technologies as MacOS, is in fact geared more toward gaming than MacOS. iOS has support for game controllers now, and the platform itself comes with standard sensors to achieve motion and AR/VR very easily. MacOS is still, at the end of the day, more of a platform for developers and content creators, rather than for gaming.
In my opinion Apple is doing the right thing ignoring gaming. The PC gaming crowd would never accept any olive branch from them.
Last edited by a moderator:
What Apple is lacking:
4. They did not create MacOS as a gaming platform. This is a more complex thing to fully answer in a few sentences, but you can read this thread for more details:
Vulkan and Metal (some observation… | Apple Developer Forums
developer.apple.com
I think it’s hilarious that you took a post I wrote 5 years ago to argue that macOS is not designed for gaming
Metal has changed considerably in the meantime. It has added a lot of advanced features and more low level control. It’s a top notch API and Apple delivers top-notch GPU debugging experience. Every M1 Mac is basically a gaming console. Macs support the same controller and ML APIs that iOS does. And Macs have faster hardware that can run more advanced games.
Last edited:
Absolutely priceless.I think it’s hilarious that you took a post I wrote 5 years ago to argue that macOS is not designed for gaming
Macs support the same controller and ML APIs that iOS does. And Macs have faster hardware that can run more advanced games.
I think it’s hilarious that you took a post I wrote 5 years ago to argue that macOS is not designed for gaming
Macs support the same controller and ML APIs that iOS does. And Macs have faster hardware that can run more advanced games.
No, I get that. Fundamentally, iOS and MacOS are built on the same base. But there is more to it than just the OS.
As far as AR/VR goes, sensors necessary to achieve that are not standard on MacOS, and I don't recall there being any support for Oculus on M1.
Also Apple does not build games themselves, nor do they work with other developers to bring more serious AAA games to Mac computers. So despite all M1 Macs being very capable, they just can't do it without... y' know, games.
In contrast, the Nintendo Switch is not a very powerful or efficient machine. In fact, most phones after 2016 should be more powerful than it. But Nintendo makes 1st party games, and they also work together with major game studios to either port games over or bring more exclusives to the Switch. That's how a gaming console should be.
OK I don't often post here, but after playing with my M1 Max (32-core) for an entire weekend mostly with gaming, here's what I can definitively tell you after benchmarking almost everything I have, including games on Crossover. Mac games have almost always performed worse than on Windows on the same hardware. Anyone who gamed on Bootcamp int he last 15 years can tell you that. So you have to look like for like to see how fast an M1 Max really is.
My M1 Max scores anywhere between 20% and 50% faster than my Desktop Vega 64 depending on the game. It's not like for like because the CPUs are different, but even under Rosetta (like SoTTR and RoTTR, Civilization etc) it was close to 50% faster in "minimum fps" and 20 - 30% faster on average. This is close to the Geekbench Metal score difference, which is exactly 30% (53000 for my Vega 64 and 68753 for my M1 Max).
3637933 (Vega 64)
3311403 (M1 Max)
Now, in Windows with 2021 drivers, a Vega 64 is about on par with an RTX 2070. So 30% faster than that is pretty much matching a desktop 3070. That's quite incredible.
You can't expect games developed 5 years ago, then ported 3 years ago from Windows to run faster than the very highest spec Windows GPU through emulation on a laptop no developer had even seen until a week ago.
My M1 Max scores anywhere between 20% and 50% faster than my Desktop Vega 64 depending on the game. It's not like for like because the CPUs are different, but even under Rosetta (like SoTTR and RoTTR, Civilization etc) it was close to 50% faster in "minimum fps" and 20 - 30% faster on average. This is close to the Geekbench Metal score difference, which is exactly 30% (53000 for my Vega 64 and 68753 for my M1 Max).
3637933 (Vega 64)
3311403 (M1 Max)
Now, in Windows with 2021 drivers, a Vega 64 is about on par with an RTX 2070. So 30% faster than that is pretty much matching a desktop 3070. That's quite incredible.
You can't expect games developed 5 years ago, then ported 3 years ago from Windows to run faster than the very highest spec Windows GPU through emulation on a laptop no developer had even seen until a week ago.