Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
what do you define as perform okay?
Given what I’ve been reading expecting anything more than 30fps on 1080p is not realistic unless you have a Max or Ultra version of the M3/M4 variants.

If I have to drop to 720p I’ll just use my son’s Xbox series x when he is not using it.
 
Given what I’ve been reading expecting anything more than 30fps on 1080p is not realistic unless you have a Max or Ultra version of the M3/M4 variants.

If I have to drop to 720p I’ll just use my son’s Xbox series x when he is not using it.
I have an M4 Pro and can get 30fps (often more) at 1080p.
 
  • Like
Reactions: Dagwood70
I think a bigger issue is getting devs to use their libraries at all. I’m sure if most games had as much effort put into optimization for Macs as they do for consoles or PC, then they would run just as well.

We’re at the point where getting a simple low effort port is difficult. Let alone expecting optimization.

EDIT: Though if I had to suggest a strategy, I think just developing DRM and/or anticheat software that works on Mac in place of the popular PC-only ones might go a long way for new titles, while just turning gptk into some compatibility layer with a permissive license to use it on older titles would fill a lot of their backlog.
The anti-cheat software problem is also on Steam Deck and on Linux, in general. Most of those companies don't seem to want to learn anything in order to add another platform and they don't trust anyone to port the software.
 
  • Like
Reactions: MRMSFC

Looks like a very heavy game even with the minimum RT that can be selected
On an M3 Max MBP, 1440p with MetalFX quality (around 960p internal) looks to be under 30fps (high preset with medium RT) in the benchmark. Even dropped to 1080p with MetalFX quality (around 720p internal) it's only just getting over 30 fps in the benchmark and in gameplay (with regular drops under 30 fps with traversal). It seems that dropping to 1080p with MetalFX balanced (around 640p internal), medium preset and dropping RT to the hideout only will get you to the high 40 fps region (and the game looks noticably worse in terms of the GI quality)

On an M4 Mac Mini, even dropping to 1080p/MetalFX performance/low cannot hold 30 fps

On an M1 Max MBP, even dropping to 720/MetalFX performance (!)/low settings just about holds 30 fps

On an M3 iMac, 720p/MetalFX performance/low settings drops below 30 fps with traversal

On an M1 MBA...you can guess what happens...

It definitely needs a patch!
 
And listing every free Mac game you find is on topic even it's not new, major or something to do with metal? Is this post I'm quoting of yours on topic? Or is telling people they're off topic on topic even though it's not really on topic?

I don’t know what’s hard to understand. I explained in detail when and why you were off topic in a particular discussion but once again you talk about the thread title while I was talking about the context in individual posts and your response in relation to those posts, not the topic of the thread. Trying to shift focus doesn’t help your credibility. This wasn’t about me but your post. This whole discussion could have never started or easily ended if you just had accepted the simple fact that the Mac in the test didn’t cost $15,000 but for some reason you can’t despite even admitting your post partly was a joke.

I joined MacRumors to get help, help others and make positive contributions to the community and forum, unlike some others who take every chance to criticize and make negative posts about Mac, Apple and Mac gaming, many times for no good reasons. It’s quite puzzling to see how PC gamers find Mac gaming news in a thread about Mac gaming off topic and seem annoyed by Mac gamers trying to enjoy free games. The thread title says ”major gaming news” because that was what it was at WWDC. It doesn’t mean all minor news afterwards are prohibited. The thread title says ”new games” because that was what it was at WWDC. It doesn’t mean news about old games afterwards are prohibited. The thread title says ”Metal” because that was what it was at WWDC. It doesn’t mean news about Mac games using other APIs afterwards are prohibited.

News about free Mac games, old or new is still news and not remotely off topic. News about free popular Mac games like Humankind or Dredge being free at launch on Epic before getting released on App Store is major for Mac gamers and serves everyone. Justifying and defending misinformation and disinformation serves no one, or at least no Mac user and apparently I wasn’t the only one thinking so.

Such news may not be interesting and seem off topic to PC/Steam Deck players who abandoned Mac years ago with great dislike for the HW/SW and Apple as company but just like PC gamers Mac gamers enjoy free stuff. This is a Mac gaming thread after all.

This for sure sounds like the Sniper Elite 4 discussion all over again with a high level of unnecessary nitpicking and trying to be right at any cost. Then in the same way I’ll try for the third time to clarify the situation with some QA in case of confusion and for better understanding.

1. What was post 3,852 about? The performance of AC Shadows on Mac Studio M3 Ultra 60c in Andrew Tsai’s video

2. What was wrong with that post? The estimated price of the Mac Studio, $15,000.

3. What’s the correct price? $3,999.

4. What was my post 3,856 about? Correction of the exaggerated price in post 3,852

5. What was your post 3,859 about? A direct response to me with complaints about the ”measly” amount of RAM and storage in the said Mac Studio and justifying the incorrect price information by talking about ”max config” and adding tax or switching to ”different dollar” to show that the mentioned Mac Studio ”very easily crosses the $15000 line”.

6. Was your post off topic in the discussion above and why? Yes because even if Mac Studio could cost a trillion dollar it wouldn’t be the Mac used in the test of AC Shadows. You don’t need a $15,000 Mac to test or play the game. The game specifications don’t require 512GB RAM and 16TB SSD for extra $8,600. You weren’t discussing AC Shadows and the performance either but how expensive and ”measly” a Mac is or can be.

7. Did anyone say Mac Studios are cheap and can’t cost $15,000? No, that wasn’t the subject of the discussion.

So what was the reason for the first poster to greatly exaggerate the price? Unclear since the poster never commented the issue afterwards and never reacted to my direct response to their post. The only explanation is that either it was a mistake or done intentionally. It’s less likely it was a mistake since Andrew Tsai has told about the specs in all his 4 videos and used the same Mac in the test. He is also the only one who has tested AC Shadows on M3 Ultra. So the specs and price are well-known.

What was the reason for you pointing out that M3 Ultra ”very easily crosses the $15000 line” in that discussion? You said your comment about ”measly” 96GB RAM was a ”joke” but otherwise haven’t given any other explanation in all your following comments. Som reasons I can think of is that you thought I was wrong which I wasn’t, you thought I was uninformed which I’m not or you think as many times before seen in your comments that Macs are overpriced with ”crummy” GPUs and ”measly” RAM and storage and wanted to take the opportunity to point that out once more even if it wasn’t the subject of the discussion. That may be all right under other circumstances but it was what made your response off topic and irrelevant. That and repeatedly making confusing posts just to call them jokes later also matches the definition of rule 7 in forum rules.

As for my other post or telling you being off topic the answer is yes, my responses were on topic discussing the issues with your posts. You chose to engage in direct discussion with me by quoting and opposing me. Do you expect people to remain silent when you quote them, especially when you tried to justify misinformation just to say it was partly a joke? Maybe if you didn’t joke repeatedly and wrote more serious and relevant answers there wouldn’t be such ”off-topic" posts as result to annoy you.

And no I wasn't joking about getting a G3 Powerbook if I can get one cheap. I wouldn't do anything with it really. It would just be for messing around with Classic macOS, maybe trying to shove Rhapsody on there. It would be more for collecting sake, than doing sake.

Regarding you wanting to get a PB G3, again I can’t tell any longer if you’re being serious or joking. That’s another confusing and contradicting statement by you. Back in January you complained in several long posts about how much effort it supposedly takes to get 32-bit games to work on Mac and how much time and money you would be wasting if you wanted to keep old macOS versions or get and repair old Macs and buy extra storage but now all of a sudden you don’t mind to put the same effort, time and money into finding and fixing an old Mac just to do nothing with it? Sounds reasonable.

No backwards compatibility has a cost too. Money and time wasted hunting for replacements to apps that no longer work, money and time wasted hunting for replacement apps because the compatible version is now some gross subscription riddled crap, money and time wasted doing solutions like buying extra drives to keep old OS versions and repair 15 year old computers because it won't work on something newer. These are all costs born by the consumer. I don't care if backwards compatibility is harder for Apple. That's their problem. They charge a premium price for their products and are almost a 4 trillion dollar company. They should be able to figure it out. Instead in Apple spaces it's made out to be the individual developers problem which is non-sensical. Why should 1000s of entities do a bunch of non-trivial non-free work because one company can't be bothered?
 
Last edited:
A screenshot would improve that person credibility since Andrew was getting more than 19 fps on a 60 core model.

Yes but Andrew Tsai never tested the game at max settings. He used mostly Low/Medium and RTGI in the Hideout. Even when he raised the RT settings he used MetalFX Performance at all time during the test.
 
100% per core, over 100% when the game uses multiple cores. Digital Foundry talked about it too.

I know but it is still higher than what is seen on PC. And of course the CPU utilization goes up when the resolution goes down. The M3 Max was around 30-50% total CPU utilization (100% for every core and it has 16 of them), which is pretty high. And it wasn't the only device to see that kind of utilization in the benchmark.
 
I know but it is still higher than what is seen on PC. And of course the CPU utilization goes up when the resolution goes down. The M3 Max was around 30-50% total CPU utilization (100% for every core and it has 16 of them), which is pretty high. And it wasn't the only device to see that kind of utilization in the benchmark.

Why would the game using 3 cores on M1 Max in total be high? 30-50% in the benchmark doesn't mean it's using 16 cores. It could also mean 50/16, 3% per core. Maybe the game doesn't use the GPU very well on Mac and relies on CPU? Who knows. Even on 7950X3D the CPU usage is 30-50% with 16 cores.

Skärmavbild 2025-03-25 kl. 11.31.10.png
 
Last edited:
Why would the game using 3 cores on M1 Max in total be high? 30-50% in the benchmark doesn't mean it's using 16 cores. It could mean 50/16, 3% per core. Maybe the game doesn't use the GPU very well on Mac and relies on CPU? Who knows. Even on 7950X3D the CPU usage is 30-50% with 16 cores.

View attachment 2495532
I saw that he mentioned that, but then never showed any screenshots with the numbers that high. I am not saying it didn't happen, but I've noticed no one really talking about it on the PC side. Then again no one is talking about it on the mac side either.

I will say the presets are dumb, I feel like there are quite a few settings that don't apparently matter (I posted a video in the AC:S thread asking if anyone saw performance improvements with some custom settings, seems like no one has tried them).

I'd like to pick the game up on mac, but with having a M1 Max 24C that seems like a bad idea. If there is a performance improvement I'll probably buy it for macOS.
 
I saw that he mentioned that, but then never showed any screenshots with the numbers that high. I am not saying it didn't happen, but I've noticed no one really talking about it on the PC side. Then again no one is talking about it on the mac side either.

I will say the presets are dumb, I feel like there are quite a few settings that don't apparently matter (I posted a video in the AC:S thread asking if anyone saw performance improvements with some custom settings, seems like no one has tried them).

I'd like to pick the game up on mac, but with having a M1 Max 24C that seems like a bad idea. If there is a performance improvement I'll probably buy it for macOS.

I have excatly the same M1 Max (Studio). Forget about it. It is what it is. Let's hope for future optimization then maybe or when it's time for a new M5-10. Menawhile we'll have Cyberpunk and Control coming.
 
  • Like
Reactions: diamond.g
I have excatly the same M1 Max (Studio). Forget about it. It is what it is. Let's hope for future optimization then maybe or when it's time for a new M5-10. Menawhile we'll have Cyberpunk and Control coming.
Speaking of, it appears Control is out now.
 
  • Like
Reactions: bousozoku

Seems a good port, though I am puzzled you get almost the same performance with crossover ? That doesn't make sense right?
Anyways Apple needs to double up the GPU performance for each chip to make RT even remotely useable.
Yeah. Baffling to me that this person described it as a good port when it get similar fps to gptk. If anything that’s a sign it has barely been optimized for the Mac.
 
I guess it's not cpu intensive, and Apple Silicon has such large single core performance that is enough even to run it with Rosetta.

GPTK performances depend a lot of which game you are running, 3d API translation layers can be quite fast, it could be Control didn't need any weird feature that can't be mapped on Metal, and so the it was fast enough even on GPTK. But for example the native versions uses quite fewer RAM.
 
So they didn't add the HDR and Ultra RT settings to the macOS version?
He says "Lastly, Control still hasn’t received the recent update on Steam—it doesn’t support HDR yet," so he is running the steam version? Have no idea about the App store version.
 
No, they mean that the App Store version doesn't have the HDR feature that the Windows version on Steam got recently.

It could come in a future update.
 
So AC:S got an update today. The Mac update is supposed to be 9GB, is that accurate, or do you have to download the whole thing?
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.