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.
Feral Releases 'Total War: Warhammer III' Optimized for Apple Silicon Macs
- Thread starter MacRumors
- Start date
- Sort by reaction score
wait wait wait
so its x86-compiled, works via rosetta, but does not support intel macs?
this gets better and better
for those screamin "hey nice its m1 optimized" - lmao
If we really want more AAA titles for Apple Silicon Macs we need to show developers that we support them. Vote with your wallets! ?
wait wait wait
so its x86-compiled, works via rosetta, but does not support intel macs?
this gets better and better
for those screamin "hey nice its m1 optimized" - lmao
Before you "lmao" you have to understand that a game can be optimized in many ways. The game can still be optimized for the GPU architechture of Apple Silicon and TBDR (Tile-Based Deferred Rendering) that it uses instead of AMD or Intel iGPU. AMD and Intel don't have TBDR features and Mac games in Metal have to use work arounds with those GPUs.
wow. $60 game. 125GB in size. this has to be good. Love to get it for the PC. must have great 4k graphics.
I also have an external 2TB Samsung 970 Evo Plus, and have pointed most User folders to it, FCPX library, videos, Music library, etc.
The main culprits at this point are iCloud Drive, Xcode the Apps folder and Logic Pro loops & instruments. I cannot move any of these things.
I tried installing my entire macOS system on the external SSD but Apple blocks it.
So, I run some benchmarks and I am disappointed. I am getting around 40Fps at 1920x1200 ultra and 15fps at 4K ultra. It’s definitely playable on my max at native resolution once you tweak the settings a bit, but it’s not even close to what the chip is capable of. Looks like another poor quality port without optimizations. The air should run it at full HD low/medium settings, but it’s missing the mark.
Really a shame, I hoped for a high quality native version (Feral all but promised that), but that’s not what we got. Oh well.
Hmm.
Something sure feels odd about that. Do you think there is something specific in those "ultra" settings that are messing things up?
Which Mac did you benchmark? Did you run multiple tests? If Rosetta is involved you may get poor results the first time. Also check the ultra settings. If supersampling SMAA is on your are rendering at higher resolution when using ultra settings.
I don’t see why it would. It shouldn’t really matter that much if a game runs natively or under Rosetta. The important part is in the algorithmic approach and it is clearly far from being optimal.
There usually is. Often one or two settings hit the bad path in the engine. But I didn’t have time to try it out. Frankly, given the state of the game and the fact that I currently have enough things to do I was considering refunding until CA brings out significant patches. So I don’t want to get over the Steam refund limit
M1 Max 16“. Rosetta is an AOT transpiler, first or subsequent times should not matter. As to the settings, I used the default ones to make it more comparable to what third?party reviews use.
Yeah, um, you missed the trend in your own chart....
| Month | Users | Growth |
|---|---|---|
| Dec | 27.97% | |
| Jan | 32.83% | 4.86% |
| Feb | 35.95% | 3.12% |
| Mar | 37.78% | 1.83% |
| Apr | 38.89% | 1.11% |
Decelerating growth.
But presumably jumps up with each new AS product release, then slowly decelerates again.
Being a 15" rMBP user, I was excited by, but didn't buy the first M1 13" MBA/MBP/Mini release. When the 14/16" MBPs came out, I waited a few months to make sure there wasn't any weird flaws, and my 16" MBP order is on the way.
I can only presume there are a handful of other users who are waiting, maybe on a headless w M1 Pro chip (a Mini Pro? It makes sense! A slightly bigger Mini with more ports and M1 Pro/Max options), or the new MBA design, or the Mac Pro, or holding out in vain for an iMac Pro or bigger iMac, or hoping for a 15" MBA.
It's pretty much 40% adoption already, which is incredibly impressive. So maybe the majority of the rest are happy to hold onto their machines that are working perfectly fine for their needs and will upgrade in due time.
My old 15" rMBP is 7 years old and still chugging along just fine. But I did skip the woeful 2016+ years. The pre-2016 models are still an absolute delight. If I had a 2016+ machine, I think I might have rushed to upgrade quicker.
Last edited:
Ultra settings are pretty hard on everything with this game. IIRC the 3090 only gets like 50fps at 4K Ultra.I don’t see why it would. It shouldn’t really matter that much if a game runs natively or under Rosetta. The important part is in the algorithmic approach and it is clearly far from being optimal.
There usually is. Often one or two settings hit the bad path in the engine. But I didn’t have time to try it out. Frankly, given the state of the game and the fact that I currently have enough things to do I was considering refunding until CA brings out significant patches. So I don’t want to get over the Steam refund limit
M1 Max 16“. Rosetta is an AOT transpiler, first or subsequent times should not matter. As to the settings, I used the default ones to make it more comparable to what third?party reviews use.
Does MacOS have the unlimited video memory option?
The M1 iPads would struggle with RAM limitations wouldn't they? If not OS limitations.
Yes, but that larger pool of existing Intel Macs mostly has crappy GPUs. Either crappy iGPUs or low end AMD discrete. Even the best 27" iMac with the best AMD Radeon Pro 5700 XT 16 GB isn't really *that* great and how many Intel Macs have that class of GPU? Not many.....
I doubt the issue is really around CPU architecture, but more around GPU capabilities.
Agree - ultra settings are basically the graphics card benchmarking/marketing setting of the industry. It’s there to sell the next generation of cards.
But in our case there may be more to it - a setting or settings that throws a pile of gravel into the tiling approach, or something else that is a bad match for AS graphics specifically.
First time I ran the benchmark in Shadow of the Tomb Raider after installation on my MBP 14" M1 Pro I got 13 fps on high at 1080p. I thought my Mac was faulty. The second time I got 45 fps as it should be. I got similar result in Metro Exodus so I do believe it's something about running the game for the first time. Rosetta kicks in the first time you start an app and I suspect that the app itself is compiled but there may be other assets that are not directly compiled before they're used? I'm not sure but that was my experience with the games the first time.
This is the thing that annoys me with game publishers, they release something gloating about how 'optimized' the game is for the hardware and then as the days goes on it becomes apparent that the game is not optimised but actually sort of optimised. This game is a very very good example 'oh it's optimized for M1, it's native to M1' but then out comes the social media posts, 'well it's not truely native because some of the stuff still needs to be in x86', then as more days goes on 'actually the game is not native for the M1 because it has to run x86 code for online player use, but it's still just as good though'. It get's worse and worse as the days and weeks go on.
It's not my chart, it's Steam's chart.
While true I didn't miss the deceleration. I was stating that the transition as a whole is happening fast despite the deceleration. It's about 10% in just the past 4 months. Like the stock market the transition rate can go up and down over a year depending on what and when Apple releases regarding to Macs. The deceleration can be due to the long delivery times. Mac Studio was released back in March but the shipping times are now up to 2-3 months.