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.
Steam now supports ARM64
- Thread starter sunny5
- Start date
- Sort by reaction score
So the gist of this is that the Steam application will now run on new Macs? I wonder how hard it is for developers to produce an ARM version of their games, and if porting it to ARM architecture means there will be a performance hit similar to what Mac porters have done in the past?
Last edited:
I'm making some rather large assumptions but here's my thoughts.
If the game already had a macOS version that was 64bits - year the work shouldn't be too intensive. Notice that I didn't say easy or a simply recompile because development work is rarely as easy as recompiling and moving on without any editing/debugging/testing.
If the macOS game is 32bits, then its highly unlikely anything will come of it, as the publisher hadn't done anything to move those games over to 64 bit.
If the game is windows based, then the work will most certainly be intensive, and time consuming as they'll basically have to develop the game to run a platform that it was not developed to run on.
Didn’t Mac porters use to use a conversion software, (that converted calls, or added interpretations) or was it painstakingly done line to line?
Yes, there were, but that was typically done by third party companies who then sold the game under their label - at least that was my experience for some of the older games such as from MacSoft. I'm not sure what the situation is now with more modern games.
Note that it's not the Steam client itself that is yet native but the overlay but it certainly makes things easier for the developers. Now they can deliver Apple Silicon native games on Steam too without separate launchers or other fixes.
Developing new games and porting them usually takes 3-5 years. When it comes to porting existing Mac games to Apple Silicon I understand why developers/porting houses take their time. One reason is Rosetta 2 speed. Last time native ports were much more needed because Rosetta 1 was so slow. It is much better now because Apple has implemented x86 features directly into the M1. This makes Rosetta 2 so fast it sometimes is faster than a native Intel app. Some apps/games also don't gain much speed or any at all with a native M1 port. There are already several examples.
Another reason is the upcoming M1X and M2 with 16-32 GPU cores or more. The current problem with bad performance in games is not Rosetta but M1 8-core GPU. Although many larger games still may lack good performance on M1 developers know that Macs in the near future, as near as the next month, will have 2-4 times better graphics performance. It means you should be able to run Borderlands 3 at 1440p ultra settings at 60 fps with 32 GPU cores in Rosetta. There wouldn't be much need for porting existing games with such performance.
One example is Feral. They have tested their existing games and ensured that they run fine. These games will fly on the future Macs for many years. For that reason they have decided to put their efforts on new native ports for Apple Silicon. As you can see Total War: Rome Remastered already has native M1 support and so will Total War: Warhammer III.
"For now we intend that all new Feral releases will offer native support for both new Apple hardware and older, Intel-based Macs. Our goal is to accommodate as many Mac gamers as possible for as long as it is technically feasible and commercially viable."
I think that's the path many developers have taken. People shouldn't focus so much on status quo and should instead have some patience.
Developing new games and porting them usually takes 3-5 years. When it comes to porting existing Mac games to Apple Silicon I understand why developers/porting houses take their time. One reason is Rosetta 2 speed. Last time native ports were much more needed because Rosetta 1 was so slow. It is much better now because Apple has implemented x86 features directly into the M1. This makes Rosetta 2 so fast it sometimes is faster than a native Intel app. Some apps/games also don't gain much speed or any at all with a native M1 port. There are already several examples.
Another reason is the upcoming M1X and M2 with 16-32 GPU cores or more. The current problem with bad performance in games is not Rosetta but M1 8-core GPU. Although many larger games still may lack good performance on M1 developers know that Macs in the near future, as near as the next month, will have 2-4 times better graphics performance. It means you should be able to run Borderlands 3 at 1440p ultra settings at 60 fps with 32 GPU cores in Rosetta. There wouldn't be much need for porting existing games with such performance.
One example is Feral. They have tested their existing games and ensured that they run fine. These games will fly on the future Macs for many years. For that reason they have decided to put their efforts on new native ports for Apple Silicon. As you can see Total War: Rome Remastered already has native M1 support and so will Total War: Warhammer III.
"For now we intend that all new Feral releases will offer native support for both new Apple hardware and older, Intel-based Macs. Our goal is to accommodate as many Mac gamers as possible for as long as it is technically feasible and commercially viable."
I think that's the path many developers have taken. People shouldn't focus so much on status quo and should instead have some patience.
They're busy working on Steam for ChromeOS. I wouldn't expect the Steam Client on macOS to be on ARM binaries until the Chromebook client is finished. Not to mention the Steam Deck kinda took priority since it was a blowout success and is doing numbers. And this isn't even factoring in SteamOS for Desktops not being finished when it was supposed to have come out a year ago.
Valve Time baby
With how long it's taking one would think it's written in assembly. Well, one day they'll have to get it done. Hopefully.
It's not so much the code but moreso how Valve operates. Valve has a specific way of doing business. Valve's philosophy is you can work on whatever project you want so long as it's providing value, and you can jump to any project you choose. Many staff jump across different projects, to the point everyone's desk at Valve are on wheels so they can easily move them to different projects and rooms.
This is such a core part of their work philosophy that a common saying around the Valve office is "oh, bring your desk," a phrase referenced and joked about in their Steam Deck tech demo Aperture Desk Job.
On one hand, it is annoying as they start so many projects but never finish them, most infamously Half Life 3 and Left 4 Dead 3 as they were working on both but the staff all lost interest due to not having the tech they needed at the time and not having a clear scope. But on the other hand because of this philosophy we have gotten so many unique left field products from them such as VR, the Steam Deck and well, STEAM. So while yes it is infuriating they aren't doing what we want, in turn they make stuff that we didn't even knew we wanted, like the Steam Deck.
In theory that’s the way it’s supposed to work, but there have been reports that senior developers use their clout to push junior developers into working on the pet projects of senior developers. https://www.pcgamer.com/ex-valve-employee-describes-ruthless-industry-politics/It's not so much the code but moreso how Valve operates. Valve has a specific way of doing business. Valve's philosophy is you can work on whatever project you want so long as it's providing value, and you can jump to any project you choose. Many staff jump across different projects, to the point everyone's desk at Valve are on wheels so they can easily move them to different projects and rooms.
![R.b22b03ae71a6e8763af33bdbc73737a9]()
![OIP.qPO21ZlG_U2TrwLj9Oi1GAHaE7]()
This is such a core part of their work philosophy that a common saying around the Valve office is "oh, bring your desk," a phrase referenced and joked about in their Steam Deck tech demo Aperture Desk Job.
On one hand, it is annoying as they start so many projects but never finish them, most infamously Half Life 3 and Left 4 Dead 3 as they were working on both but the staff all lost interest due to not having the tech they needed at the time and not having a clear scope. But on the other hand because of this philosophy we have gotten so many unique left field products from them such as VR, the Steam Deck and well, STEAM. So while yes it is infuriating they aren't doing what we want, in turn they make stuff that we didn't even knew we wanted, like the Steam Deck.
I believe this person is exaggerating, but it does bring up the point that Valve is opaque, and to assume it’s a kind of developer candyland is silly. I don’t know if it’s got better in the time since then.
There’s also the fact that to finish Half Life Alyx they had to move to a standard structure. https://arstechnica.com/gaming/2020...ing-half-life-3-in-new-steam-documentary-app/
The fact is that Valve does have structure and bosses when it wants to.