Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

RTX 5090 and M4 MacBook Air: Can It Game?

scottjg.com

Post reply on HN
Print view View original

181–190 of 195 posts

Re: RTX 5090 and M4 MacBook Air: Can It Game?

#181

Post by bigyabai »

astrange wrote: post #176
bigyabai wrote: post #152

Earlier quoted context omitted.

Porting games natively to macOS is a waste of developer time. Apple has already depreciated vast swathes of 32-bit games that were never updated to support 64-bit x86 or Apple Silicon. Developers that give macOS the same level of attention as Windows don't get the same level of support that Microsoft offers in return. Not to mention that Mac owners are a minority share of the PC gaming market. Linux has the right ide…

> Apple has already depreciated vast swathes of 32-bit games that were never updated to support 64-bit x86 or Apple Silicon. They had literally 15 years of warning about this.

Okay, now literally count how many developers went back to update their vintage macOS games.

I'm not going to blame the developers here because it's not their fault.

Re: RTX 5090 and M4 MacBook Air: Can It Game?

#182

Post by bigyabai »

fgfarben wrote: post #175
bigyabai wrote: post #168

Earlier quoted context omitted.

We're quite lucky that Nvidia didn't ship a 256gb system at sub-500gb/s transfer rate, is my point.

> Nvidia didn't ship a 256gb system at sub-500gb/s transfer rate DGX Spark has 128 GB and only 273 GB/s BW. Are we lucky that NVIDIA did ship something even worse than what you specified? I'm confused. People have been complaining [1] about how little VRAM NVIDIA ships with their GPUs for decades. Their whole game has been "oh, you want more VRAM? Buy more or pay us 50x for server grade with 10x as much VRAM. The mor…

The DGX Spark is also a niche, arbitrarily limited machine that will not displace serious datacenter workloads. It's targeted directly at the homelab LARPers and arguably a waste of money versus similarly priced GPU clusters. A 256gb Spark at LPDDRX5 transfer rates would be a genuine travesty.

You can try to weasel out any sort of edge justification you want - these are not industry-grade machines. They are slow, expensive, bandwidth-constrained SOCs that don't hold a candle to either datacenter GPUs or even decade-old gaming GPUs. It's worth criticizing when Apple does it, and also worth criticism when Nvidia does it. The only difference being that Nvidia has natural datacenter buy-in, while Apple can't even justify their own hardware in the face of TPU inference costs: https://9to5mac.com/2026/03/02/some-apple-ai-servers-are-rep...

Re: RTX 5090 and M4 MacBook Air: Can It Game?

#183

Post by zer0zzz »

traderj0e wrote: post #107
zer0zzz wrote: post #51

Once egpus work on Apple Silicon there will be little reason to own a pc

Been hearing this for over a decade, except back then it was eGPU in Intel Macs which were closer to other PCs if anything. Even if this didn't require so much DIY and if Thunderbolt could do PCIe speeds, most people don't want to add drama when they can just use a PC with regular PCIe slots and native compatibility with Nvidia. The native way already has enough edge cases without adding an unusual setup.

What would be native enough here? What if they got Asahi working with NV gpus for rendering and running cuda kernels? Would eGPU on asahi be sufficient or do you really only see pcie worthwhile?

Some of us mainly want more gpu options on a high performance consumer arm machine (for Linux).

Re: RTX 5090 and M4 MacBook Air: Can It Game?

#184

Post by traderj0e »

zer0zzz wrote: post #183
traderj0e wrote: post #107

Earlier quoted context omitted.

Been hearing this for over a decade, except back then it was eGPU in Intel Macs which were closer to other PCs if anything. Even if this didn't require so much DIY and if Thunderbolt could do PCIe speeds, most people don't want to add drama when they can just use a PC with regular PCIe slots and native compatibility with Nvidia. The native way already has enough edge cases without adding an unusual setup.

What would be native enough here? What if they got Asahi working with NV gpus for rendering and running cuda kernels? Would eGPU on asahi be sufficient or do you really only see pcie worthwhile? Some of us mainly want more gpu options on a high performance consumer arm machine (for Linux).

Thunderbolt still doesn't provide the full PCIe bandwidth, but even if it did, I'd want PCIe itself. I don't trust the encapsulated version over Thunderbolt to work the same.

Virtualized Linux would be ok though. That's what datacenters already do with their GPUs, albeit on x86 not ARM. Doesn't need to be Asahi, cause that's unlikely to completely work.

Re: RTX 5090 and M4 MacBook Air: Can It Game?

#185

Post by traderj0e »

jaimex2 wrote: post #147
zer0zzz wrote: post #51

Once egpus work on Apple Silicon there will be little reason to own a pc

The only thing Apple silicon has going for it is power use and that gap is getting closed. I can't really see any reason why I would switch to Mac, it just seems like you pay a lot more for a closed expensive environment that fights you at every step. I'll never pay anyone for a developer licence or fee either. They can sponsor me to port my software to their platform.

Mac lets you run any software you want, but I understand the principle of not wanting to support them.

Re: RTX 5090 and M4 MacBook Air: Can It Game?

#186

Post by traderj0e »

zer0zzz wrote: post #178
lowbloodsugar wrote: post #85

Earlier quoted context omitted.

Just built a workstation with an older Threadripper Pro. It has 128 PCIE lanes, for 7 16-lane PCIE slots. An egpu has 4. I have one GPU, at x16, and I can add more. Most people don't need that, but most people don't need an eGPU either. The number of gamers who would switch to Macbook+eGPU is negligible. It's just not compelling. For LLMs, hanging a 5090 off the thunderbolt port makes prompt processing fast, but I wi…

FWIW, I am partial to eGPUs not for laptop gaming but for space. I want to write cuda kernels at home but I dont want a big tower. I have a Minisforum mini-pc the size of an M4 Mac Mini attached to one of those egpu enclosures and the whole setup was pretty easy and sits on my desk nicely.

Yeah the desire makes sense. The Macs are very nice hardware. You can get a mobo/case that's not much larger than the GPU itself, but it's still clunky. It's just, unfortunately eGPUs are unlikely to become more than a curiosity.

Re: RTX 5090 and M4 MacBook Air: Can It Game?

#187

Post by fgfarben »

bigyabai wrote: post #182
fgfarben wrote: post #175

Earlier quoted context omitted.

> Nvidia didn't ship a 256gb system at sub-500gb/s transfer rate DGX Spark has 128 GB and only 273 GB/s BW. Are we lucky that NVIDIA did ship something even worse than what you specified? I'm confused. People have been complaining [1] about how little VRAM NVIDIA ships with their GPUs for decades. Their whole game has been "oh, you want more VRAM? Buy more or pay us 50x for server grade with 10x as much VRAM. The mor…

The DGX Spark is also a niche, arbitrarily limited machine that will not displace serious datacenter workloads. It's targeted directly at the homelab LARPers and arguably a waste of money versus similarly priced GPU clusters. A 256gb Spark at LPDDRX5 transfer rates would be a genuine travesty. You can try to weasel out any sort of edge justification you want - these are not industry-grade machines. They are slow, exp…

What even is an industry grade machine?

Would you own a computer if the smallest computer you were allowed to buy was a $27,000 Supermicro rack that draws 900 watts all the time?

Re: RTX 5090 and M4 MacBook Air: Can It Game?

#188

Post by ActorNightly »

zer0zzz wrote: post #177
ActorNightly wrote: post #142

Earlier quoted context omitted.

Man, Apple fans are still proving the stereotype to be accurate after 20 years. Ignoring the fact that the Mac OS gets in your way every time you try to do something that Apple doesn't like, with no guarantee that an update won't break anything existing, ignoring the fact that Macs are non repairable, non upgradable, ignoring the fact that they don't support multiple displays flawlessly, I hope you realize that egpu…

> Man, Apple fans are still proving the stereotype to be accurate after 20 years. Who is this straw man you're flogging? > Ignoring the fact that the Mac OS gets in your way every time you try to do something that Apple doesn't like, with no guarantee that an update won't break anything existing, ignoring the fact that Macs are non repairable, non upgradable, ignoring the fact that they don't support multiple display…

Mac will appear to give some leeway to fake being dev friendly, but they are not. There is a reason why still Asahi is in its state - lack of any real documentation from Apple. If Apple was dev friendly, they would just bring those people on board and give them the documentation and have them develop a fully working linux for free. But Apple fundamentally DGAF about linux users.

RDMA over Thunderbolt is going to be used only with Mac devices. Apple has a history of keeping things within their own ecosystem. You gotta be insane to think that they are going to just magically allow you to plug in a graphics card and it will work natively.

The point of bringing up Android is because that is what being dev friendly. Samsung or Google have nothing to really gain to enable the desktop mode. But they do it anyway because it increases the usability of their devices. Ask yourself again, if Apple already runs arm on all of its devices, why not enable a desktop mode for the iPhone? Its EXACTLY for the reason to squeeze more money from consumer. Its why they do the thing they do with app store, that why they own all the advertising streams on their devices.

So if you wanna stay deluded about what Apple does, be my guest. Just don't be surprised when nothing turns out like you hoped.

Re: RTX 5090 and M4 MacBook Air: Can It Game?

#189

Post by zer0zzz »

traderj0e wrote: post #184
zer0zzz wrote: post #183

Earlier quoted context omitted.

What would be native enough here? What if they got Asahi working with NV gpus for rendering and running cuda kernels? Would eGPU on asahi be sufficient or do you really only see pcie worthwhile? Some of us mainly want more gpu options on a high performance consumer arm machine (for Linux).

Thunderbolt still doesn't provide the full PCIe bandwidth, but even if it did, I'd want PCIe itself. I don't trust the encapsulated version over Thunderbolt to work the same. Virtualized Linux would be ok though. That's what datacenters already do with their GPUs, albeit on x86 not ARM. Doesn't need to be Asahi, cause that's unlikely to completely work.

That's understandable.

Re: RTX 5090 and M4 MacBook Air: Can It Game?

#190

Post by zer0zzz »

traderj0e wrote: post #186
zer0zzz wrote: post #178

Earlier quoted context omitted.

FWIW, I am partial to eGPUs not for laptop gaming but for space. I want to write cuda kernels at home but I dont want a big tower. I have a Minisforum mini-pc the size of an M4 Mac Mini attached to one of those egpu enclosures and the whole setup was pretty easy and sits on my desk nicely.

Yeah the desire makes sense. The Macs are very nice hardware. You can get a mobo/case that's not much larger than the GPU itself, but it's still clunky. It's just, unfortunately eGPUs are unlikely to become more than a curiosity.

I am definitely enjoying mine, but I've found for what I do some of the hosted kernel writing tools are easier to deal with. I actually had to do a ton of work arounds for my tb5 egpu on Linux recently for it to not drop off the pci bus every ~48 hours.
Post reply on HN