Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Absolutely accurate. However the AVP's current limitations revolve around neither speed/ray tracing/CPU nor display quality. The AVP's current limitations are visionOS competence. Much like with iOS, improving the OS is the bottleneck.
I’d say limitations also (probably mainly) revolve around hardware.

The VP display quality is great compared to the competition, but unfortunately the PPI needed to simulate reality next to your eyes must be absurdly high. And there are other factors like fov. It’s much more convincing than what we had seen (maybe enough-ish), but there’s a lot of room for improvement.

Also, doing a detailed analysis of CPU needs would be too much for me, but I think it’s obviously a lot: the amount of pixels to handle, combined with the need to process data from multiple LiDARs with almost no acceptable lag, keeping windows perfectly anchored, etc. I bet Apple considered M1/M2+R1 as the minimum requirement, but we’re not even close to comfortable “overkill” levels (as we have with iPhones or Macs). We’re talking about covering the basic interface of the device. And if we get better screens, it gets even worse.
 
I am considering getting an AVP but I know that M5 is rumored to be coming "some time in 2025." Personally, though, I have not heard of anyone complaining about CPU bottlenecking on the AVP. Would an M5 upgrade even be noticeable? Is it worth the wait?
I am waiting for vision OS to evolve and M5 can potentially give better battery life. I am also hoping Apple can bring down the cost. I am mostly focussed on high res screens when traveling, don’t really care about Apple intelligence.
 
I’d say limitations also (probably mainly) revolve around hardware.

The VP display quality is great compared to the competition, but unfortunately the PPI needed to simulate reality next to your eyes must be absurdly high. And there are other factors like fov. It’s much more convincing than what we had seen (maybe enough-ish), but there’s a lot of room for improvement.

Also, doing a detailed analysis of CPU needs would be too much for me, but I think it’s obviously a lot: the amount of pixels to handle, combined with the need to process data from multiple LiDARs with almost no acceptable lag, keeping windows perfectly anchored, etc. I bet Apple considered M1/M2+R1 as the minimum requirement, but we’re not even close to comfortable “overkill” levels (as we have with iPhones or Macs). We’re talking about covering the basic interface of the device. And if we get better screens, it gets even worse.
Except IMO the hardware presentation is already good. Sure it can improve, but no one is currently whining that PPI or FOV or latency is making AVP unworkable.
 
  • Like
Reactions: MakaniKai
Found this article. I can’t speak for its credibility.
But the part about the Neural Engine being the bottleneck sounds cromulent.



  • VisionOS 2 is not getting any Apple Intelligence features, despite the fact that the Vision Pro has an M2 chip. One reason is that VisionOS remains a dripping-wet new platform — Apple is still busy building the fundamentals, like rearranging and organizing apps in the Home view. VisionOS 2 isn’t even getting features like Math Notes, which, as I mentioned above, isn’t even under the Apple Intelligence umbrella. But another reason is that, according to well-informed little birdies, Vision Pro is already making significant use of the M2’s Neural Engine to supplement the R1 chip for real-time processing purposes — occlusion and object detection, things like that. With M-series-equipped Macs and iPads, the Neural Engine is basically sitting there, fully available for Apple Intelligence features. With the Vision Pro, it’s already being used.
 
Absolutely accurate. However the AVP's current limitations revolve around neither speed/ray tracing/CPU nor display quality. The AVP's current limitations are visionOS competence. Much like with iOS, improving the OS is the bottleneck.
In my experience, that’s insufficient. The hardware is also not up to the task, at least as my eyes see it. It’s a remarkable tech achievement, and the software definitely has the furthest to go, however M and R and screens all need to improve to get my (financial) vote of confidence.
 
  • Like
Reactions: MakaniKai
In my experience, that’s insufficient. The hardware is also not up to the task, at least as my eyes see it. It’s a remarkable tech achievement, and the software definitely has the furthest to go, however M and R and screens all need to improve to get my (financial) vote of confidence.
Interesting. For me the improvements needed are literally 100% on the software side. Of course software always requires hardware, but the M2/R1 do not seem to be bottlenecking things. The fact that AVP uses a base level M2 when Pro or Max were available suggests that CPU is not limiting.

That said, using a newer stronger chip in a v2 AVP will help allow software to do more. Having lots of CPU headspace available allows for easier, more comprehensive software engineering.
 
  • Like
Reactions: MakaniKai
I am considering getting an AVP but I know that M5 is rumored to be coming "some time in 2025." Personally, though, I have not heard of anyone complaining about CPU bottlenecking on the AVP. Would an M5 upgrade even be noticeable? Is it worth the wait?

AR/VR needs all the GPU power we can get at the moment. It's the GPU that will be more relevant on M5. Any upgrade to GPU until we get past 16k res per eye at 200-300 fps is going to be beneficial.

I'm not saying wait or anything like that. But M4, M5, M6, M7 are all going to be intermediate steps until we get to properly high quality AR/VR.
 
  • Like
Reactions: MakaniKai
AR/VR needs all the GPU power we can get at the moment. It's the GPU that will be more relevant on M5. Any upgrade to GPU until we get past 16k res per eye at 200-300 fps is going to be beneficial.

I'm not saying wait or anything like that. But M4, M5, M6, M7 are all going to be intermediate steps until we get to properly high quality AR/VR.
Without doubting what you are saying, why under that logic of GPU bottlenecking would Apple not have used an M2 Pro or M2 Max chip in the AVP? Apple could have more than tripled the GPU cores and bandwidth using M2 Max, and the $3500 price of AVP would support such a higher end chip.
 
  • Like
Reactions: MakaniKai
When it comes to an entirely new category of technological products, the third generation is often considered a milestone where the product becomes sufficiently advanced, stable, and user-friendly for a broad audience.

Take the iPhone as an example: The first iPhone was revolutionary, but it lacked many features we now take for granted, such as the App Store and video recording. The iPhone 3G introduced 3G connectivity and the App Store, but limitations remained, like the absence of video recording. The iPhone 3GS, however, offered better performance, video recording, and a stable user experience. By that point, most of the major shortcomings had been addressed.

What changes does AVP need to become its “3GS”? And I don’t mean just hardware, but features.
 
What changes does AVP need to become its “3GS”? And I don’t mean hardware, but features.
What AVP needs is for the software to work. Currently I cannot do an hour test without exposing broken software. Basic Apple apps not fully functional, let alone trying to run non-Apple apps.
 
  • Like
Reactions: MakaniKai
Without doubting what you are saying, why under that logic of GPU bottlenecking would Apple not have used an M2 Pro or M2 Max chip in the AVP? Apple could have more than tripled the GPU cores and bandwidth using M2 Max, and the $3500 price of AVP would support such a higher end chip.
Maybe the requirements for electric power and thermal dissipation?
If the headset becomes too hot, or if the fans are too noisy, or if the battery empties too fast, all this could prevent the use of the Pro or Max versions of the M2 chip.
 
I am waiting for vision OS to evolve and M5 can potentially give better battery life. I am also hoping Apple can bring down the cost. I am mostly focussed on high res screens when traveling, don’t really care about Apple intelligence.
I can survive with the short battery life, I think most of the time I will be sitting at my desk anyway where I can have it plugged in all the time. If I go outside to work I doubt I'd be out there for more than 2 hours. Even on a plane there's usually power outlets.

The cost definitely hurts but idk..I've always been a tech nerd and I really enjoy playing with brand new tech even when it's a little screwy. I think I would get over it.

I don't care about AI at all to be honest. It has been mildly useful on my iPhone with the notification summaries, but that is pretty much the only time I ever use it. And it has been woefully incorrect many times, recently. If the M5 realistically is only going to bring AI support, then I will probably just get the M2.

I guess my fear is that I will buy it and then just a month or two later they will release the M5 model for the same price. Even if it would make no difference to me, I'd still feel poopy
 
  • Like
Reactions: fatTribble
Without doubting what you are saying, why under that logic of GPU bottlenecking would Apple not have used an M2 Pro or M2 Max chip in the AVP? Apple could have more than tripled the GPU cores and bandwidth using M2 Max, and the $3500 price of AVP would support such a higher end chip.

Price? Power consumption? heat? battery life? Wearables aren't quite as simple as "stick the biggest baddest processor that will fit in it".

My bet is most likely heat and power consumption. The M2/m3/m4 max consume a lot of power and put out a lot of heat, relative to the space available to cool the thing without a fan in a head mounted device.

It's not that the device couldn't use the power - more that it would likely introduce a series of other problems; from what i understand the vision pro is pretty heavy on your face as it is; it would be worse with more cooling (increased weight) and even worse if it had a fan exhausting hot air near your head (comfort, fans small enough to fit are noisy), which will already get warm with something like this stuck to your face (based on my experiences with the quest2, quest 3, hololens and PSVR2).

It will likely get something similar power to m2 max when that is available in something smaller and more power efficient - or if the processing is offloaded to a part of the device that wirelessly connects to the headset so that hotter/heavier component is not carried on your head. But - that poses its own challenges; VR is very sensitive to latency and moving components like that out of the headset may introduce additional latency purely due to the wireless connectivity...


Depending what is going on, my 6900XT struggles with VR at times, and that's a 300+ watt GPU that will smoke my m4 max at GPU things.

We've got a very long way to go with AR just yet, which is why its clear to me that this is an early adopter device for software developers and curious people to play with - it's not a mass market device because the technology simply isn't available to get the level of visual fidelity in a mass market product yet. From any vendor. everybody is making different compromises and one of the compromises apple made was cost - they didn't build down to a price so much as say, Meta did with the quest series.

As to the cost of AVP being able to support the M2 max (discounting the heat/packaging concerns above)? Not so sure on that. The thing has expensive sensors, screens and lenses in it. It's not subsidized with user data like the quest series by meta is. Its a tier or two above what you'll get in any competing ar/vr headset and the price reflects that. The M (+R1) series SOC is a fraction of the actual BOM cost of that headset and i really doubt apple have a lot of margin on it. It's expensive because its expensive to make.
 
Last edited:
Without doubting what you are saying, why under that logic of GPU bottlenecking would Apple not have used an M2 Pro or M2 Max chip in the AVP? Apple could have more than tripled the GPU cores and bandwidth using M2 Max, and the $3500 price of AVP would support such a higher end chip.

M2 is a 30W chip. M2 Pro is 60W. That basically means adding a copper heatsink (weight) and much faster fans. The headset already weighs a ton.
 
I am considering getting an AVP but I know that M5 is rumored to be coming "some time in 2025." Personally, though, I have not heard of anyone complaining about CPU bottlenecking on the AVP. Would an M5 upgrade even be noticeable? Is it worth the wait?

The Vision Pro is one of the few cases where nobody would complain if they made it thinner/lighter.

As for the processor? Meh…
 
I guess my fear is that I will buy it and then just a month or two later they will release the M5 model for the same price.
I’m no expert, but I haven’t read anything about an upcoming Pro. I’m still in awe every time I use it. Just a couple of days ago I found a new game (Gears and Goo) that’s become a bit addictive. Yes you could do it on an iPad but seeing this huge game map in front of me in 3D is such a visual treat.

In the words of Marge Simpson, your purchase will “help the economy “ ☺️
 
Last edited:
Interesting. For me the improvements needed are literally 100% on the software side. Of course software always requires hardware, but the M2/R1 do not seem to be bottlenecking things. The fact that AVP uses a base level M2 when Pro or Max were available suggests that CPU is not limiting.

That said, using a newer stronger chip in a v2 AVP will help allow software to do more. Having lots of CPU headspace available allows for easier, more comprehensive software engineering.
Base M chips are more power efficient and the VisionPro depends on a tiny battery. This is very much a condition of YMWV: perhaps you see none of the things I’m annoyed by when I used it. That is, in fact, probable. I’m super picky with my screen tech (the only screens that have entered my life as approaching “good enough” are my M1 MBPro, my gaming OLED and my tandem OLED iPP and you’ll notice that the oldest there has just passed three years since introduction).

I don’t need the device to have better tech unless it directly improves the viewing experience. I expect to use the device near on 100% linked to a future Mac Studio rather than as a standalone device.

But the question was: what will an M5 bring to the VisionPro? And for myself, mesh shading and ray tracing are really big deals particularly as I intend to develop apps for the platform in future.
 
Price? Power consumption? heat? battery life? Wearables aren't quite as simple as "stick the biggest baddest processor that will fit in it".

My bet is most likely heat and power consumption. The M2/m3/m4 max consume a lot of power and put out a lot of heat, relative to the space available to cool the thing without a fan in a head mounted device.

It's not that the device couldn't use the power - more that it would likely introduce a series of other problems; from what i understand the vision pro is pretty heavy on your face as it is; it would be worse with more cooling (increased weight) and even worse if it had a fan exhausting hot air near your head (comfort, fans small enough to fit are noisy), which will already get warm with something like this stuck to your face (based on my experiences with the quest2, quest 3, hololens and PSVR2).

It will likely get something similar power to m2 max when that is available in something smaller and more power efficient - or if the processing is offloaded to a part of the device that wirelessly connects to the headset so that hotter/heavier component is not carried on your head. But - that poses its own challenges; VR is very sensitive to latency and moving components like that out of the headset may introduce additional latency purely due to the wireless connectivity...


Depending what is going on, my 6900XT struggles with VR at times, and that's a 300+ watt GPU that will smoke my m4 max at GPU things.

We've got a very long way to go with AR just yet, which is why its clear to me that this is an early adopter device for software developers and curious people to play with - it's not a mass market device because the technology simply isn't available to get the level of visual fidelity in a mass market product yet. From any vendor. everybody is making different compromises and one of the compromises apple made was cost - they didn't build down to a price so much as say, Meta did with the quest series.

As to the cost of AVP being able to support the M2 max (discounting the heat/packaging concerns above)? Not so sure on that. The thing has expensive sensors, screens and lenses in it. It's not subsidized with user data like the quest series by meta is. Its a tier or two above what you'll get in any competing ar/vr headset and the price reflects that. The M (+R1) series SOC is a fraction of the actual BOM cost of that headset and i really doubt apple have a lot of margin on it. It's expensive because its expensive to make.
I do disagree with "the technology simply isn't available to get the level of visual fidelity in a mass market product yet." IMO the visual fidelity of the AVP is competent now.

But I get the point that the high cost of achieving sales to whatever qualifies as "mass market" might be by definition limiting. However IMO reported sales of v1 with alpha-level software were adequate at the $3500 price.

None of the issues Price? Power consumption? heat? battery life [n/a to me]? are IMO currently particularly limiting, based on my admittedly limited three tests, the longest 90 minutes. The only reason I do not own AVP is the current beta-level of AVP software; I will buy one to do real work with, and I do not allow beta into my real work.
 
Last edited:
  • Like
Reactions: MakaniKai
I’m no expert, but I haven’t read anything about an upcoming Pro. I’m still in awe every time I use it. Just a couple of days ago I found a new game that’s become a bit addictive. Yes you could do it on an iPad but seeing this huge game map in front of me in 3D is such a visual treat.

In the words of Marge Simpson, your purchase will “help the economy “ ☺️
I’m curious, what game did you find “addictive“. I’m always on the lookout…
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.