Probably because mobile data standards evolve faster than the typical lifespan of a laptop. i.e. people upgrade their phones, and TelCos upgrade their networks more frequently than people upgrade their laptops.
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.
A14X Chip for First Apple Silicon Mac and New iPad Pro to Enter Mass Production in Fourth Quarter
- Thread starter MacRumors
- Start date
- Sort by reaction score
Probably because mobile data standards evolve faster than the typical lifespan of a laptop. i.e. people upgrade their phones, and TelCos upgrade their networks more frequently than people upgrade their laptops.
Or more realistically: You use a small handset that fits in your pocket quite a bit more than you would use your laptop. Plus every smartphone can create a mobile hotspot, so that's a cost factor to consider too. You would need an additional SIM card that costs you something every month, for the exact same functionality your smartphone already has.
What are you using your computer for? Interested to know what applications need so much compute power!
All good points! It's not a feature I've ever missed on my laptop, because I always have my phone with me. I have an iPad with 4G, but have *never* put a data SIM in it because I'm OK tethering my phone. For some people though (my wife included), they prefer to have a completely independent data connection because they often use the iPad without having a phone nearby.
depends on what capable as a computer means, like slow lower end netbook? Or like Alien Gaming Desktop?
I will have to see it running 3D Max and Cyberpunk 2077 4K to believe it. They are moving to ARM yes, but I thought its specially designed CPU for desktop use...not actually an iPad chip.
That Reddit leak mentions an "M14" chip (from this family) on a new 14" Macbook Pro next week. That seems like a reasonable naming convention.
Apple is probably trying to ship AS macs soon as they can because it must be getting harder to move Intel macs with the performance/efficiency gains apple is promising in the wings...
Apple is probably trying to ship AS macs soon as they can because it must be getting harder to move Intel macs with the performance/efficiency gains apple is promising in the wings...
If anything, Steve would've wanted a Mac Pro even less. It was under him that the original Macs didn't have any system expansion whatsoever; Gassée's Macintosh II with internal cards happened long after Steve was gone.
And the iMac does go up to an i9-10910, which… close enough?
If you see some of my other posts you’ll see my doubts about Apple Silicon. On phones and non Pro iPads (I don’t own a pro so I can’t say) the apps run well but not as good as my Mac version of that software. I’m not against Apple developing their own chips-they’ve done a good job with phones and iPads but they aren’t as fast or capable as a mid to upper end laptop or iMac. The app version of software usually has features disabled or missing from the full computer version.
I waited a couple of years to avoid the PowerPC to Intel cross so I never went though needing a program to make old software run on my iMac. But this time I do have a software investment in the “old” hardware. And I know how Apple doesn’t like to do more than “hint” at things literally up until it’s officially announced.
I had plans to upgrade my desktop in the next year or so, and those are now on hold. I probably won’t buy any major software for my desktop either if I can avoid it.
Maybe all of the benchmarks are correct and things will run fine and developers will all be eager to do whatever they need to for recompiling for the new processors. Maybe.
Data standards change about once a decade, no? people don’t keep their laptops that long. And when they do, it’s not as if the old data standard stops working right away.
The “reddit leak” is completely false, though. Everyone who published the leaks he says he is responsible for says he was not the guy who leaked anything to them.
Well, nobody outside of Apple knows if he's speaking the truth, but we'll find out in a few days.
Last edited:
Primarily because Intel’s forced to make processors the way they do because that’s what their market wants. There’s a big sized portion on every Intel processor that primarily takes the variable sized instructions it receives and slice those into chunks that the processor actually uses to do work. If Intel could tell the entire industry (including any competitors like AMD) that everyone’s going to one instruction size, they could remove that chunk and the complexity that goes along with it. (As an aside, when Apple went to pure 64-bit, that opened up processor die space for more cores of various types.) Apple’s job to create a processor was easier, in a way, because they know the detailed specs of every current system that will use the chip and has a say into any future hardware/software combination that could possibly use the chip. As a result, they’ll always be more efficient than Intel or even AMD will ever be at executing macOS workloads.
I’m sure some want to spend a little on more cores for the iPad Pro. It’s certainly potentially leaving money on the table. However, there won’t be a “better” on Intel’s ARK database to compare against. You get the fastest Mac (that will be faster than the prior year’s) and that’s it. I could see how binning might provide some performance increases around the clock speed (rather than number of cores) and they could offer those to folks that REALLY want to know they have the “best”. But, it makes sense to me that Apple would reset expectations in this area.We’ll see in a few weeks if they plan to replicate the complexity of an Intel lineup OR the simplicity of the iOS processor lineup or something in between
What would be interesting would be if we finally see a Mini Pro primarily because Apple can now actually bring a more efficient Mac Pro class CPU into the smaller form factor.
I think it’s likely a smaller group of users altogether, but perhaps one that Apple could see a benefit in supporting going forward?
Programming, 3D modelling, Photography(batch processing scales well with cores and memory), video editing(8k now)
And to complicate things even further Apple has its own team spreading rumours........
Yeah, absolutely pointless to upgrade to the i5 or i7, unless you loooove that fan sound.
I just checked, one person "liked" what I wrote, and one person "downvoted" it, ha ha Apple is a deity for some people, and I think I just blasphemed!
Big tech companies are the new "Gods" and Silicon Valley is the new Mount Olympus.
Considering the recent trend on Intel processors has been fairly modest year-on-year improvements < 10%, I think Apple will meet expectations with a 30-50% performance improvement and >50% battery life improvement on laptops.
I'm expecting "considerably better", but not 200-300%, which hasn't been seen very often in IT hardware developement (maybe PowerPC -> Intel transition?)
That is firmly in the professional and not consumer market segment and not where Apple makes most of its money.
The first Apple desktop ASi Mac needs to compete against existing iMacs, not the Mac Pro. (Incidentally, the new iMac 27 8 & 10 core models are already very competitive against the 8 to 12-core Mac Pros, with the latter representing quite poor bang-for-your-buck).
All Apple need to do is to make an ASi iMac at least 20% faster than the current 8-core iMac, and ensure that graphics performance matches the Radeon 5500M at least, and they will have met expectations for the market segment. I hope it is better than this, but this would be sufficient. It would probably require 12 ARM-performance cores to match the 8-core Intel 10th gen.
And again, what are you doing that requires 64-cores on a desktop?? Pretty sure there are no consumer applications that require or even benefit from this size of computer. Professional ones....sure...but I bet it's a very small number of them.
There are few applications that really need 64 cores. Video editing tends to level off after 24 cores and there is little benefit in more cores (see https://www.pugetsystems.com/labs/a...formance-AMD-Threadripper-3990X-64-Core-1659/). Even multiple GPUs don't add benefit beyond 2 or 3, depending on the application.
In servers, especially those running VMs or containers, sure 64 cores or more is common. But these are running dozens or hundreds of separate instances for multiple users / applications.
I'm really interested to know which single application needs or greatly benefits from really high core-counts.
Running lots of VMs is a reasonable use case, but I'd have to question whether the most cost-effective way to do it is to have a super-expensive workstation computer. You may find that cloud platforms such as AWS, GCP or Azure are better options - both in terms of cost and the management effort required to run and configure them on your own hardware. It depends on your use case. Are you spinning VMs and containers up and down for development, or running them 24x7?
In my industry (enterprise cloud computing), I have seen a rapid decline in the number of clients using VMs on their own hardware, and most of those are ESXi servers in their data center, not workstations. I used to use a lot of VMWare or VirtualBox VMs for test environments, but now it's easier, less time-consuming, and often cheaper to just spin them up in the cloud.
Video editing / rendering has diminishing returns over about 24 cores (Premiere Pro CPU performance: AMD Threadripper 3990X 64 Core) - with the 64 core Threadripper running slower than the 32-core version.
Photoshop only shows minimal improvements with >8 cores (https://www.pugetsystems.com/recomm...-Adobe-Photoshop-139/Hardware-Recommendations)
You could run thousands of Docker containers on a 64 core machine with sufficient RAM, so this is overkill to run 50-100 containers (which is a relatively complex micro-service architecture).
You might argue that need to do all of the above *at the same time*....but this is definitely an edge use case.
By all means, spend your money on £4000 CPUs....you probably aren't getting your money's worth though
Apple certainly need to make the launch of Apple Silicon a success by demonstrating that it has obvious and significant advantages in user experience (performance, power consumption, software options) over Intel.
I don't think they need to completely crush Intel performance to do this though. A 20-30% improvement should be sufficient.
More is better of course - but this may be harder to achieve that we think, particularly for their GPUs which are largely unproven against desktop dGPU alternatives.
ASi will need more cores than current Intel designs to match or exceed Intel, but I doubt we will see 30 core iMacs. Mac Pros - yes - I would expect 32-64 power cores.
As an AWS user, I'd say those costs add up quickly. So, easy yes, cost effective, I'd be skeptical. It's also introduced a new kind of anxiety of forgetting to ramp down hardware when unused. YMMV.
Besides, if Apple Silicon is so great - it should have compute to spare, no?