Agreed! It’s just, as a group of edge cases, this one’s a bit edgier.
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.
Mac Pro Enthusiasts Raise Concerns Over Upgrade Limitations of Apple Silicon
- Thread starter MacRumors
- Start date
- Sort by reaction score
People who buy the Mac pro for upgradability are missing the point. It's all about the wheels!
But seriously, what is it about Apple silicon that makes external RAM and GPU impossible? There are great GPUs out there, is Apple just assuming that their customers would never have a need for them? It's kind of dumb.
But seriously, what is it about Apple silicon that makes external RAM and GPU impossible? There are great GPUs out there, is Apple just assuming that their customers would never have a need for them? It's kind of dumb.
Last edited:
Nvidia will always be better at Nvidia’s proprietary workflows than anything else in the world. No one should ever think that anything Apple can do would change that.
The other problem is that the M2 Ultra is still substantially behind even mid range intel desktop parts and even many of the alder lake/raptor lake mobile parts. When it comes to single threaded performance the intel chips are over 50 percent faster and in multithreaded performance they are twice as fast. And this is with only mid range parts. (I5's and I7s) It would get even worse if you started comparing it to the upcoming workstation parts. The only reason why the M series chips compare so well to the old Intel Mac Pro is it is still using a Skylake era cpu architecture which was outdated even when the mac pro came out.
It gets even worse when you start involving dedicated graphics cards. The M1/M2 Ultra can't even compete with the RTX 3060 let alone the higher end stuff which are multiple times faster than that.
The reality is that Apple really should keep intel around for the desktops.
It gets even worse when you start involving dedicated graphics cards. The M1/M2 Ultra can't even compete with the RTX 3060 let alone the higher end stuff which are multiple times faster than that.
The reality is that Apple really should keep intel around for the desktops.
Last edited:
That all seems highly dependent on which benchmarks you're looking at? In GeekBench land, Intel has a small lead in performance, but requires way more power.
The kicker is, Apple chip development is mobile first, so Apple's "perforamnce per watt" is king.
Intel is far behind in process node, but luckily they have really good engineers and have mitigated that deficiency with tons of electricity and heat. Apple's problem is: at the pro desktop space, nobody cares about performance per watt.
Had Apple been able to ship a (Max * 4) chip it wouldn't look so bad; but right now, Apple's high end looks weak.
Time will tell if Apple can ship the high-end Macs that they were rumoured to be planning on.
Why is everyone suprised? The Apple M series was, has and is NOT a real computer. This has nothing to do with x86 vs ARM as Alterra has no issue running the GPU compute nodes at work.
The Apple M series is a glorified tablet. It lacks any meaningful peripheral interfaces or bus and has minimal PCIE expansion capacity. I had hoped that the move to M2 would have changed this but apparently not. The amount of work that would be required to do this would put a system like the one needed for the Mac Pro years away; not 3.
For those defending the unified memory mess, why hasn’t Apple chosen to use HBM? That is the logical conclusion is it not? Plus having a dual strata memory structure is quite possible just look at Sapphire Rapids.
Apple has completely lost the plot and unless something is done before Intel get their tile based packaging corrected with Meteorlake it’s going to get much much worse.
I don’t understand why people put so much faith in Apple reinventing the wheel as if AMD and Intel engineers weren’t capable of better. Minimal engineering was put into the M1 and M2 design wise; what were people expecting?
The fact that the M2 which has full USB4 support and thunderbolt lacks EGPU capability is beyond me. Oh wait? That’s because it is NOT a real computer…
The Apple M series is a glorified tablet. It lacks any meaningful peripheral interfaces or bus and has minimal PCIE expansion capacity. I had hoped that the move to M2 would have changed this but apparently not. The amount of work that would be required to do this would put a system like the one needed for the Mac Pro years away; not 3.
For those defending the unified memory mess, why hasn’t Apple chosen to use HBM? That is the logical conclusion is it not? Plus having a dual strata memory structure is quite possible just look at Sapphire Rapids.
Apple has completely lost the plot and unless something is done before Intel get their tile based packaging corrected with Meteorlake it’s going to get much much worse.
I don’t understand why people put so much faith in Apple reinventing the wheel as if AMD and Intel engineers weren’t capable of better. Minimal engineering was put into the M1 and M2 design wise; what were people expecting?
The fact that the M2 which has full USB4 support and thunderbolt lacks EGPU capability is beyond me. Oh wait? That’s because it is NOT a real computer…
Last edited:
Far behind how? Transistor size or density because TSMC can make em small just not well or nearly as compact as Intel can. The fact that Intel 4 on 10nm produces the same die space as TSMC on 5nm is pitiful.
These people aren’t miracle workers and TSMC isn’t god
Aren’t the wheels an upgrade, though?
The wheels are available in six colors:
- PRODUCT(RED)
- Starlight
- Midnight
- Blue
- Pink
- Green
Intel’s bigger problem is that they have no path to the ultraportables that look to make up a significant portion of the market going forward. If you look at the number of different processor lines being produced in a given year by Intel, while Apple’s not beating the top end, they are outperforming a larger percentage of them every year.
Given the way things are going, I can foresee a time where Apple’s processors are running circles around the majority of what Intel has on the market at that time. By some single core metrics, that’s true now as even the slowest FIRST M1 is in the top 60, out of a list of over 230. If we only count the M2’s, even the POOREST performing M2 is in the top 30.
No. This simply can never happen. The entire reason the M1/M2 is so fast is because of unified memory. The old Intel computers used different memory in the CPU and GPU and then connected it with a PCIe bus. They moved data. the software had to push the data over the PCIe bus. So the number of bus lanes and the speed of each lane was a big deal because it was a bottleneck.
But with unified memory, the data NEVER MOVES. It is not that the bus is faster, there is no bus.
As soon as you have expandable RAM then you need some kind of bus to place it on. this ruins the entire advantage of unified RAM. Now all the data ging to and from you add-on memory has to go over a bus that connects the sockets to the main memory.
With the "M" series Apple removes a huge bottleneck which is the need to move data. They simply can not give his up without going back to having different kinds of memory
And why, really, do you need to add memory later? Just buy it up front and be done with it. The Mac Pro is for running some specific app that you need, and the cost of the Mac Pro is likely trivial compared to the cost of the work being done. No one is caring about saving $200 by adding RAM later.
Because PCIe cards, and the power supply to run them, can generate lots more heat, and Apple has probably designed this chassis for the M* Quadra which might make about twice as heat as the M* Ultra?
Not sure that's true, but in any case it will probably end as Intel is catching up to TSMC on the process technology. End of this year they will release the first mobile CPUs ("Meteor Lake") using the Intel 4 process (for the CPU tile) and TSMC N3 or N5 (for the GPU tile). They also have a CPU on their roadmap that is supposedly redesigned from the ground up for ultra mobile platforms using their 18A process ("Lunar Lake"). From their earnings call this week it sounds like it is all on track.
Yes, I want to know the technical reason. So far it seemed to me that a computer was made of a CPU, a GPU and RAM, among other things, and that the RAM and GPU could be swapped if they weren't soldered in.
I get that AMD, Intel, and Apple make different types of CPUs. But both AMD and Intel allow you to have whatever RAM and GPU you like. Why not Apple? I suspect it's just "we want to force you to pay 5x price for RAM and we want you to get the M2 Pro Max Ultra just to get an extra GPU core".
But I think it's a huge, huge shame that there are companies like nVidia who make great GPUs, they innovate in the AI revolution, there are amazing games that make use of raytracing and all kinds of cutting edge technology, there's VR... and Mac users don't even know what the hell any of that is. Apple is like the North Korea of tech. "Our tech is perfectly good, you're not allowed to try anything else."
Just a few years ago we had nVidia GPUs in MacBook Pros. Then for some odd reason it was limited to AMD GPUs. But you could still get external GPUs, like the BlackMagic eGPUs via Thunderbolt. It seemed like things were expanding, that more options were becoming available... and now all Mac users are stuck with integrated graphics made by Apple regardless of how much they spend on their computer, a company that has no experience making GPUs and has not brought gaming, AI or VR to the Mac in any form. They're just using their new CPU as an excuse to lock down their already massively crippled ecosystem even more. Now the entire computer is just a single, proprietary, closed source, locked down chip that can only run software made by a single company and can only work with hardware made by that company, too. You can put it in an iPad, an iMac, a MacBook Pro or Air, or a Mac Pro tower, and it's the exact same thing. The only difference that you're paying for is the size of the fan in the machine. That's it.
Need more RAM? Buy a new computer.
Need another GPU? Buy a new computer (no actually there are no GPU options).
Need to upgrade the CPU? Buy a new computer.
You'd think you're paying extra so that you get to keep using the machine for longer, because it will be more future-proof, more durable, repairable, upgradeable... but it's the exact opposite. It's just more expensive and you'll have to buy a new one sooner.
Last edited:
Just check any list that shows ALL the processors that benchmarks have been run against. You’ll ALWAYS see a long list from as high as high dollar, high performance workstations, all the way down to “grandad’s small machine”. Draw a line across the middle of that chart and ALL of Apple’s Silicon processors will be in the upper half. Most of what Intel sells to customers is outclassed by Apple’s M2’s. And, will be outclassed by Apple’s M3’s because neither Intel nor AMD are EVER going to stop making low performing processors for cheap systems. They both have the option of making mobile processors that have VERY similar single core scores to their high end, Apple’s shown it’s possible. They just won’t do it.
Everything is ALWAYS on track in Intel’s earning calls. Right up until they start sliding the dates and increasing the TDP’s.
For folks that need to run Windows, Intel can miss their goals and come out with a processor that gets 3 hours of battery with an impressive fan, it’s not like those folks are going to go out and buy Macs. Intel doesn’t HAVE to beat Apple in performance per watt for the same reason Apple doesn’t have to beat Intel in peak performance. The OS one wants defines the architecture they use these days.
But TSMC was on track to be at N3 large scale production by now, then Covid hit.
You won’t find anything but praise from me on the M2 line.
Apple has a cadence when entering a new market. You have your breakthrough release, and then iterate rapidly while the competition is still trying to figure out how to catch up to the 1st gen.
When it comes to Apple silicon I think we’re going to see Apple pull ahead in areas that matter to them, and part of that initial push relied on N3. I think when M3 is announced we’re going to see performance that isn’t entirely due to the N3 process, but wasn’t feasible without it. I think N5 doesn’t give Apple the transistor budget they need to include things that are waiting in the wings.
Hopefully that clarifies my thoughts on AS? I think most tech oriented people still don’t understand how big of a game changer it is, because we haven’t seen the true features going it alone on chip design are going to bring quite yet (and what we’ve already seen has been astonishing from a perf per watt perspective).
Well, to begin with, having user-upgradeable memory makes the "just buy it up front" part easier for customers. It has almost always been the case that even when Apple was using "commodity" RAM in their machines, it was generally cheaper for someone to buy an Apple machine, throw out the included RAM, and then install aftermarket RAM in its place in order to get a usable spec. I made a pretty good living in the early to mid 00's doing exactly this while working at an Apple specialty store. Even with an insanely high mark-up (think 50%+) after-market RAM was both cheaper and generally better quality (and backed by a better warranty) than Apple's own RAM.
The other major reason is simple longevity. In many cases if someone's computing needs change, the only thing they really need to upgrade to get more years out of their machine is RAM and storage. Sometimes GPU as well (but even that only applies to certain users).
The only thing that shows is that Apple does not serve the low cost and embeded markets. And your claim that they were "outperforming a larger percentage of them every year" seems to imply that Apple was somehow pulling away. But I don't think that's the case at all.
Huh? Apple doesn't make any low end processors. As a result, you can buy x86 laptops for less than half the price of the cheapest Macbook. By all estimates the M-series CPUs are very expensive since they use TSMC's cutting edge manufacturing and packaging technologies and are made in relatively small numbers.
There is thriving competition between manufacturers in the PC market, as well as between Intel and AMD. We currently see in their financial results what happens if they let themselves fall behind.Everything is ALWAYS on track in Intel’s earning calls. Right up until they start sliding the dates and increasing the TDP’s.
I think that's exactly what they are aiming to do with products like "Lunar Lake".
Apple doesn’t play in the high volume, low margins game. That’s always been their business strategy. I don’t view it as a critique because why should they?
Year after year they gobble up most of the profits in the industry, and NEW Mac users (people who didn’t own a Mac) are still the majority of buyers every quarter.
I’ve got my issues with Capitalism, but by its own definitions I don’t understand how Apple can be considered anything but a massive ongoing success story in this industry.
You can not like it, and I certainly have my gripes, but that doesn’t change that this strategy is clearly working.
I have a hard time believing some of this. I just doesn't make any sense. They did an about-face on the trashcan, openly apologized, and even kinda told people "we're working on a replacement" which was uncharacteristic. Why go through all of that to put out a single generation product and then totally redesign it with all the problems you just corrected?
I don't care that much because I've never owned nor cared about "workstation" class machines. But seriously, it would be weird if Apple does what they did with the MacBook Pros—cave to legitimate and whiner concerns—then return to the very thing that lead to all the backlash with its most expensive machines.
I don't care that much because I've never owned nor cared about "workstation" class machines. But seriously, it would be weird if Apple does what they did with the MacBook Pros—cave to legitimate and whiner concerns—then return to the very thing that lead to all the backlash with its most expensive machines.
Let ChatGPT enlighten us on Apple... Wait, maybe it's trying to tell us the problem isn't Apple, it's us!
Attachments
It wasn't a "critique" of Apple's business strategy, but of the previous poster's weird statements about the number of CPU models that Intel makes. Obviously Intel is a supplier that wants to cover all bases, while Apple is a maker of premium consumer products, and the M-series CPU is only made for a very specific niche.
Saying businesses do or don't upgrade memory after the fact is an over-generalization of what "businesses do" in both regards. I work with many businesses, and there has always been a mix of both. It all comes down to the purpose of the machines and how quickly their computing needs expand.
Focusing on this topic, however, the most common machines to be upgraded are workstation-class machines. Nearly all of the manufacturing-focused clients I have buy these for CAD and engineering applications, and they very commonly upgrade/reconfigure the machines' specs during those machines' lives.
I am hoping it will be something like.
Existing case
Motherboard with swappable compute card M Chip.
PCIE support with swappable GPU
Dual level Ram: on SOC and on motherboard.
We can all dream right?
Existing case
Motherboard with swappable compute card M Chip.
PCIE support with swappable GPU
Dual level Ram: on SOC and on motherboard.
We can all dream right?