Then I'll just say wait and see what happens.
Apple has proven that ARM is the future. Microsoft and the PC industry will be trying their best to imitate this and x64 will likely be dead in 7 years. That's what I think at least.
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.
M1 as an "Industry Game Changer"
- Thread starter Yebubbleman
- Start date
- Sort by reaction score
Apple has proven that ARM is the future. Microsoft and the PC industry will be trying their best to imitate this and x64 will likely be dead in 7 years. That's what I think at least.
Performance per watt is the advantage. We already had 6W CPUs, they were slow.
Not just the thermal paste, the amount of dust accretion after a few years is insane. I know from cleaning it out. With a dust mask on.
x86 I presume?
If software support for ARM spreads to Windows machines, we may see the demise of Wintel anx x86, but I think it'd take a couple of decades. Old devices are getting better at clinging to usability.
Not sure why "investments" is in quotes like that. That's likely how TSMC managed to get to 5nm relatively smoothly, if industry rumours about Apple's involvement in TSMC's 7nm process are true.
True, but in the case of OpenGL it was depreciated a long time ago and the axe never came. I'm quite sure it will stay there for a while, if not indefinitely — Apple has nothing to lose by continuing to ship their wrapper. It's not like depreciation of 32-bit software or some legacy UI frameworks which had some real support cost and which messed up the software infrastructure.
What I am trying to say is complaining that "Apple disallows users to run OpenGL apps" is a bit silly, because those apps still run fine.
I've been seeing this sentiment quite a while, and I can't say that I understand it. I now have an ultracompact laptop that's faster at running stats and compiling code than some high-end desktops. For many people, performance matters, and Apple silicon means that they can move to more convenient form factor without sacrificing performance or keep the same form factor for vastly improved performance.
Once consumer gets used to powerful and yet long battery life of the MacBooks in a thin and light form factor, it'll create a lot of pressure for other notebooks manufacturers to follow suit. This in itself will shake up the industry.
You may have missed that even ARM A series cores are advancing with roughly 20%+ performance gains per year - very similar to Apple Cores. The progress is easily outperforming both Intel and AMD. For example this year the 30% IPC gains of the Cortex X1 compared to Cortex A77 are outperforming A13->A14 IPC gains. Apple still gained roughly 20% but only via increasing the frequency.
I do not think an architecture license is required, as the standard ARM cores are progressing so nicely.
It is true, but I don't think the situation is as simple as you make it sound. Cortex X1 represents a big jump in performance simply because it is the first attempt by ARM to design CPU IP focused on performance rather than on die area and low power consumption. It is far from certain they will be able to keep this momentum going.
I think it's quite interesting that maximal performance of all the existing chips seems to be fairly similar. Intel, AMD, Apple — their peak single-threaded performance is more or less comparable despite different approaches they take. Ok, Intel and AMD use similar design philosophy, but Apple is quite different with their low clock and wide core.This makes me wonder whether Apple is purposely holding their CPUs back to fit a more reasonable power profile, or whether there is indeed some sort of "industry performance cap" that is common to all sufficiently sophisticated CPU designs.
The smaller form factor is a consequence of lower power usage. Similar performance was already available in the same price range for those willing to use a larger laptop.
Same price range? Not so much if we are looking at the premium segment. The $999 Air is faster in many tasks than $2000+ laptops (especially if we stay with Macs). Besides, let's not forget that higher-end Apple Silicon Macs are coming. You will get significantly improved performance and efficiency at every tier.
That is exactly the message Apple has been pushing: power-efficiency = performance.
Definitely true. It will be extremely interesting to see how AMD's 86x performs against Apple's ARM, using the same node, from the same TSMC in the future.
Right now, I think the best comparison we can make is the M1 vs the 4800u, both 15W:
The "premium segment" is an artificial restriction, because it implies a thin case with a low cooling capacity.
For cost-effective performance, we should be looking at gaming laptops and similar hardware. For $1000, we should then get something like a Ryzen 7 4800H and a Radeon RX 5600M. That would yield lower single-core performance, similar multi-core performance, and higher GPU performance than in an M1 Mac.
Future higher-end Macs will be faster, but they will also be much more expensive. In that price category, I expect that Apple Silicon laptops will have faster CPUs but much slower GPUs than non-Apple laptops.
I maybe wrong but it seems Apple has addressed specialized instructions to make their SOC perform better:
A) under macOS (storing/releasing objects)
B) with x86 instructions (identifying when x86 code is running and altering, I guess, out of order rules)
So it’s not just the ARM isa in their secret sauce.
Microsoft IS ALREADY serious about Arm. They have ported Windows to Arm, not just the desktop version that we talk about but the server versions. They build their own servers for Azure and many of them are Arm running Windows server for Arm. They have undertaken the large task of porting Office to Arm, not just for Apple silicon but for Windows as well. They have hired a large number of CPU designers to produce their own cpus. Purely for Arm.
Here’s a thought- why does someone else have to develop arm desktops for it to get Traction. Apple silicon runs Windows fantastically in virtualization, and Microsoft can ( and I believe will) offer a Version that boots natively in Apple silicon.
These machines are fantastic and are only the tip of the iceberg of what Apple will produce over the next year.
Yes Apple silicon has changed the industry. Two Tuesdays ago will be noted as the beginning of the end of X86.
Last edited:
Yes. But that chart is from a few weeks ago. Which is a long time ago in this space now. The new version adds both the Amd processors and the M1. The 8 and 16 core threadrippers are the only processors that beat the M1. And they better enjoy their position there. Because the very next M chip will leave them behind. They will never be able to catch up.
openGL is supported on Apple Silicon (though its deprecated) : https://developer.apple.com/documentation/xcode/porting_your_macos_apps_to_apple_silicon
For now its an Apple gamechanger that will impact its ecosystem - particularly future software development for the Apple ecosystem since mobile, desktop, laptop, and ipad software is written for the same architecture.
If Apple proves to be successful with this going into the future - others will follow with their own approach to using ARM (RISC) with a SOC approach. The impact will be that resources will flow into ARM development and pull resources away from x86 development. One could envision Intel and AMD migrating their focus to this.
If Apple proves to be successful with this going into the future - others will follow with their own approach to using ARM (RISC) with a SOC approach. The impact will be that resources will flow into ARM development and pull resources away from x86 development. One could envision Intel and AMD migrating their focus to this.
Windows is architecture independent too. It uses a hardware abstraction layer, and has at one point or another been compiled for x32, x64, ARM, RISC, and PowerPC, in addition to the x86-64 version most commonly used on AMD and Intel hardware today.
Any Windows 10 apps that only use native libraries, can be compiled for ARM with the press of a button. Windows isn't married to x86-64 at all.
The Secret sauce is that these two things go together. Every device has a maximum thermal envelope. And at every given level, AS offers more performance at a lower wattage. Wattage leads to heat. Less wattage =less heat. So if a cpu is more efficient at power per watt then you can have more performance in a given thermal envelope. I saw a test running Cineabench 23. The Apple silicon system completed the test , never drawing more than 13 watts. The 2020 iMac completed it drawing 125 watts. The AS system stayed cool and it’s fan never left idle. The iMac ramps up its fans to max in less than 30 seconds. When Apple takes the large thermal envelope of the iMac and puts a more powerful M series processor in it, they will be able to bring much more performance than in today’s iMac because more performance is possible because the AS brings more performance per watt.
They'll also enable quieter computers in the same thermal envelope. Not a big deal for me, but vital for someone who makes music or edits video.
Exactly my point. And focusing on one side of the medal while disregarding the other one is throwing out half the story.
This degree of power in a tiny, fanless computer is an absolute stepchange to what was possible using x86 chips. The Intel 12” MacBook, the only other fanless Mac, was weak on CPU performance and especially weak on GPU performance. The AS MacBook Air trades blows with a 16” MacBook Pro with discrete graphics (albeit quite entry-level discrete graphics).
Does that matter for the wider industry? At the moment, I don’t think the pieces are there on the Windows side. But this might be a catalyst to start getting them in place. It’s worth noting neither Intel or AMD have any answer to the M1 for fanless ultrabook class devices (nor does Qualcomm for the moment). So Apple has the opportunity to expand and take the cream off this lucrative market segment, which might be a worry for other OEMs where the overall PC market is stagnant. Unless Intel or AMD can pull off a quick comeback, we might see redoubled interest from Microsoft and their OEMs for using Arm in this market segment, eventually forcing x86 out if it is successful. I suppose it hinges on whether Intel/AMD or Qualcomm/Microsoft/any consortium of OEMs are first to, or more successful at mounting a challenge here.
Does that matter for the wider industry? At the moment, I don’t think the pieces are there on the Windows side. But this might be a catalyst to start getting them in place. It’s worth noting neither Intel or AMD have any answer to the M1 for fanless ultrabook class devices (nor does Qualcomm for the moment). So Apple has the opportunity to expand and take the cream off this lucrative market segment, which might be a worry for other OEMs where the overall PC market is stagnant. Unless Intel or AMD can pull off a quick comeback, we might see redoubled interest from Microsoft and their OEMs for using Arm in this market segment, eventually forcing x86 out if it is successful. I suppose it hinges on whether Intel/AMD or Qualcomm/Microsoft/any consortium of OEMs are first to, or more successful at mounting a challenge here.