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.
Qualcomm revealed X Elite's benchmark scores
- Thread starter sunny5
- Start date
- Sort by reaction score
They won't change anytime soon for sure. It's also harder changing from your own proud product to someone else's - especially the someone else who powers your competitors (Android and Windows).
If Qualcomm somehow became 3x better, then maybe Apple would think about it. But that isn't going to happen. More likely Apple just keeps growing and innovating and pushing the marker further ahead, which is good for all.
This is good news for ARM adoption in the computer world and for pushing all parties to make faster and more efficient chips.
This is just a nothing article.
Software is really important. Doesn't matter how good the hardware is if the software is not properly optimized.
So we have a great QC chip but we can only run Windows on arm.
Who wants that? Sure you have more native apps but unless Microsoft did what Apple has done and just ends support for x86 and switches full OS support to arm, gets all third party software vendors on board and kills off Nvidia and AMD and Intel for integrated SOC then who will really use these?
Since none of what I just said will happen any time soon then arm chips just are an outlier. Sure arm might be more efficient but AMD and Intel are already good enough and laptops get 6-10 hours generally. If Arm based laptops double that it will be a good incentive but and performance is good it will be a big incentive but the software restrictions and lack of dedicated GPU for gamers will be a problem. Also gamers are not going to buy an Arm chip.
Arm on Windows has enormous potential or even risc-v but without a cohesive and comprehensive roadmap forward I don't see it happening. Simply because Microsoft is too entrenched in legacy and x86 products they have a real barrier to adopting another architecture because of it.
One of the biggest benefits of Windows for many people is backwards compatibility and the amount of software available for it and arm will destroy both of those benefits for end users which is I think the biggest block to adoption.
It is frustrating too because as a Windows user I would love to have the complete Windows experience on a arm or risc-v based chip.
Software is really important. Doesn't matter how good the hardware is if the software is not properly optimized.
So we have a great QC chip but we can only run Windows on arm.
Who wants that? Sure you have more native apps but unless Microsoft did what Apple has done and just ends support for x86 and switches full OS support to arm, gets all third party software vendors on board and kills off Nvidia and AMD and Intel for integrated SOC then who will really use these?
Since none of what I just said will happen any time soon then arm chips just are an outlier. Sure arm might be more efficient but AMD and Intel are already good enough and laptops get 6-10 hours generally. If Arm based laptops double that it will be a good incentive but and performance is good it will be a big incentive but the software restrictions and lack of dedicated GPU for gamers will be a problem. Also gamers are not going to buy an Arm chip.
Arm on Windows has enormous potential or even risc-v but without a cohesive and comprehensive roadmap forward I don't see it happening. Simply because Microsoft is too entrenched in legacy and x86 products they have a real barrier to adopting another architecture because of it.
One of the biggest benefits of Windows for many people is backwards compatibility and the amount of software available for it and arm will destroy both of those benefits for end users which is I think the biggest block to adoption.
It is frustrating too because as a Windows user I would love to have the complete Windows experience on a arm or risc-v based chip.
PowerPC was Apple's own proud product (along with IBM and Mot), and they left it for Intel, who powered their competitors.
My point was simply that this is a no-lose situation for Apple. As you say, Apple will continue to develop at a rapid pace-- and they've proven themselves to be among the best in the world at it. If they decide it's not worth it anymore, they'll change vendors. It doesn't make sense to ask if AS is in danger-- AS isn't a product, it's a very, very expensive and resource intensive means to an end.
It's weird how many people think Windows is just never going to change...
HW/SW is a bit of a chicken and egg problem. Qualcomm, Nvidia and allegedly AMD are working on Arm devices for PCs. Microsoft will follow. I'd be surprised if these companies were putting all this effort in without having been in discussions with Microsoft already. There's no upside to MS announcing anything before a product ships though.
It might take a generation or two to work out the kinks, but Apple and now others are demonstrating why it's worth it.
And why wouldn't a gamer buy an Arm? I mean they do already, in the hundreds of millions in mobile devices. Nvidia chose it for their shield. Consoles regularly use non x86 processors. I'm not sure why they'd suddenly care...
You just said it. There is no upside until all of Microsoft vendors are ready.
So this chip is nothing sauce sprinkled with hype. I have no doubt that arm and risc-v will be what powers everything at a certain point but with Intel and AMD showing road maps several years in advance with x86 architecture. It takes years to r&d and then start producing and they don't even have a product or road map with any arm chips. So no matter what you say we are many years for full adoption of arm on Windows.
Microsoft doesn't have the control that Apple does and it has enterprise clients using old hardware and paying to keep it going and I doubt Microsoft is going to risk alienating their best customers and the entire customer based who all own primarily x86 based products. How many OEM's produce Windows based devices? You need to coordinate with them and the processor vendors and although I think it will happen it is a long ways off.
I will tell you why a gamer won't buy an arm PC because no games will run on it and certainly not with the frame rates gamers want. Of course gamers would want better battery life but it isn't a priority. Power and speed is everything and the ability to run a powerful separate GPU instead of an integrated one. There are certain constraints that an soc on arm with integrated graphics will have which a dedicated GPU won't.
Even if you use the best M2 Max configuration for GPU it doesn't beat Nvidia so I think the GPU side will be a wait and see for gamers.
And honestly Windows hasn't changed much since Windows 2000. Sure a lot of GUI layers were added, menus cleaned up and now with Windows 11 more window dressing than change. The one area where Windows has changed is security. But the underlying system, kernel, file system, etc has changed little. There are tons of old elements throughout the OS from Windows 98. Windows has been very much the exact same product for decades. A company that ditched the mobile space is going to shift to arm quickly is wishful thinking at best.
Yes, Windows is very slow to change substantially and generally only makes UI changes and adds more bloat with each release.
I liked Windows a lot more when they actually charged for each release of the OS. With this rolling release BS Microsoft has no incentive to really make a better product. They are trying to become Google with their revenue streams and it is a mistake.
But I digress. I really hope that Windows did push arm and moved the entire base off x86 ASAP. I am just being realistic. I don't see how it happens except for niche device releases for special use cases. Mainstream arm on Windows is a ways away.
I hope you prove me wrong though!!
Apple Silicon runs x86 applications at near native speed via Rosetta. I don't know that Qualcomm has the same translation support that Apple included, but I'd imagine they would if they really want to play in the PC market so I'd imagine Arm will be running x86 Windows at near native speed (or at least native speed a generation back, which is fine for legacy code).
I'm guessing the first generation Arm PCs will be at parity or better with x86 but somewhat hobbled by the Windows experience and the fact that the PC world treats everything as optional so a lot of applications won't be native at first. Shortly there after, Arm and x86 Windows will be at parity for a while. Then x86 Windows will be the lesser experience for a while before it sunsets native x86 support entirely.
It has to start somewhere though. It started with underperforming Arm chips and crappy Windows on Arm. Now we're getting competitive processors and will almost certainly get an improved Windows experience with this generation. In a few years we'll look back on it like the 64 bit transition-- a bit messy, but necessary and now behind us.
You're looking at this Qualcomm chip like it's a Mac that runs Windows. Qualcomm, Nvidia and AMD are looking at making PCs, so they'll be made for that market. Even this part clock scales in a way that Apple doesn't, has a boost mode similar to Intel, and seems to be more performance focused than efficiency focused. There's no reason to assume support for high performance graphics will be removed from the PC market.
Games will be written for whatever platform is performant and mass market enough to sell games. If Arm outruns x86, and I expect it will, and if a machine is made with sufficiently high performance graphics, and I expect one will, then games will run on it. I mean, Nvidia themselves are suggesting they'll get into Arm processors for PCs, so...
Will this Oryon do all that? Probably not. The excitement here is that it's an indication that the transition is beginning in earnest. Apple wasn't an outlier, they were simply first. Apple isn't out to devour Intel, but they chummed the water. Oryon is the first actual shark to arrive.
It's the biggest benefit of Windows, and I love it, and I'll mostly use windows on x86/64 and be happy as a clam.
That said, Windows on Arm has gotten a lot better in the last few builds, it's almost good enough to compete -- it's just got to have a good hardware platform to go on.
That is correct. And SnapdragonX laptop you can actually buy will be available in 2024. So it will compete with M3 class cpus from apple that you might be able to buy in 2023. This feels like Qualcomm had to make announcement because otherwise they would need to remove comparison with M2 and redo it for M3 were it might not be that rosy. Anyway, it is good that competition started. Intel & AMD are not exactly producing right HW for laptops. Yeah, AMD is a bit better..but still..
When Qualcomm cpu in laptop can report something like this:
CPU Power: 170 mW
GPU Power: 25 mW
ANE Power: 0 mW
Combined Power (CPU + GPU + ANE): 195 mW
many users will be happy. This is typical consumption for task "browsing the web" on M1 Max
M3 is based on 3nm while X elite is 4nm or another 5nm just like M2 series which is totally different.
Considering that these guys worked on the mobile ARM chips of Apple, it's not a surprise they can design a chip that can beats their own M-series chip they worked on.
And in terms of software, they can simply open it up for Linux or BSD which has better ARM support than Windows.
And in terms of software, they can simply open it up for Linux or BSD which has better ARM support than Windows.
Yeah, peaks are bigger, but that is not the point. Most of the time you just staring to the page, or typing something. And yes, most of the power usage takes display.That's a joke, right.
It's easy to measure the power consumption in real-time:
```
sudo powermetrics -i 1000 --samplers cpu_power,gpu_power -a --hide-cpu-duty-cycle --show-usage-summary --show-extra-power-info
```
On my MBP 16" with M1 Pro, combined power was around 1w to 3w when playing 4k Youtube video. If you load/refresh a web page, peak power usage could shoot to 5w and more depending on the complexity of web pages.
Apple's official spec for 2021 M1 MBP 16" was 14 hours wireless web. Given the 100Wh battery, it ends up with about 7.14w power draw for the full system (including the LCD panel with 50% brightness (8 clicks from bottom)). That seems fair.
But again, the power consumption of Apple Silicon is amazing.
True but they are using the N4P node which is pretty close to N3 for performance and power. The major difference is the density improvements on 3nm. Attached the table from Anandtech.
5nm Technology - Taiwan Semiconductor Manufacturing Company Limited
TSMC’s 5nm (N5) Fin Field-Effect Transistor (FinFET) technology successfully entered volume production in the second quarter of 2020 and experienced a strong ramp in the second half of 2020.
So much a failure that Qualcomm just bet its entire product range on it. Sometimes it helps to think before speaking.
A bit worrying that they can’t do better than Firestorm in IPC. The single-core boost is merely circumstantial and it’s not entirely clear what cost it commands.
The table is clear to me for both power and performance for N4P vs N5 and N3 vs N5.4nm is just another 5nm which TSMC officially said. Beside, where does it say N4P is close to N3? N4 is a failure anyway just like A16 proved.5nm Technology - Taiwan Semiconductor Manufacturing Company Limited
TSMC’s 5nm (N5) Fin Field-Effect Transistor (FinFET) technology successfully entered volume production in the second quarter of 2020 and experienced a strong ramp in the second half of 2020.www.tsmc.com
22% vs 25-30% - power
11% vs 10-15% - performance
4nm is just a 5nm failure and 3nm is also a failure, but you also can’t compare 3nm to 4nm because it’s a huge advantage.
Got it.
Got it.
same old sunny5 making funny comments4nm is just another 5nm which TSMC officially said. Beside, where does it say N4P is close to N3? N4 is a failure anyway just like A16 proved.5nm Technology - Taiwan Semiconductor Manufacturing Company Limited
TSMC’s 5nm (N5) Fin Field-Effect Transistor (FinFET) technology successfully entered volume production in the second quarter of 2020 and experienced a strong ramp in the second half of 2020.
As long as arm windows works beautiful on M Macs....is not the arm windows issue...its about the translation/emulation layer. Even their own surface arm windows worked like a garbage
So, a chip is not enough to work at least normal, like an ultrabook at minimum, not even talking for professional work
After over 4 years trying non mac arm windows based devices, i have almost 0 hopes for this to be for semi-professional work
Lets hope that they can compete with current ultrabooks devices at least...otherwise it will be another failure.
In reality, it's not that simple and like the chart said, it's just an advertisement value.
Where this chip would be useful is for desktop systems. On the desktop, we don't care so much about power usage but we want speed. Apple's chips are not optimized for desktops.View attachment 2301539
Qualcomm Unveils Snapdragon X Elite CPU PC Benchmarks: Oryon Core Faster & Efficient Than Intel 13th Gen & Apple M2 Max, GPU Faster Than AMD RDNA 3
Qualcomm has unveiled first benchmarks of its Snapdragon X Elite CPU for upcoming Windows PCs which will tackle Apple, Intel, and AMD.wccftech.com
View attachment 2301538
Maybe Apple Silicon is in a danger?
In any case, Qualcomm does NOT have the fastest ARM chip, not even close. I think Ampere wins. Just read their specs. (These are real chips that have been on the market for some time now. They are widely used in data centers.)
- 192 Cores
- 2 MB Private L2 Cache per Core
- 8 channel DDR5
- 128 lanes PCIe Gen5
- 350 W
More about Ampere here: https://amperecomputing.com/briefs/ampereone-family-product-brief