Qualcomm Will Try To Have Its Apple Silicon Moment in PCs With 'Snapdragon X' (arstechnica.com) 32
Qualcomm's annual "Snapdragon Summit" is coming up later this month, and the company appears ready to share more about its long-planned next-generation Arm processor for PCs. ArsTechnica: The company hasn't shared many specifics yet, but yesterday we finally got a name: "Snapdragon X," which is coming in 2024, and it may finally do for Arm-powered Windows PCs what Apple Silicon chips did for Macs a few years ago (though it's coming a bit later than Qualcomm had initially hoped). Qualcomm has been making chips for PCs for years, most recently the Snapdragon 8cx Gen 3 (you might also know it as the Microsoft SQ3, which is what the chip is called in Surface devices). But those chips have never quite been fast enough to challenge Intel's Core or AMD's Ryzen CPUs in mainstream laptops. Any performance deficit is especially noticeable because many people will run at least a few apps designed for the x86 version of Windows, code that needs to be translated on the fly for Arm processors.
So why will Snapdragon X be any different? It's because these will be the first chips born of Qualcomm's acquisition of Nuvia in 2021. Nuvia was founded and staffed by quite a few key personnel from Apple's chipmaking operation, the team that had already upended a small corner of the x86 PC market by designing the Apple M1 and its offshoots. Apple had sued Nuvia co-founder and current Qualcomm engineering SVP Gerard Williams for poaching Apple employees, though the company dropped the suit without comment earlier this year. The most significant change from current Qualcomm chips will be a CPU architecture called Oryon, Qualcomm's first fully custom Arm CPU design since the original Kryo cores back in 2015. All subsequent versions of Kryo, from 2016 to now, have been tweaked versions of off-the-shelf Arm Cortex processors rather than fully custom designs. As we've seen in the M1 and M2, using a custom design with the same Arm instruction set gives chip designers the opportunity to boost performance for everyday workloads while still maintaining impressive power usage and battery life.
So why will Snapdragon X be any different? It's because these will be the first chips born of Qualcomm's acquisition of Nuvia in 2021. Nuvia was founded and staffed by quite a few key personnel from Apple's chipmaking operation, the team that had already upended a small corner of the x86 PC market by designing the Apple M1 and its offshoots. Apple had sued Nuvia co-founder and current Qualcomm engineering SVP Gerard Williams for poaching Apple employees, though the company dropped the suit without comment earlier this year. The most significant change from current Qualcomm chips will be a CPU architecture called Oryon, Qualcomm's first fully custom Arm CPU design since the original Kryo cores back in 2015. All subsequent versions of Kryo, from 2016 to now, have been tweaked versions of off-the-shelf Arm Cortex processors rather than fully custom designs. As we've seen in the M1 and M2, using a custom design with the same Arm instruction set gives chip designers the opportunity to boost performance for everyday workloads while still maintaining impressive power usage and battery life.
Comment removed (Score:5, Interesting)
Re: (Score:3, Informative)
Re: (Score:3)
Conversely, if you want to live in the real world, all OEMs have had these struggles from time to time. The Nexus 6P was such a turd it ended in one of the few class action lawsuits [topclassactions.com] that paid me real money. I got $400 for that $600 paperweight. The Nexus 6 and first generation Pixel had hardware defects too and those owners (I had both) didn't get anything for their trouble.
LOL, why in the hell would I argue conversely when you have listed every reason I'm mocking this? Consumers are the reason class-action lawsuits are a calculated business expense now.
I could use these examples as an excuse to shit on Google's QA/QC process, or I could just accept the fact that these are super complicated devices and sometimes the OEMs get it wrong.
Thank you for confirming that you have a creative way of saying some don't know what the fuck they're doing from time to time. The reason they don't care? Consumers.
Re: (Score:2, Interesting)
Re: (Score:2)
Not sure exactly what your beef is here boss. You think you could design a smartphone completely free of software and hardware glitches? Go for it. Apple and Google will fight to the death to license your design. You'll have Oprah money before the year is out.
For two people seemingly standing on the same side of frustration as often-screwed consumers, you certainly are quite optimistic as to how the above scenario would actually play out in the Twenty First Century.
Does someone really need to try and stand up a business behind diapers.com to validate my point about what mega-corps like Amazon are prepared to do in order to win? Do we really need to list all of the former, current, and pending lawsuits between market giants Samsung and Apple? Or Samsung/Apple v
Re: (Score:2)
I haven't had this experience with my Apple laptops. The M1/M2 chips are also faster, in general + media use, than anything x86/amd64 I've seen.
Re: (Score:2)
Re: (Score:1)
I've been running PC games at full framerates under wine using rosetta on my trusty M1 Max and I've yet to hear the fans kick in (Oh I suppose they are, but its an exceedingly quiet design), and its always cool to touch.
Re: (Score:1)
Re: (Score:2)
On the other hand, Baldur's Gate 3 (native ARM64 code) on my MacBook Pro with M2 Max burns more than 100W, and the fans are pretty loud. If I use a docking station or 100W USB-C supply rather than the 140W wall wart, the battery charge drops quite quickly while playing.
Re: (Score:2)
Re: (Score:2)
It's not just cooling capacity. Laptops set a power limit for the CPU and GPU too.
You can get beefy workstation laptops with good cooling, depending on your workflow it might be better to have lots of cores or a GPU that can do encoding/decoding for you. Apple has the advantage of being a monoculture when it comes to GPUs, so support for them in apps is sometimes better than it is on Windows/Linux. Their CPUs can't come close to AMD and Intel ones though, even when thermally constrained, outside of very spe
Re: (Score:2)
I have a Dell Ryzen laptop that throttles to 400 MHz constantly because of insufficient cooling, it's very loud too.
Re: (Score:2)
As someone who uses Windows ARM every day... (Score:2)
... I am looking forward to seeing how this performs. I've been using a Surface Pro X (running an ARM CPU) for most of my mobile computing for over a year now. Most of the applications I use are now ARM native and run very well, and the few that are not have been running perfectly via the x86/x64 compatibility layer, albeit a little bit slower. Overall the experience has been very good, and the few minor performance hiccups I experience are worth it when you consider the tradeoffs -- the battery life is
Re: (Score:2)
What advantages are these?
So far for general computing ARM comes up short. There's no standardization of a platform that an OS can support, no standardized hardware tree or boot loader. I understand that Windows ARM runs on laptops that have a form of UEFI, but it's completely locked down. Outside of the very small Windows ARM ecosystem, things are not great. You can't just grab a standard Linux ISO, stick it on a usb stick and install it on your favorite ARM computer. All current ARM chips require ker
Re: (Score:3)
Re: (Score:2)
Asking for a friend: you've had issues with the Unifi manager on a Pi? I was seriously considering it...
Re: (Score:2)
Re: (Score:2)
Due to their hanging on to Java8 and MongoDB 3.6, the only practical way to run the Unifi controller is in a docker container.
I admit to not running it on Raspbian anymore, most recently I'm running it on Debian 11.7 on an RK3399 NanoPC-T4, and a docker image that I think is based on lscr.io/linuxserver/unifi-controller.
Note that it should work, with the above container, on RaspberryOS 64bit. I just haven't a Pi4 to try it with, and 1GB RAM isn't enough to run the controller with a Pi3.
Re: (Score:2)
That's not what I mean. Yes the packages are available. But you can't go to debian.org or fedora.org and download their ISO, put it on a usb stick and put it on your pi and all the other ARM devices. Each device requires a custom kernel fork, funky gpu drivers, and incompatible boot processes. If your goal is to compete with x86 devices, this does not cut it.
I've got an OPi5 on the desk here that I bought in a weak moment. Amazing system on paper, but what a nightmare in reality. And most other ARM d
Re: (Score:2)
Serious question, why isn't establishing such an standard a main goal of the people behind Raspberry Pi? They are given the unique opportunity to control a hardware specification that everybody will want to be compatible with (like Arduino).
Re: (Score:2)
There is a project called DAS-U-boat that aims to be a universal u-boot that can allow supported SoCs to all boot the same way, supporting usb, sd-card, mmc, etc.
Really UEFI will do everything, but that adds cost, so SoCs will never have it. You can buy arm laptops right now with UEFI, but they are locked down tight and can only boot a factory-installed version of Windows 10 or 11.
I think it's untrue that Pi controls the hardware specification. In reality they just choose chips from a vendor (Broadcom) an
Re: (Score:2)
There are updated images from Armbian, albeit they're currently "nightly" quality, running kernel 6.6.0-rc1. So there's hope for the OPi5.
I have no idea when the GPU support will be fixed.
Re: (Score:2)
There Is No Try. . . (Score:2)
And no matter what, Qualcomm will always be at least two, and soon to be three major generations behind the Mega Corporation, Apple, with not only more practical ARM experience than basically anyone in the world; but a Mega Corporation that has bet the entire farm on balls-to-the-walls continuous R&D on that Platform.
Apple has absolutely nothing to worry about here. Qualcomm needs to stick to their core competency: RF MODEMS; at which they excel.
They suck at SoC design; and even poaching a few team memb
Re: (Score:2)
Re: (Score:2)
Qualcomm doesn't need to be better than Apple with ARM processors. They need to make ARM processors that Windows uses will find as good or better than Intel/AMD (x86_64) offerings.
Still wouldn't hold my breath.
Qualcomm sucks at processors.
DOA (Score:3)
The whole point of using Windows is to run existing x86 Windows applications, otherwise Arm or not, you can just as well run Linux, Android or ChromeOS (I know, I repeat myself, all of those are Linux) on your device. So if what you want is running x86 Windows apps on Windows, better stay with an x86 CPU.
Re: (Score:2)
Microsoft and Qualcomm are going to need to help and possibly pay companies to migrate apps over to ARM quickly. The entire adobe suite is one example, but there are many others if they want their apple M1 downgrade moment. For apple, blowing off gamers didn't matter too much and it cost them counter strike. Microsoft actually needs gamers. Where's the ARM kit and compatibility toolkit for that? Did anyone get valve on board? Epic? Activision Blizzard?
Arm based macs are only good at youtube video cre