Qualcomm Reportedly Developing Apple M1 Competitor Dubbed Snapdragon SC8280

erek

[H]F Junkie
Joined
Dec 19, 2005
Messages
10,894
"The Qualcomm Snapdragon SC8280 was found in two configurations one with 8 GB of LPDDR5 RAM while the other was coupled with 32 GB of LPDDR4X memory. This is double what Apple offers with the M1 chip which is only available in 8 GB and 16 GB LPDDR4X configurations. The chip has also seen a 13% die size increase coming in at 20 mm x 17 mm up from the 20 mm x 15 mm on the 8cx Gen 2. This new processor is still under development and it is yet to been seen how it will compare with the M1 or the rumored 12-core Apple processor."

https://www.techpowerup.com/277399/...-apple-m1-competitor-dubbed-snapdragon-sc8280
 
Ya offering more RAM is nice but not the real determining factor is it. Considering how Qcom has always been a gen at least behind in mobile I doubt they are the laptop savior for the rest of the ARM market... never mind desktops. By the time Qcom has this ready to ship and assuming it can even show a small bump in performance over M1... Apple will be on to M1s replacement and the next bump up for imacs. Sounds like Qcomm is going to be performing the same dance. It was enough to get by in the mobile market as if you don't want Apple your options are not exactly plentiful in the Android world. In the Laptop/SFF M1 market.... Qcom competition is not Apple, It will be x86, and they already got spanked there the first go round of windows ARM I'm not convinced round 2 will go any different if their current target is = Apple product that has been shipping for months by the time they get this out the door.

If you really really want ARM... why not go Apple. They seem like strong products and unlike ANYTHING that comes out running windows ARM Apple is committed long term no worries your year old product and its entire software ecosystem will end up EOL.

If you just want a non Apple mobile... why go ARM at all frankly. Unless Qcomm and MICROSFT are offering something interesting... why switch from x86. (and I'm the ARM will win guy)

It chaffs my ass but I can already see Nvidia riding in when they close on ARM... with the real ARM CPU/GPU SOC for the non Apple ARM switch.
 
Last edited:
Apple will succeed because they control the entire ecosystem - hardware and software.

Qualcomm making chips doesn't do diddly squat in the big picture because they don't have a software ecosystem to connect it to that they control.

If Microsoft decided to switch Win to Arm entirely, that might be something but there are no other players as large as Apple or as monopollistic as Apple on the hard/software fronts to be able to do what they are doing.
 
I wonder if this has anything to do with the NUVIA acquisition
 
If you really really want ARM... why not go Apple. They seem like strong products and unlike ANYTHING that comes out running windows ARM Apple is committed long term no worries your year old product and its entire software ecosystem will end up EOL.
This developer will disagree with you. Apple going ARM is making it hard to justify the extra cost to port games to their platform. Apple not supporting Vulkan and apparently their Metal API driver is crap are reasons why some developers won't be supporting Apple and their venture to ARM. Qualcomm products at least support Vulkan and can probably have Linux installed at least, assuming they give you the privilege to install it.
If you just want a non Apple mobile... why go ARM at all frankly. Unless Qcomm and MICROSFT are offering something interesting... why switch from x86. (and I'm the ARM will win guy)
Why go ARM at all? X86 is an open platform which allows you to install just about any OS you want without restrictions. My nephews 2011 Macbook Air can't go past High Sierra but I was able to install Linux Mint and make it look like Mac OS X. I could even put Windows on it if I want to, though not sure about newer Intel Macs. Almost anything using ARM is locked down and you use and install what the company wants, not what you want. Until ARM has a more unified standard in what it allows it's users to install, I would stick to x86.
 
After reading that it seems more to me that he just isn't interested in learning Metal. He more or less states that Vulcan also has caused him a lot of headaches inside of Windows and in Linux but it was something he wanted to learn. There is a note about drivers and other issues inside of macOS, but even that stuff aside if he was getting paid enough to do it he still would develop on macOS. So, I wouldn't necessarily even take that account into a big negative for programming for Metal. (And there are a lot of iOS devs that will gladly program for desktop that already know Metal inside and out. I think if anything iOS devs will have a great opportunity to work for a lot of companies to develop for macOS).

Also as has been beaten to death, there already exists translation software: MoltenVK. From a programming perspective, what I understand is that Vulkan and Metal aren't really that different.

Apple going ARM is making it hard to justify the extra cost to port games to their platform. Apple not supporting Vulkan and apparently their Metal API driver is crap are reasons why some developers won't be supporting Apple and their venture to ARM. Qualcomm products at least support Vulkan and can probably have Linux installed at least, assuming they give you the privilege to install it.
We'll see won't we? It's far too early to see what adoption rates onto Arm Mac are going to be like. If Apple gets appreciable market-share it won't matter how hard it is to program for. It could be the hardest damn platform on the planet, but if devs want that base and that money they'll do it. For an example of this look up the history of the cell processor and the PS2. If 3/4 HS and College kids move to a Mac, you'd better be damn sure that EA will have the next Call of Duty on the platform. It won't be the first time or the last.

Why go ARM at all? X86 is an open platform which allows you to install just about any OS you want without restrictions. My nephews 2011 Macbook Air can't go past High Sierra but I was able to install Linux Mint and make it look like Mac OS X. I could even put Windows on it if I want to, though not sure about newer Intel Macs. Almost anything using ARM is locked down and you use and install what the company wants, not what you want. Until ARM has a more unified standard in what it allows it's users to install, I would stick to x86.
With this I don't disagree. For most Windows and Linux junkies it probably just makes more sense to stay on x86 for the time being. The "issue" is that M1 being even an entry level low powered SOC is already destroying systems that cost 2x-3x as much for half the cost. While M1 isn't the best at general compute and obviously for Mac ARM to have a future it will have to have better GPU capabilities - it's prowess is telling. If it gets to the point in which a Mac is 2-4x as powerful as an x86 machine for half the cost it will seem dumb to be on an x86 machine anymore - despite the restrictions I know you hate so much about being on macOS.

Intel is in flux and although it's now known that they're moving to TSMC, it's still very unknown how long it will take them to get their ish together. AMD is doing well for the first time in a long time, but they still have nothing that can match speed and low power consumption of ARM. Not that any of this will convince you. I'm certain you'll more or less make the statement that you'd rather pay 10x more for an x86 machine that's 600% slower rather than ever move to a Mac, but whatever. That's your prerogative.


As for the discussion in the OP: I think Qualcomm sees the writting on the wall and also sees an opportunity to move into a new market: desktop/laptop CPU's. I'll agree with everyone else's assessments though, they have to beat Apple and frankly I don't think they can. nVidia has a much better chance at doing so if/when their ARM acquisition goes through than Qualcomm does. I think there is a very high possibility that the CPU landscape is going to look very different in 10 years in comparison to today. The players might be entirely different companies than we've been used to for the past 35+ years.
 
Last edited:
Why go ARM at all? X86 is an open platform which allows you to install just about any OS you want without restrictions. My nephews 2011 Macbook Air can't go past High Sierra but I was able to install Linux Mint and make it look like Mac OS X. I could even put Windows on it if I want to, though not sure about newer Intel Macs. Almost anything using ARM is locked down and you use and install what the company wants, not what you want. Until ARM has a more unified standard in what it allows it's users to install, I would stick to x86.

For cutting-edge performance.

While it's not a 100% slam dunk yet, with their first try Apple is already delivering performance multiples better than comparable Intel mobile cpus in certain tasks like video editing. Even running Rosetta, Apple's M1 is faster with native x86 applications than comparable Intel processors.

If that gap widens in the future with their higher performance variants and further software adoption, it will just be a no-brainer for the mobile space. I think desktops will be more competitive for a long time (I don't see Apple beating my 3990x for a while), but in mobile I think Apple will make staying on x86 pretty stupid over the next 3-5 years.
 
...

Why go ARM at all? X86 is an open platform which allows you to install just about any OS you want without restrictions. My nephews 2011 Macbook Air can't go past High Sierra but I was able to install Linux Mint and make it look like Mac OS X. I could even put Windows on it if I want to, though not sure about newer Intel Macs. Almost anything using ARM is locked down and you use and install what the company wants, not what you want. Until ARM has a more unified standard in what it allows it's users to install, I would stick to x86.
x86 is very far from an open platform. The only performant and truly 100% open "platform" (bootloader, firmware, microcode running on-chip ie. ME and power management type stuff) is the Raptor POWER9 Blackbird.

Complete sources are available:
https://git.raptorcs.com/git/

That is an actual open platform without any unknown/third-party/proprietary binary blobs (you can build it all yourself).
 
This developer will disagree with you. Apple going ARM is making it hard to justify the extra cost to port games to their platform. Apple not supporting Vulkan and apparently their Metal API driver is crap are reasons why some developers won't be supporting Apple and their venture to ARM. Qualcomm products at least support Vulkan and can probably have Linux installed at least, assuming they give you the privilege to install it.

Why go ARM at all? X86 is an open platform which allows you to install just about any OS you want without restrictions. My nephews 2011 Macbook Air can't go past High Sierra but I was able to install Linux Mint and make it look like Mac OS X. I could even put Windows on it if I want to, though not sure about newer Intel Macs. Almost anything using ARM is locked down and you use and install what the company wants, not what you want. Until ARM has a more unified standard in what it allows it's users to install, I would stick to x86.
I see the developers point... and on the other hand. Who cares. I don't think anyone is buying M1 stuff for games really, at least not real AAA level stuff. Ultra portables of any OS and MFG are not exactly known as high end gaming machines. I agree Apple is stupid to stick to their own API but ya if your looking for a gaming machine ultra portable M1 and apple are probably not high on your shopping list. I don't expect their desktops are going to get great AAA type support either... but they will have at least a very good selection of casual gaming which is the bulk of the market no matter what most of us think.

I agree if ARM is going to take over the PC world beyond Apple they are going to need to have a bit more of a plan. But don't underestimate how badly MS wants to emulate Apple. And how badly Nvidia wants to Kill Intel... and take their place as the king of CPUs (not just GPUs) for the next few decades. Microsoft is going to push ARM if they have to make the chip themselves (which they are probably doing if rumors are true). Nvidia didn't shell out all the money for ARM to just do what they are doing now. Expect a full on ARM platform from Nvidia in the next year or two.... I wouldn't put it past Nvidia to build both High end ARM/NV GPU SOC to compete with M1, and full on Motherboards with socket ARM chips... and PCIx slots for GPUs to compete directly with Intel AMD and x86.

The next few years are going to be wild. Windows ARM will be a real market thing... MS silicon. Nvidia will release chips we will have to compare to Ryzens and Intels i123456789000+ stuff. Does Nvidia get solidly behind Windows ARM... or does MS fall on their face. As crazy as it seems Nvidia could end being a Linux pusher... and go all in on their own Apple like 100% nvidia solution. Imagine Nvidia getting friendly with the open source community and building building a AAA ARM Linux gaming system. As crazy as that sounds... hey don't be shocked if they try. Imagine a steam machine with a 12-16 core Nvidia ARM chip... with either a decent SOC GPU, or a full fledged PCIx port for GPUs, all running Nvidias version of a SteamOS. Nvidia could push developers to ARM native faster then anyone.
 
Last edited:
I see the developers point... and on the other hand. Who cares. I don't think anyone is buying M1 stuff for games really, at least not real AAA level stuff. Ultra portables of any OS and MFG are not exactly known as high end gaming machines. I agree Apple is stupid to stick to their own API but ya if your looking for a gaming machine ultra portable M1 and apple are probably not high on your shopping list. I don't expect their desktops are going to get great AAA type support either... but they will have at least a very good selection of casual gaming which is the bulk of the market no matter what most of us think.

I agree if ARM is going to take over the PC world beyond Apple they are going to need to have a bit more of a plan. But don't underestimate how badly MS wants to emulate Apple. And how badly Nvidia wants to Kill Intel... and take their place as the king of CPUs (not just GPUs) for the next few decades. Microsoft is going to push ARM if they have to make the chip themselves (which they are probably doing if rumors are true). Nvidia didn't shell out all the money for ARM to just do what they are doing now. Expect a full on ARM platform from Nvidia in the next year or two.... I wouldn't put it past Nvidia to build both High end ARM/NV GPU SOC to compete with M1, and full on Motherboards with socket ARM chips... and PCIx slots for GPUs to compete directly with Intel AMD and x86.

The next few years are going to be wild. Windows ARM will be a real market thing... MS silicon. Nvidia will release chips we will have to compare to Ryzens and Intels i123456789000+ stuff. Does Nvidia get solidly behind Windows ARM... or does MS fall on their face. As crazy as it seems Nvidia could end being a Linux pusher... and go all in on their own Apple like 100% nvidia solution. Imagine Nvidia getting friendly with the open source community and building building a AAA ARM Linux gaming system. As crazy as that sounds... hey don't be shocked if they try. Imagine a steam machine with a 12-16 core Nvidia ARM chip... with either a decent SOC GPU, or a full fledged PCIx port for GPUs, all running Nvidias version of a SteamOS. Nvidia could push developers to ARM native faster then anyone.
I would love to see what NVidia could propose to Nintendo in terms of a next-gen console, I love my Switch for what it is but I would love to see what Nintendo could do with whatever NVidia's response to the M1 would be in that case, or more likely the M2 or M3 by the time they get the thing ready to actually ship.

That or launch a beefy version of the Shield, not sure if SteamOS would be the best option to go with but it would be at least an option. NVidia has enough pull in the industry that they could help fund ARM ports, make it worth the developer's while. The real question is can NVidia work out a deal with Apple to get their video hardware up in the Mac's, Vulkan to Metal translation layers exist and they aren't resource prohibitive to implement, and if they could work out that deal then it would go a very long way towards convincing publishers to support an ARM launch if there is already a potentially large install base.
 
Last edited:
  • Like
Reactions: ChadD
like this
I would love to see what NVidia could propose to Nintendo in terms of a next-gen console, I love my Switch for what it is but I would love to see what Nintendo could do with whatever NVidia's response to the M1 would be in that case, or more likely the M2 or M3 by the time they get the thing ready to actually ship.

That or launch a beefy version of the Shield, not sure if SteamOS would be the best option to go with but it would be at least an option. NVidia has enough pull in the industry that they could help fund ARM ports, make it worth the developer's while. The real question is can NVidia work out a deal with Apple to get their video hardware up in the Mac's, Vulkan to Metal translation layers exist and they aren't resource prohibitive to implement, and if they could work out that deal then it would go a very long way towards convincing publishers to support an ARM launch if there is already a potentially large install base.
Unfortunently Nintendo doesn't care about hardware at all they never have... which is why they laughed in Microsofts face when they proposed building a Nintendo Xbox. MS offered to handle 100% the building of the latest greatest next gen for them every few years for ever and Nintendo said na... the games we want to make don't need that. And they don't. (as cool as a high end switch could be... Nintendo just won't chase that dragon) Even the earlies Nintendo systems where actually based on old hardware... home consoles where just not being done after the Atari face plant. The hardware used in the first NES was already 2 or 3 years old when they launched it in the US.

As for Nvidia... I have no doubt they have HUGE plans that are awaiting regulatory approval on their purchase of ARM. I have a feeling they are coming for Intel hard.... its not Apple Intel really has to worry about that was the cannery in the mine so to speak. Ya it shows what ARM is capable of if designers and companies take it seriously in terms of performance first (and not just chips that won't cook your leg when you have it stuffed in your pants... no one wants law suits cause their Newton device burned someone's shlong)
Intels realy hurdle is coming when Nvidias purchase clears... and I'm sure it will once NV greases the right palms. At that point ya Nvidia who has always hated Intel is coming not just for the crown but the Intel killing blow(s)... they will come at consumer, and more importantly they are coming after the server space. Amazon is already killing x86 with Graviton and G3 will be out before NV closes on ARM I'm sure... but when they do expect big Nvidia Server announcements (imagine a graviton 3 type chip sold by Nvidia directly to server farms right along side their GPU server stuff... there also going to go after some prestige super computer contracts as well I'm sure, Jensen is busting with excitement about talking about a 100% NV super computer in the #1 slot.)

In the game world I am sure Microsoft will fall all over themselves to help Nvidia power Windows ARM. Gaming really won't need any porting at all for older titles if Microsoft does translation properly which could be helped a ton if Nvidia does like Apple and builds their chip properly to emulate x86 cache and memory writes in hardware. Newer titles will just naturally migrate over if Nvidia manages to start getting their ARM CPUs out into the market. And they now how to market to gamers.... expect less upgradeable stuff... but it will still have that all important GPU loving PCIx slot. They may socket the CPU who knows.... that would be something though if they came up with a Nvidia Socket, owning ARM would make them the to to to set such a standard.

I feel for Intels new CEO... yes they needed a technically minded CEO again to steer the ship. But man what a mine field Intel faces in the next few years. Turn the wrong way and he'll have a short 1-2 year go as well. I doubt their big.Little is going to save them consumer side from the incoming ARM onslaught... its for sure going to do nothing in the server space where ARM is not using big.Little they are running ARM Neoverse cores, which Intel really has no answer for... so far their answer is 5 year old core designs on a very solid 14nm++++ process that are no doubt dialed in but ARM just gets better and better. It will be interesting to see which products they are farming out to TMSC... are they going to try and farm their server stuff, their consumer stuff Both ? (TMSC isn't likely to give them the capacity to do both) Not sure Intel has 4 years to fix things.... or if their shareholders will give Pat Gelsinger that sort of time. Short term there going to have to do some insane juggling act its going to be a train wreck probably. (a 14nm++ 8core backport isn't going to save them on the high end.... and their next Intel fabbed chip isn't going to compete with anything its going to have 4 atom cores for F sake, and probably still burn more power then Apples and perhaps even AMDs mobile chips.)

As for NV going in Macs... na never going to happen. Apple is going 100% Apple in all things. Their M1 has a Apple GPU... make no mistake their imac is going to have its own Apple GPU, and the real story of the Mac pro desktops when they hit may be more about the Apple designed GPU/Gcompute bits they have.
 
Last edited:
The real question is can NVidia work out a deal with Apple to get their video hardware up in the Mac's, Vulkan to Metal translation layers exist and they aren't resource prohibitive to implement, and if they could work out that deal then it would go a very long way towards convincing publishers to support an ARM launch if there is already a potentially large install base.
Apple going back to Nvidia chips is about as likely as them going back to Intel.
 
Back
Top