Qualcomm Snapdragon X Elite Put Through Graphics Tests, Beats AMD Radeon 780M iGPU in 3DMark

[Battler624]

I want one, but does linux have an answer to rosetta2 yet?

There is x64/86 emulation on linux via box64/box86.

Also, iirc someone was able to run rosetta2 on amazon servers running linux, 3 hiccups happened, one was kernel (some kernel 4kb or 16kb stuff), some memory ordering stuff that is different between arm/x86 (apple follows intel memory when running a rosetta app), and iirc it required a binary taken from an ARM-based mac.

The last one can be reverse engineered/remade in software, the first 2 require hardware level stuff (and thus require to be emulated on un-supported hardware). And not every application required the issues to be solved. This is all for near-native speeds, if you dont care about near-native then yea all can be done in software.

 

[tabascosauz]

I didn't expect the 780M to beat M2's GPU in anything, but it does fairly well in Control. The M2 enjoys access to a 8 MB SLC whereas the 780M has to make do with a 2 MB L2 cache for the GPU. Is there any word on the memory configuration for the x86 laptops?

It's a pretty mid showing for 780M, but not surprising considering the wattage limitations ("23W").

-HS Phoenix immediately screams early release gaming laptop as opposed to more probability of LPDDR for -U. Some of those gaming laptops not only come with JEDEC DDR5 but also come out of the factory with a single channel config. But the low perf probably still comes more from the power budget.

As long as the node of their choosing is well paired with the clock envelope they want, it doesn't seem surprising that Qualcomm should be able to pull it off. It is a new custom chip after all, look what they already could do in an ultrabook form factor with their low effort Compute Platform rebrands.

 

[mplayerMuPDF]

I am happy that Qualcomm is investing their significant resources into desktop ARM and bringing some competition to Intel and AMD, but I will never trust or like them. I am much more interested in an Nvidia PC ARM chip. We need both ARM and RISC-V to get some competition and diversity in the PC SoC market instead of this endless duopoly where garbage such as Pluton and ever more DRM gets pushed down our throats. I will never understand those who insist that we should stick with x86/AMD64 forever and who are deeply opposed to RISC-V and/or ARM for some reason.

Regardless, the primary purpose of the Linux demo was to showcase that Linux was working on the Snapdragon Elite X as well – that it’s not just for Windows – as Qualcomm has aims of getting the SoC into Linux laptops as well.

From the AnandTech review @Tek-Check linked to; very interesting!

 

[AnotherReader]

I am happy that Qualcomm is investing their significant resources into desktop ARM and bringing some competition to Intel and AMD, but I will never trust or like them. I am much more interested in an Nvidia PC ARM chip. We need both ARM and RISC-V to get some competition and diversity in the PC SoC market instead of this endless duopoly where garbage such as Pluton and ever more DRM gets pushed down our throats. I will never understand those who insist that we should stick with x86/AMD64 forever and who are deeply opposed to RISC-V and/or ARM for some reason.

I've got two words for you: backwards compatibility. That's perhaps the biggest reason PC users use PCs and stick to Windows or Linux instead of migrating to Mac OS.

 

[mplayerMuPDF]

I've got two words for you: backwards compatibility. That's perhaps the biggest reason PC users use PCs and stick to Windows or Linux instead of migrating to Mac OS.

How many people are actually using legacy apps these days? Not saying that it is an insignificant number but probably not a majority. Most people doing basic stuff (mostly in the browser) who are not playing hard core games are not going to be using AMD64 or x86-only applications. I feel like this gets way overstated during these kinds of discussions. Probably due to the bias of this site, where most people are heavily focused on playing e.g. Steam/GOG/Origin/whatever games to the extent that people believe that ISA compatibility is actually a major issue on Linux when it really is not. Hard core gamers should probably stick with their custom built latest gen Ryzen/Core Ultra desktops, but that has nothing to do with the average user who would like to get long battery life browsing, watching Netflix or working on essays/presentations out of a thin and light laptop.

 

[CosmicWanderer]

Linux doesn't need any emulation & MS will probably eventually get native ARM builds like in the past.

Native ARM builds are already available. The latest Surface Pro 9 has an ARM variant.

 

[AnotherReader]

How many people are actually using legacy apps these days? Not saying that it is an insignificant number but probably not a majority. Most people doing basic stuff (mostly in the browser) who are not playing hard core games are not going to be using AMD64 or x86-only applications. I feel like this gets way overstated during these kinds of discussions. Probably due to the bias of this site, where most people are heavily focused on playing e.g. Steam/GOG/Origin/whatever games to the extent that people believe that ISA compatibility is actually a major issue on Linux when it really is not. Hard core gamers should probably stick with their custom built latest gen Ryzen/Core Ultra desktops, but that has nothing to do with the average user who would like to get long battery life browsing, watching Netflix or working on essays/presentations out of a thin and light laptop.

I agree that for the average consumer, with the move to browser based apps and software as a service, backwards compatibility isn't a big concern, but you're wrong about Linux. As recently as 2022, the latest kernel supported the 486 which was released in 1989. Then there are enterprise customers who definitely want backwards compatibility.

 

[mplayerMuPDF]

I agree that for the average consumer, with the move to browser based apps and software as a service, backwards compatibility isn't a big concern, but you're wrong about Linux. As recently as 2022, the latest kernel supported the 486 which was released in 1989. Then there are enterprise customers who definitely want backwards compatibility.

I am not sure what your point is with Linux still supporting the 80486, to be honest. Linux supports many architectures and old CPUs. What I am talking about is that most Linux users get virtually all their software from the official repositories and that these days there is virtual parity with AMD64 in the ARM64 (or aarch64 as it is sometimes also called) part of those repositories in terms of what packages are available. The biggest issue is probably Flatpak/Flathub, which also has many (relatively) popular proprietary applications, but if these Qualcomm-powered laptops were to really catch on I think that ARM64 versions would relatively quickly be offered there.

I agree that for the average consumer, with the move to browser based apps and software as a service, backwards compatibility isn't a big concern, but you're wrong about Linux. As recently as 2022, the latest kernel supported the 486 which was released in 1989. Then there are enterprise customers who definitely want backwards compatibility.

I am not sure what your point is with Linux still supporting the 80486, to be honest. Linux supports many architectures and old CPUs. What I am talking about is that most Linux users get virtually all their software from the official repositories and that these days there is virtual parity with AMD64 in the ARM64 (or aarch64 as it is sometimes also called) part of those repositories in terms of what packages are available. The biggest issue is probably Flatpak/Flathub, which also has many (relatively) popular proprietary applications, but if these Qualcomm-powered laptops were to really catch on I think that ARM64 versions would relatively quickly be offered there.

 

[AnotherReader]

I am not sure what your point is with Linux still supporting the 80486, to be honest. Linux supports many architectures and old CPUs. What I am talking about is that most Linux users get virtually all their software from the official repositories and that these days there is virtual parity with AMD64 in the ARM64 (or aarch64 as it is sometimes also called) part of those repositories in terms of what packages are available. The biggest issue is probably Flatpak/Flathub, which also has many (relatively) popular proprietary applications, but if these Qualcomm-powered laptops were to really catch on I think that ARM64 versions would relatively quickly be offered there.

The point is that it reflects a difference of mindset. Unlike Apple, Linux and Windows don't force users of older platforms to stay on the upgrade treadmill to keep their applications running.

 

[RayneYoruka]

I want one, but does linux have an answer to rosetta2 yet?

IDK tbh

 

[TheinsanegamerN]

How many people are actually using legacy apps these days? Not saying that it is an insignificant number but probably not a majority. Most people doing basic stuff (mostly in the browser) who are not playing hard core games are not going to be using AMD64 or x86-only applications. I feel like this gets way overstated during these kinds of discussions. Probably due to the bias of this site, where most people are heavily focused on playing e.g. Steam/GOG/Origin/whatever games to the extent that people believe that ISA compatibility is actually a major issue on Linux when it really is not. Hard core gamers should probably stick with their custom built latest gen Ryzen/Core Ultra desktops, but that has nothing to do with the average user who would like to get long battery life browsing, watching Netflix or working on essays/presentations out of a thin and light laptop.

There's all sorts of apps and interfaces in manufacturing, medical, media production, and financial sectors that is strictly windows only. Sure, it may not even be MOST apps today, but in business it is 100% or 0%, none of this "well it works well enough" stuff some consumers choose to accept.

You can poo-poo gamers all you want, but they make up a significant portion of remaining PC users. Steam alone has 120 million monthly active users (not total users). That 9% of the global windows population, not just consumer. Consumer only, it's likely far higher. And they spend big money.

The average user who does everything via web browser has already started jumping to chromebooks, mac OS, linux, or most commonly, phones and tablets, which make up an ever growing % of total online devices. The current PC industry is made up of people who need legacy support. You may not value it, but clearly there's 1.4 billion users who do.

If it didnt matter, then windows RT, windows 10 S, and windows on ARM would have become major successes. Instead they have all, so far, failed dramatically. More importantly, if you dont need legacy apps, WHY USE WINDOWS AT ALL? MS is well aware (or at least it used to be) that legacy IS its market, without legacy support windows will die out, that's why they've backed down from any attempt to replace the win32 API.

 

[dj-electric]

I feel like bringing up another topic that wasn't brought up here.

Qualcomm is using 2 design reference units. Config A is at 80W and config B is at 23W.
What happened here? Why is power scalability so incredibly poor between 23W and nearly 4 times as much power?

If anything, this shows me that the PPW curve is heavily lower scale leaning. Its not that the Config B sample shows particularly poor performance, but I honestly don't know why Qualcomm decided 80W is a good TDP as a secondary, high performance mode one.

To me, from a silicon perspective this proves one thing. Well, two things. One - keep the SD X Elite at where it performs efficiently. Two - maybe above 25W think of a more power efficient core complex instead of pushing the hardware so hard with frequency you gain 30% performance for nearly 4 times the power consumption. How about you shove more X4 cores into it instead, Qualcomm?

I have no doubt they will, and hope this kind of more higher performance tuned chip will rely on multiple X4 cores

 

[TheinsanegamerN]

I feel like bringing up another topic that wasn't brought up here.

View attachment 319588

Qualcomm is using 2 design reference units. Config A is at 80W and config B is at 23W.
What happened here? Why is power scalability so incredibly poor between 23W and nearly 4 times as much power?

If anything, this shows me that the PPW curve is heavily lower scale leaning. Its not that the Config B sample shows particularly poor performance, but I honestly don't know why Qualcomm decided 80W is a good TDP as a secondary, high performance mode one.

To me, from a silicon perspective this proves one thing. Well, two things. One - keep the SD X Elite at where it performs efficiently. Two - maybe above 25W think of a more power efficient core complex instead of pushing the hardware so hard with frequency you gain 30% performance for nearly 4 times the power consumption. How about you shove more X4 cores into it instead, Qualcomm?

I have no doubt they well, and hope this kind of more higher performance tuned chip will rely on multiple X4 cores

Well, two ideas.

1 - chasing a graph leader. We see it every day with GPUs and CPUs pushed well out of the efficiency curve for 1-3% gains. No reason it would be different here.

2- it's a TDP, not measured power draw. Just because the A config is 80w does not mean the CPU itself draws 80w. Remember it has a big iGPU in there too, and that is included in that TDP. We will have to wait and see, but its likely the 80w model isnt pulling 80w under just CPU or GPU load. Just like the apple M2 max, or the ryzen 7800hswhateveritscalled.

 

[dj-electric]

Well, two ideas.

1 - chasing a graph leader. We see it every day with GPUs and CPUs pushed well out of the efficiency curve for 1-3% gains. No reason it would be different here.

2- it's a TDP, not measured power draw. Just because the A config is 80w does not mean the CPU itself draws 80w. Remember it has a big iGPU in there too, and that is included in that TDP. We will have to wait and see, but its likely the 80w model isnt pulling 80w under just CPU or GPU load. Just like the apple M2 max, or the ryzen 7800hswhateveritscalled.

Running contemporary chips at a render test like multicore Cinebench will squeeze their defined TDP / power curve dry. It rarely doesn't. The only thing that can stop it is running deep into thermal throttling

Maybe SD X Elite works differently, but either way, this is not a complimenting look efficiency wise. I would hope that a chip design made for such significantly higher TDP would be able to actually use it and convert it into significantly higher performance. From all shown results so far, this is not the case. Not in single core, not in multicore, not in graphics compute.

 

[Assimilator]

I'm willing to bet good money that actual tests of actual devices will show this all to be complete BS. Wouldn't be the first time that an Arm chip vendor has straight up lied about their iGPU performance, hell Apple does it regularly.

 

[mplayerMuPDF]

There's all sorts of apps and interfaces in manufacturing, medical, media production, and financial sectors that is strictly windows only. Sure, it may not even be MOST apps today, but in business it is 100% or 0%, none of this "well it works well enough" stuff some consumers choose to accept.

And how much of that manufacturing and medical software is running on brand new thin and light laptops? The vast majority of it is probably running on some old desktop. You say that the "media production" and financial software is Windows only; how exactly is that a problem? We are not talking about switching to some new OS here. They just need to recompile and release a new version in most cases. In some cases some NEON/SVE focused optimization may be necessary.

You can poo-poo gamers all you want, but they make up a significant portion of remaining PC users. Steam alone has 120 million monthly active users (not total users). That 9% of the global windows population, not just consumer. Consumer only, it's likely far higher. And they spend big money.

The average user who does everything via web browser has already started jumping to chromebooks, mac OS, linux, or most commonly, phones and tablets, which make up an ever growing % of total online devices. The current PC industry is made up of people who need legacy support. You may not value it, but clearly there's 1.4 billion users who do.

And that is exactly why MS is pushing Qualcomm, AMD and Nvidia to release ARM SoCs to compete with those because they are worried about bleeding users to new MacBooks (and to a lesser extent perhaps Chromebooks). And I doubt that the average user is switching to Linux. As for the people who are "switching" to phones/tablets; I think that most who are going to do so would have done so already and those who will in the future are mostly those who still have an old computer that they haven't upgraded in years, so they don't matter from the perspective of the PC/SoC manufacturers and MS anyway.

If it didnt matter, then windows RT, windows 10 S, and windows on ARM would have become major successes. Instead they have all, so far, failed dramatically. More importantly, if you dont need legacy apps, WHY USE WINDOWS AT ALL? MS is well aware (or at least it used to be) that legacy IS its market, without legacy support windows will die out, that's why they've backed down from any attempt to replace the win32 API.

Because Windows is what people are typically used to and because it is the default. Some people don't even understand the distinction between Windows and the hardware of their computer; Windows is just a standard component of an appliance to them. And it is not clear to me that macOS is clearly superior to Windows at this point; people will have their preferences and biases but that is not the same as objective superiority of one over the other. ChromeOS arguably has some advantages but that is because it is dumbed down and locked down and that comes with disadvantages too. If MS understands the situation that they are in so well according to you, then surely there must be a good reason for them pushing Qualcomm, AMD and Nvidia to launch ARM SoCs aimed at Windows according to Reuters sources?