Wednesday, July 17th 2024

Qualcomm Snapdragon X "Copilot+" AI PCs Only Accounted for 0.3% of PassMark Benchmark Runs

The much-anticipated revolution in AI-powered personal computing seems to be off to a slower start than expected. Qualcomm's Snapdragon X CPUs, touted as game-changers in the AI PC market, have struggled to gain significant traction since their launch. Recent data from PassMark, a popular benchmarking software, reveals that Snapdragon X CPUs account for a mere 0.3% of submissions in the past 30 days. This is a massive contrast to the 99.7% share held by traditional x86 processors from Intel and AMD, which raises questions about the immediate future of ARM-based PCs. The underwhelming adoption comes despite bold predictions from industry leaders. Qualcomm CEO Cristiano Amon had projected that ARM-based CPUs could capture up to 50% of the Windows PC market by 2029. Similarly, ARM's CEO anticipated a shift away from x86's long-standing dominance.

However, it turns out that these PCs are primarily bought for the battery life, not their AI capabilities. Of course, it's premature to declare Arm's Windows venture a failure. The AI PC market is still in its infancy, and upcoming mid-tier laptops featuring Snapdragon X Elite CPUs could boost adoption rates. A lot of time still needs to pass before the volume of these PCs reaches millions of units shipped by x86 makers. The true test will come with the launch of AMD's Ryzen AI 300 and Intel's Lunar Lake CPUs, providing a clearer picture of how ARM-based options compare in AI performance. As the AI PC landscape evolves, Qualcomm faces mounting pressure. NVIDIA's anticipated entry into the market and significant performance improvements in next-generation x86 processors from Intel and AMD pose a massive challenge. The coming months will be crucial in determining whether Snapdragon X CPUs can live up to their initial hype and carve out a significant place in the AI PC ecosystem.
Sources: PassMark (on X), via Wccftech
Add your own comment

28 Comments on Qualcomm Snapdragon X "Copilot+" AI PCs Only Accounted for 0.3% of PassMark Benchmark Runs

#1
AusWolf
ARM in PC, and the whole AI thing was always going to be a failure, as long as there isn't any benefit to a Regular Joe. Is "know your customer base" not a thing anymore?
Posted on Reply
#2
clopezi
AusWolfARM in PC, and the whole AI thing was always going to be a failure, as long as there isn't any benefit to a Regular Joe. Is "know your customer base" not a thing anymore?
AI would be, for sure. ARM can be a game changer for laptops, x86 has been failing for years to make responsible battery use and proper suspension states.
Posted on Reply
#3
randomUser
Is passmark really the way to tell that ARM failed? lol no. Just because consumer doesn't care about passmark, doesn't mean nobody buys the laptops.
0,3% is alot, considering they just released. 99,99% of people don't even run benchmarks on PCs they buy, so those 50+ runs are most likely done by the reviewers and enthusiasts.
Posted on Reply
#4
JWNoctis
clopeziAI would be, for sure. ARM can be a game changer for laptops, x86 has been failing for years to make responsible battery use and proper suspension states.
Arguably not quite a real problem, when both major x86 CPU manufacturer's mobile offerings idle well below 5W and with the cores fully parked and powered down, unless we are talking about tablets and netbooks. But it's true that more battery life is always welcome.

Either way, the figure's probably not as meaningful without a number detailing how often users of either types run PassMark, a specific system benchmark among numerous choices.
Posted on Reply
#5
john_
Passmark is not the most accurate way to get market share numbers. Maybe desktops, where people might run Passmark's benchmark, by I doubt users of laptop/server systems care or even know Passmark's benchmarks.
AusWolfARM in PC, and the whole AI thing was always going to be a failure, as long as there isn't any benefit to a Regular Joe. Is "know your customer base" not a thing anymore?
Battery life is an important reason for people constantly on the go. With more apps getting native ARM versions and compatibility improving, I see ARM getting a good slice of windows market in the near future. Also some big OEMs like Dell seem to be friendlier to Qualcomm than AMD, so I am expecting them to promote ARM based laptops more than AMD based laptops. That will also help ARM in the laptop market. And that's before Mediatek getting in the game, meaning cheaper ARM based laptops and Nvidia getting in the game, meaning more powerful in gaming ARM based laptops.
I can see a 20% market share by 2027.
Posted on Reply
#6
AusWolf
john_Battery life is an important reason for people constantly on the go. With more apps getting native ARM versions and compatibility improving, I see ARM getting a good slice of windows market in the near future. Also some big OEMs like Dell seem to be friendlier to Qualcomm than AMD, so I am expecting them to promote ARM based laptops more than AMD based laptops. That will also help ARM in the laptop market. And that's before Mediatek getting in the game, meaning cheaper ARM based laptops and Nvidia getting in the game, meaning more powerful in gaming ARM based laptops.
I can see a 20% market share by 2027.
Battery life is an important factor you can sell a laptop by, I'll give you that. But AI isn't, especially when there is no way any home user could or would want to use AI for anything.
Posted on Reply
#7
Eternit
Windows Phone had better market share and MS killed it. I wouldn't mind to buy a small tablet with ARM and Windows, but only if it was cheaper than with Processor (formerly Celeron). I don't care about copilot, but about time on battery.
Posted on Reply
#8
john_
AusWolfBattery life is an important factor you can sell a laptop by, I'll give you that. But AI isn't, especially when there is no way any home user could or would want to use AI for anything.
While Microsoft and Qualcomm will be pushing AI in marketing, many consumers will be going in the store asking simply for a windows laptop with the best battery life. Many of them will be ending with an ARM based laptop at hand.
AI marketing affects as mostly, by breaking our nerves for now, but it might sell to typical consumers who will be amazed by it, even not knowing what exactly is.

Another thing I remembered from a Linus sponsored review is the quality of the webcam on ARM based laptops. That could also sell a few.
Posted on Reply
#9
AusWolf
john_While Microsoft and Qualcomm will be pushing AI in marketing, many consumers will be going in the store asking simply for a windows laptop with the best battery life. Many of them will be ending with an ARM based laptop at hand.
AI marketing affects as mostly, by breaking our nerves for now, but it might sell to typical consumers who will be amazed by it, even not knowing what exactly is.

Another think I remembered from a Linus sponsored review is the quality of the webcam on ARM based laptops. That could also sell a few.
Consumer: "Do you have a laptop with good battery life?" :)
Seller: "Yes, we have this one with a Qualcomm CPU, it's amazing on battery." :rolleyes:
Consumer: "That's awesome! What else does it have?" :D
Seller: "It's got AI!" :cool:
Consumer: "AI as in artificial intelligence? Wow! What does it do?" o_O
Seller: "Ehm... well... it can... search the web for (inaccurate) answers, I guess?" :slap: :roll:
Posted on Reply
#10
JWNoctis
john_While Microsoft and Qualcomm will be pushing AI in marketing, many consumers will be going in the store asking simply for a windows laptop with the best battery life. Many of them will be ending with an ARM based laptop at hand.
AI marketing affects as mostly, by breaking our nerves for now, but it might sell to typical consumers who will be amazed by it, even not knowing what exactly is.

Another thing I remembered from a Linus sponsored review is the quality of the webcam on ARM based laptops. That could also sell a few.
Remember when they sold "64-bit" Pentium 4 systems installed with 32-bit Windows XP, back when there was zero driver support for a lot of things if you actually went 64-bit? At least AI actually does something.

That's marketing for you. :oops:
Posted on Reply
#11
AusWolf
JWNoctisAt least AI actually does something.
Like what?
Posted on Reply
#12
JWNoctis
AusWolfLike what?
As opposed to a capability entirely unused and at the time unusable, yet still marketed like it's some secret sauce.

But point conceded. :p
Posted on Reply
#13
Ferrum Master
I read on Phoronix adreno drivers are getting blacklisted currently, it is a such a mess code wise.

Go on... after 5-10 years :D
Posted on Reply
#14
phanbuey
the core issue here is that this is a great "surf the internet" machine that costs $1300-$1500.
Posted on Reply
#15
Neo_Morpheus
AusWolfARM in PC
I think that should be a thin and we have to give credit to Pi’s on this.
Heck, Pi foundation should offer a “Pro” option to help advance the market.
clopeziuse and proper suspension states.
I wonder who's fault is it here, hardware or MS?
I remember a video done by LTT about how bad Windows was in this compared to Macs.
JWNoctisArguably not quite a real problem, when both major x86 CPU manufacturer's mobile offerings idle well below 5W and with the cores fully parked and powered down,
See above and will add, running intensive task, I wonder how far behind x86 currently is, compared to these Qualcomm chips.
john_Also some big OEMs like Dell seem to be friendlier to Qualcomm than AMD
Dell is still taking bribes from Intel to keep AMD away, so I wonder what is their bribe arrangement with qualcomm and MS in this case.
JWNoctisRemember when they sold "64-bit" Pentium 4 systems installed with 32-bit Windows XP, back when there was zero driver support for a lot of things if you actually went 64-bit? At least AI actually does something.

That's marketing for you. :oops:
Actually, was AMD Athlon 64 that was released first and MS screwed them by withholding the release of XP 64, to give intel time to release their version of the AMD64 extensions. Did try to find the source of the article explaining this, but it was a long time ago and since search sucks these days…
AusWolfLike what?
Amazing cats images of course! :)

Posted on Reply
#16
Lew Zealand
Snapdragon X CPUs account for a mere 0.3% of submissions in the past 30 days. This is a massive contrast to the 99.7% share held by traditional x86 processors from Intel and AMD, which raises questions about the immediate future of ARM-based PCs. The underwhelming adoption comes despite bold predictions from industry leaders. Qualcomm CEO Cristiano Amon had projected that ARM-based CPUs could capture up to 50% of the Windows PC market by 2029.
So after 1.8% of the time predicted it's 0.6% of the way there. As these things are never linear, that seems a reasonable market penetration in all of a month. But of course that kind of analysis rarely grabs headlines and eyeballs so we get the copy above.
Posted on Reply
#17
persondb
Lew ZealandSo after 1.8% of the time predicted it's 0.6% of the way there. As these things are never linear, that seems a reasonable market penetration in all of a month. But of course that kind of analysis rarely grabs headlines and eyeballs so we get the copy above.
The thing is that there were thousands to ten of thousand of units shipped to reviewers. And reviewers are the kind of users that are heavily biased into running benchmark software such as Passmark, which likely mean that the current rate is over represented due to the large portion of reviewers. This happen in a lot of product cycles, specially ones that have more limited stock, i.e. paper launch or didn't have good initial sales(think similarly if it was the MSI claw, wouldn't it be more likely that the benchmarks are done by reviewers considering that people barely brought into it?).

So this does paint a very bad look into early market penetration and that in my opinion it seems like it's not selling well.

There might also be an argument that early users might also be more biased to run benchmark software due to this being the 'new class of hardware'(windows on arm) that promised great performance. There is no proof for that though, but you might think it makes sense. If true to some extent, it could make up for an even worse picture.
Posted on Reply
#18
phanbuey
Lew ZealandSo after 1.8% of the time predicted it's 0.6% of the way there. As these things are never linear, that seems a reasonable market penetration in all of a month. But of course that kind of analysis rarely grabs headlines and eyeballs so we get the copy above.
New products usually see an initial spike in first adopters - unless it flops like the apple VR headset. This is a hard flop.

If you look at most 'new' tech that's good it moves alot of initial volume. The initial intel macs and the M1 sold like hotcakes, ipod, iphones, AMD Zen (couldn't keep mobos in stock), the list goes on. Very rarely does a product ease into market penetration without that initial excitement - this is more like an Intel ARC or a microsoft Zune.

I have some friends who work at dell and best buy -- the sales on these are abysmal, and the return rates are also very high. I don't see this gaining share.
Posted on Reply
#19
Minus Infinity
Forget the AI BS, it's the lack of native software or compatibility that's the real problem, even if you ignore prices.

Not a chance in hell I would buy into ARM based computer for at least 2-3 years until we see most software running natively. Not interested in emulation rubbish. Softeware companies need to get off their lazy arses and port programs over.
Posted on Reply
#20
uuee
I think Pat was more reasonable about the future of ARM when he said that Intel will win back Apple.
Posted on Reply
#21
Dr. Dro
Other than some low-cost Compaq Presario-branded laptops with the Snapdragon 7c, I haven't seen anything ARM let alone Copilot+ being sold here at all. Maybe if it was available people would give it a try.
Posted on Reply
#22
Lew Zealand
persondbThe thing is that there were thousands to ten of thousand of units shipped to reviewers. And reviewers are the kind of users that are heavily biased into running benchmark software such as Passmark, which likely mean that the current rate is over represented due to the large portion of reviewers. This happen in a lot of product cycles, specially ones that have more limited stock, i.e. paper launch or didn't have good initial sales(think similarly if it was the MSI claw, wouldn't it be more likely that the benchmarks are done by reviewers considering that people barely brought into it?).

So this does paint a very bad look into early market penetration and that in my opinion it seems like it's not selling well.

There might also be an argument that early users might also be more biased to run benchmark software due to this being the 'new class of hardware'(windows on arm) that promised great performance. There is no proof for that though, but you might think it makes sense. If true to some extent, it could make up for an even worse picture.
phanbueyNew products usually see an initial spike in first adopters - unless it flops like the apple VR headset. This is a hard flop.

If you look at most 'new' tech that's good it moves alot of initial volume. The initial intel macs and the M1 sold like hotcakes, ipod, iphones, AMD Zen (couldn't keep mobos in stock), the list goes on. Very rarely does a product ease into market penetration without that initial excitement - this is more like an Intel ARC or a microsoft Zune.

I have some friends who work at dell and best buy -- the sales on these are abysmal, and the return rates are also very high. I don't see this gaining share.
Bunch of assumptions and guesses there but no data. I'll stick with the data and we'll see how it goes over the next 6 months or so.

I don't think it'll proceed very quickly because WOA doesn't have the money behind it that Apple did for the transitions you mention but you can also use Apple to argue just the opposite: The transition from OS 9 to OS X was very slow in penetration and happened while Apple was at an ebb and many saw bankruptcy looming. All the decks were stacked against success. I certainly didn't think they were gonna go anywhere and it took a long time before they did. But they did.
Posted on Reply
#23
ARF
JWNoctisArguably not quite a real problem, when both major x86 CPU manufacturer's mobile offerings idle well below 5W and with the cores fully parked and powered down, unless we are talking about tablets and netbooks. But it's true that more battery life is always welcome.
5 watts is extremely high. Should be less than a watt. 0.5 watts.
Neo_MorpheusI wonder who's fault is it here, hardware or MS?
Both. The CISC x86-64. And MS.
I know intel wanted a brand new architecture without the legacy instructions...
Posted on Reply
#24
Dr. Dro
Lew ZealandBunch of assumptions and guesses there but no data. I'll stick with the data and we'll see how it goes over the next 6 months or so.

I don't think it'll proceed very quickly because WOA doesn't have the money behind it that Apple did for the transitions you mention but you can also use Apple to argue just the opposite: The transition from OS 9 to OS X was very slow in penetration and happened while Apple was at an ebb and many saw bankruptcy looming. All the decks were stacked against success. I certainly didn't think they were gonna go anywhere and it took a long time before they did. But they did.
So slow that it had to be performatic, remember the OS 9 funeral lol

Steve Jobs is truly missed
Posted on Reply
#25
phanbuey
Lew ZealandBunch of assumptions and guesses there but no data. I'll stick with the data and we'll see how it goes over the next 6 months or so.

I don't think it'll proceed very quickly because WOA doesn't have the money behind it that Apple did for the transitions you mention but you can also use Apple to argue just the opposite: The transition from OS 9 to OS X was very slow in penetration and happened while Apple was at an ebb and many saw bankruptcy looming. All the decks were stacked against success. I certainly didn't think they were gonna go anywhere and it took a long time before they did. But they did.
I mean the data right now is bad - that's the point. It's certainly missed expectations at an OEM that knows the data.... I think 6 months is a bit soon - end of 2025 when they release v2 I think it has a good chance, but the initial release was way overhyped.
Posted on Reply
Add your own comment
Oct 7th, 2024 19:59 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts