Discussion Qualcomm Snapdragon Thread

Page 17 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

gdansk

Platinum Member
Feb 8, 2011
2,894
4,383
136
Don't listen to him, don't let marketing misuse the word ultimate. They've already ruined Pro(fessional).
 

Tigerick

Senior member
Apr 1, 2022
695
601
106

Dell actually sells a Snapdragon laptop. This is surprising.

What's even better is that for $500, you are getting an 8cx Gen 3.
Half of price of ThinkPad X13s, this is real cost of Qualcomm 8cx Gen3.

I am afraid X Elite will price above $1K initially due to OEMs positioning ...
 

FlameTail

Diamond Member
Dec 15, 2021
3,881
2,315
106
Half of price of ThinkPad X13s, this is real cost of Qualcomm 8cx Gen3.

I am afraid X Elite will price above $1K initially due to OEMs positioning ...
Hmm. Idk. That Inspiron 14 is a much lower end laptop than the Thinkpad X13s.

That and the fact that this Inspiron 14 seems to be on 'sale'. Dell does stuff like this. For instance you can buy the entry level SKU of the XPS 13 for just $699, which is a discount from the original $999 price.

Also the 8cx Gen 3 is a 2 year old chip by this point, on an older and cheaper node at that. Undoubtedly, the X Elite would be costlier.
 

Tup3x

Golden Member
Dec 31, 2016
1,080
1,080
136
My AMD Ryzen 7 PRO 4750U work laptop is well past three years so technically it should have been replaced already. However... Changing it now would be a bit meh since much better options are going to be released than what's currently available.

If X Elite laptops turn out great, I'd definitely consider those. Git for Windows is basically in beta but I'd normally use WSL2 Ubuntu anyway... It would be nice if Google drive had ARM application since that makes backuping easier. Other than that all tools/random software that I would use have native builds available.

By the way, LibreOffice has native builds now. GIMP too.
 
Reactions: Tlh97 and Thibsie

FlameTail

Diamond Member
Dec 15, 2021
3,881
2,315
106
I doubt he has any control over the naming anyway. He's an engineering VP, marketing controls product naming not the guys who design the products.
Even if he isn't directly involved in the naming, he might atleast some influence in the matter. Anyways, he could pass on the feedback to the relevant people.
 
Last edited:

FlameTail

Diamond Member
Dec 15, 2021
3,881
2,315
106
My AMD Ryzen 7 PRO 4750U work laptop is well past three years so technically it should have been replaced already. However... Changing it now would be a bit meh since much better options are going to be released than what's currently available.

If X Elite laptops turn out great, I'd definitely consider those. Git for Windows is basically in beta but I'd normally use WSL2 Ubuntu anyway... It would be nice if Google drive had ARM application since that makes backuping easier. Other than that all tools/random software that I would use have native builds available.

By the way, LibreOffice has native builds now. GIMP too.
LibreOffice has native ARM version? That's cool. So the Officd software is covered then (MS Office is also native).
 

FlameTail

Diamond Member
Dec 15, 2021
3,881
2,315
106

This chart counts both dGPUs and iGPUs.

We know Nvidia only sells dGPUs, but they only have 18% marketshare!

That means the majority of PCs don't have a dGPU.

So tell me, is it really that important that X Elite should pair with a dGPU?
 

SpudLobby

Senior member
May 18, 2022
976
670
106
View attachment 93009
First A18 leak is here, and it puts Oryon in a bad light.

Oryon CPU in 8G4 does ~2800 in GB6 ST.

If this A18 leak is true, then A18 will have a comprehensive 25% ST performance advantage over 8G4 Oryon.

What's the use of the expensive Oryon custom core project if they are not able to atleast come close to Apple's ST performance?
Lmao, I straight up don’t believe Apple is going to get a 20% boost on N3 just going N3B to N3E with a tweak. The A17 already was the new core generation that as rumored from years back by Dylan and while I think they will certainly change a few things, they are at 3.78GHz and on N3 already.

Parametric yield imrpovements on N3E might allow them some very slightly lower power or possibly some extra frequency iso-power but I doubt it’s much and the power is already getting a bit higher than it was. The architecture iso-frequency and (largely) iso-node could use more power if it’s larger as well without necessarily smarter tweaks.

No I don’t think the A18 is hitting 3500 GB6 at the same power as the A17 Pro and I don’t think it’ll even be that high regardless.

Would bet on another year of very incremental improvement if not outright low single digit gains on ST and a focus on improved (power too) efficiency in some ways or another.m
 

Doug S

Platinum Member
Feb 8, 2020
2,751
4,684
136
Do we know that A17 really was a major core upgrade? Given the very lackluster performance gains (almost all from frequency increase) either it wasn't (and perhaps A18 might be where we see that) or it was but the initial version was mostly laying the groundwork for future improvements that couldn't be made on N3B (in other words, maybe the redesign relies on FinFlex to be able to place faster transistors in certain blocks)

Like I said before I'm highly skeptical of the A18 leak just because of how early in the year it is, but when AMD is claiming 15% or greater IPC gains with Zen 5, and most aren't too skeptical of those claims, why would it be hard to believe Apple could have similar IPC gains, especially when they have not been gaining much on that front in the last few years and could be seen as "overdue" for bigger IPC impacting core changes.
 

SpudLobby

Senior member
May 18, 2022
976
670
106
Do we know that A17 really was a major core upgrade? Given the very lackluster performance gains (almost all from frequency increase) either it wasn't (and perhaps A18 might be where we see that) or it was but the initial version was mostly laying the groundwork for future improvements that couldn't be made on N3B (in other words, maybe the redesign relies on FinFlex to be able to place faster transistors in certain blocks)

Like I said before I'm highly skeptical of the A18 leak just because of how early in the year it is, but when AMD is claiming 15% or greater IPC gains with Zen 5, and most aren't too skeptical of those claims, why would it be hard to believe Apple could have similar IPC gains, especially when they have not been gaining much on that front in the last few years and could be seen as "overdue" for bigger IPC impacting core changes.
Well it was a redesign they were straightforward about unlike with the A15 and A16 where at best they mentioned some power gains for the cores alone in the A16, likely due to a new phy des. This matches (Apple mentioning it) Dylan’s leak. Major at this point is somewhat relative. There’s not a really *good* reason to believe Apple have a Zen 5-tier (rumors put Zen 5 above 15% btw) they’ll pull out of a hat one off year when they’ve lost chief talent and have taken a very incremental approach for several years now — even the A13 to A14 was mid single digits. I’ve heard this story before about the A15, the A16, then the magnitude of the A17/M3 changes. Mind you, Apple updates more frequently than Intel and AMD, so this adds up, and even the A14 -> A16 had some nonzero IPC gain probably partly from greater L2.


Anyway, I do think the main difference is that AMD’s core structures are slightly undersized relative to the performance they have thanks to smarter prefetching and prediction: there’s more obvious room for scaled up improvement and they haven’t lost chief architects like Apple did.



I think it’s actually possible that N3E improved yields, a new physical design tweak with FinFLEX (I’ve thought the same thing, too, Doug) in tandem with another architectural improvement bring another 5-10% in total maybe, I just wouldn’t bet on 20% given where they already are with power (higher than ever, sort of pushing it even with improved efficiency), frequency, architectural size, node.
 
Reactions: Tlh97

Doug S

Platinum Member
Feb 8, 2020
2,751
4,684
136
I think it’s actually possible that N3E improved yields, a new physical design tweak with FinFLEX (I’ve thought the same thing, too, Doug) in tandem with another architectural improvement bring another 5-10% in total maybe, I just wouldn’t bet on 20% given where they already are with power (higher than ever, sort of pushing it even with improved efficiency), frequency, architectural size, node.


"Given where they already are with power" - that assumes A17's higher power draw is because of design issues, and not N3B's yield issues. Apple may have been forced through a combination of poor yields and TSMC's "known good die" pricing to have to use some number of marginal dies that are "working" only if more power than Apple would normally prefer is delivered to them. Dies which Apple normally would discard, since good yields allow you to bin not just on "working" but "working within acceptable power ranges".

Theoretically Apple could test every die and figure out the voltage it requires - the best ones may work at around half the voltage of the worst, and have each use only what is necessary to function at the standard clock speed. But then you'd have a situation where some people would get iPhones with noticeably better battery life / heat than others. To my knowledge however, the only time that's happened was with A9 when Apple was dual sourcing from TSMC and Samsung.

So rather than do the extra work of characterizing each die like Intel/AMD would (which pays for itself because they can sell the ones that operate at very low voltages / higher clocks at a premium as "U" class / top speed bins) and risk customer ire (and people returning perfectly good phones hoping to get one with a "golden" SoC) they may set them all to the "worst case" voltage. That would cause all A17s to use more power than they would have had N3B not been a broken process. Since N3E is supposed to have typical TSMC yields, they won't have to do that with A18 so I think there's a good chance its power efficiency will be more like A14/A15/A16 than A17, regardless of whether its performance improves by 5% or 20%.
 
Reactions: lightmanek

eek2121

Diamond Member
Aug 2, 2005
3,100
4,398
136
View attachment 93451
This chart counts both dGPUs and iGPUs.

We know Nvidia only sells dGPUs, but they only have 18% marketshare!

That means the majority of PCs don't have a dGPU.

So tell me, is it really that important that X Elite should pair with a dGPU?
That is a garbage chart and most certainly does NOT represent market share. The PC TAM is far too large for there to be quarter to quarter swings like that. They are likely measuring units sold, instead.

Also the issue isn’t about dGPU vs APU, it is about compatibility. Intel found out all about this when they launched their GPUs. Writing compatible, functional, and performant GPU drivers is hard work. IIRC Most QC GPUs have been limited to OpenGL ES support. Qualcomm faces the daunting task of needing to support DX12U, Vulkan, and OpenGL on an unproven platform (WoA), and I suspect they are not ready and that day one drivers will be a joke.

GPUs are also not just used for gaming. AI, content creation, etc. also use the GPU (the NPU isn’t a replacement for many of these workloads, including AI)

Windows uses hardware acceleration for rendering, and even uses the GPU to accelerate virus scans.
 

FlameTail

Diamond Member
Dec 15, 2021
3,881
2,315
106
"Qualcomm Adreno GPUs have better gaming (graphics) performance, but Apple GPUs have better compute (GPGPU) performance".

How true is this statement?
 

trivik12

Senior member
Jan 26, 2006
331
305
136
I am willing to bet this is -profanity removed-. Last year also at this point we had all kinds of crazy rumors around A17. No way they are bumping 20%. There are limits to increasing clock speed for mobile soc and its already 9-wide core with tons of cache. Considering SRAM is at N5 level size than N3B, there is no way they will increase the cache.

Profanity is not allowed in the tech forums.

Daveybrat
AT Moderator
 
Last edited by a moderator:

FlameTail

Diamond Member
Dec 15, 2021
3,881
2,315
106
I am willing to bet this is bullshit. Last year also at this point we had all kinds of crazy rumors around A17. No way they are bumping 20%. There are limits to increasing clock speed for mobile soc and its already 9-wide core with tons of cache. Considering SRAM is at N5 level size than N3B, there is no way they will increase the cache.
Sir this is Qualcomm thread
 

mikegg

Golden Member
Jan 30, 2010
1,835
459
136
"Qualcomm Adreno GPUs have better gaming (graphics) performance, but Apple GPUs have better compute (GPGPU) performance".

How true is this statement?
It's not true that Adreno GPUs are better at gaming. iOS games usually target higher resolutions, better textures, and more advanced effects. You can't compare FPS to FPS between iOS/Android.

The fact that Apple GPUs are better at compute should already tell you that they're also just straight up more powerful.
 
Reactions: Tlh97

FlameTail

Diamond Member
Dec 15, 2021
3,881
2,315
106
It's not true that Adreno GPUs are better at gaming. iOS games usually target higher resolutions, better textures, and more advanced effects. You can't compare FPS to FPS between iOS/Android.

The fact that Apple GPUs are better at compute should already tell you that they're also just straight up more powerful.
It is possible to pull up an objective comparison using benchmarks?

Because in graphics benchmarks, Snapdragon comes out on top.




Now, cab somebody find the compute benchmarks?
 
Last edited:

FlameTail

Diamond Member
Dec 15, 2021
3,881
2,315
106
Hey. Did we forget that Qualcomm also makes SoCs for AR/VR headsets?

How does the latest Snapdragon XR2+ Gen 2 compare to M2/R1 in Apple Vision Pro?
 

FlameTail

Diamond Member
Dec 15, 2021
3,881
2,315
106
My prediction for laptops battery life in late 2024:

Apple > Qualcomm > AMD = Intel

The lack of E-cores, and a less efficient GPU would result in X Elite not reaching the efficiency level of the latest Apple M3.

However, it would still be a bit ahead of AMD/Intel.

_____

Do you call bull on the prediction?
 

FlameTail

Diamond Member
Dec 15, 2021
3,881
2,315
106

This is the Samaung Galaxy S24 Ultra's motherboard. (Taken from This video).

Look at how ridiculously compact it is.

This thing contains the WiFi, Bluetooth, RAM (upto 24 GB*), Storage (upto 1 TB), a bunch of other chips and an SoC as powerful as the M1.

Yes, the Snapdragon 8 gen 3 is as strong as the legendary M1.

8 Gen 3
• CPU : Geekbench 6 = 2400/7400
• GPU : 3DMARK WLE = 32 FPS
• Memory bandwidth = 76 GB/s

M1
• CPU : Geekbench 6 = 2300/8300
• GPU : 3DMARK WLE = 30 FPS
• Memory bandwidth = 68 GB/s

Now imagine a Windows on ARM laptop with this motherboard and SoC. It would have a 99 Wh battery (thanks to all the space savings from the compact motherboard).

The result would be a laptop that is;

• Thin and light
• Has insane battery life
• 5G connectivity
• Ultimate portability and mobility (due to above factors)

It would be perfect for students.
 
Reactions: Tlh97

FlameTail

Diamond Member
Dec 15, 2021
3,881
2,315
106
Qualcomm's Adreno GPU is interesting in that it uses a "wide and slow" approach compared to others who are doing "narrow and fast".

Exynos 2400 = 768 ALU
SD 8G3 = 1792 ALU
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |