• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Discussion Qualcomm Snapdragon Thread

Page 153 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
To be more efficient, E cores are workhorse of Soc in mobile, P cores are just for burst perfomance, 6 E cores would definetely help reducing power usage compare to 4 E cores.

More efficient only when all cores are being used, though, and that's his point here, that almost nothing in the real world is going to leverage all these cores in mobile.
 
IMG-20241022-WA0008.jpg
SoCFP32 ALUsClock Speed3DMark Steel Nomad Light
8 Elite15361.15 GHz2600 points
D940015361.6 GHz2600 points
A18 Pro7681.5 GHz2100 points

This means the Adreno 830 has better "performance-per-clock" than the Immortalis G925 MP12 in the Dimensity 9400, which means the architecture is better. But the "performance-per-ALU" is still not as good as A18 Pro, which has half the ALU count.
 
Last edited:
Main thing I want to see is single core efficiency, as multi core efficiency is going to be better by simply having more cores and reducing frequency to raise efficiency.
 
This means the Adreno 830 has better "performance-per-clock" than the Immortalis G925 MP12 in the Dimensity 9400, which means the architecture is better. But the it's still not as good as A18 Pro, which has half the ALU count.
Your table states that 8 Elite PPC is 2600/1.6 while G925 is 2600/1.15. That would suggest PPC of G925 is better. Unless you have a mistake in your table.
 
The unrelenting march of Oryon CPU generational advancement has only just begun. Gen 1 set the bar in PC. Gen 2; advance the bar in peak performance, halve the power, introduce a new tier of performance core and maintain industry leading prime core area efficiency.

Gen 2 sets the foundation of the next generation of Oryon. Gen 3...er, you'll have to wait and see. 😉

Below are two salient slides from the keynote:

1729546409962.png
1729547493133.png
 
The unrelenting march of Oryon CPU generational advancement has only just begun. Gen 1 set the bar in PC. Gen 2; advance the bar in peak performance, halve the power, introduce a new tier of performance core and maintain industry leading prime core area efficiency.

Gen 2 sets the foundation of the next generation of Oryon. Gen 3...er, you'll have to wait and see. 😉

Below are two salient slides from the keynote:

View attachment 109923
View attachment 109929
A little humiliating comparing a phone SoC's benchmarks to a laptop's... so I guess we'll be seeing this in laptops then? 🙂
 
GB6 ST

A18 Pro = 3500
A19 Pro = 4000 (speculation)

I think there is a high possibility that 8 Elite Gen 2 will exceed the ST performance of A19 Pro.
 
GB6 ST

A18 Pro = 3500
A19 Pro = 4000 (speculation)

I think there is a high possibility that 8 Elite Gen 2 will exceed the ST performance of A19 Pro.
The release of the M4 has already proved that Apple will not give up. In an interview with Geekerwan, Johny Srouji said that their goal is the best ST performance in the industry. But in any case, competition is good.
 
Sorry guys but I want to see what Qualcomm is preparing for the 7, 6 and 4 series... 8 Elite looks great, but I want to see the 7 Plus.
 
GB6 ST

A18 Pro = 3500
A19 Pro = 4000 (speculation)

I think there is a high possibility that 8 Elite Gen 2 will exceed the ST performance of A19 Pro.
Not impossible, but Qualcomm is already using frequency headroom that Apple is still expanding into. Snapdragon remains behind despite having much faster, higher bandwidth memory and higher clocks.

I think the clocks will be higher on future nodes in addition to whatever architecture tweaks may come. And Apple now has more room to turn that frequency knob than Qualcomm.
 
A little humiliating comparing a phone SoC's benchmarks to a laptop's... so I guess we'll be seeing this in laptops then? 🙂
Qualcomm faces a serious uphill battle in laptops. They can claim single thread and battery leadership, but it is all by not much. Any AI leadership will also be by not much.

The gaps the M1 held was often jaw dropping. Oryon is impressive, but is no head turning feature. Both Strix Point and Lunar Lake are very snappy with long battery life. They both have better GPUs and Strix is the multithreaded leader.

If you see someone excited about a Strix laptop, what can you say to redirect them to Qualcomm? Barely perceptible extra snappyness? 15 hours of battery life instead of 12 1/2 hours? The ability to generate an article summary in 8.2 seconds instead of 8.8 seconds?

What would compel a Snapdragon purchase?
 
Not impossible, but Qualcomm is already using frequency headroom that Apple is still expanding into. Snapdragon remains behind despite having much faster, higher bandwidth memory and higher clocks.

I think the clocks will be higher on future nodes in addition to whatever architecture tweaks may come. And Apple now has more room to turn that frequency knob than Qualcomm.
Agreed and certainly not in performance per clock. There is still quite a gap to narrow.
My guess the Oryon Gen 3 core will be on ARMv9.2 so maybe that might help claw back some. And they’ll likely focus more on IPC/PPC next time around since they’re already at 4.32GHz in a phone.
 
Qualcomm faces a serious uphill battle in laptops. They can claim single thread and battery leadership, but it is all by not much. Any AI leadership will also be by not much.

The gaps the M1 held was often jaw dropping. Oryon is impressive, but is no head turning feature. Both Strix Point and Lunar Lake are very snappy with long battery life. They both have better GPUs and Strix is the multithreaded leader.

If you see someone excited about a Strix laptop, what can you say to redirect them to Qualcomm? Barely perceptible extra snappyness? 15 hours of battery life instead of 12 1/2 hours? The ability to generate an article summary in 8.2 seconds instead of 8.8 seconds?

What would compel a Snapdragon purchase?
Lower SoC manufacture cost and PCB complexity: better screen, build, connectivity iso-cost... Gamers are just 15% of PC users; whomever would have been impressed by Strix Pt. would have been among them and not the market Windows on ARM is initially targetting...
 
Agreed and certainly not in performance per clock. There is still quite a gap to narrow.
My guess the Oryon Gen 3 core will be on ARMv9.2 so maybe that might help claw back some. And they’ll likely focus more on IPC/PPC next time around since they’re already at 4.32GHz in a phone.
The genius is that they're MORE efficient at those clocks!! Nothing wrong w/ hitting 5Ghz as long as you're ultimately more efficient...
 
Back
Top