• 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 173 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Meteor Late said:
ARM standard core is further behind that it may seem at first glance, because, while the performance and efficiency difference is small (in favor of Oryon), there is a decent difference in area, Oryon V2 is quite a bit smaller.
So we have two factors that lead me to believe Qualcomm will be ahead:
-Good difference in area, there is much more potential to be realized by Qualcomm there
-Qualcomm has really good talented Apple engineers with the Nuvia team.

In fact, Apple P core is like 40% larger or so than Oryon V2, depending on which source I've checked. It wouldn't surprise me if Qualcomm catches Apple in single core in not much time.
Click to expand...
Oryon v3 is their next chance. Got some time to make it count!
Of course then it’s going up against M6/A20.
 
FlameTail said:
M5/A19
Click to expand...
Maybe. But the leaked Dell chart has it posted in late 2027. So that’s what I’ve used; maybe their (and the other OEMs) cadence is just slower for laptops.

But even if so (or is just in phones), it should be interesting to see how it plays out.
 
jdubs03 said:
Maybe. But the leaked Dell chart has it posted in late 2027. So that’s what I’ve used; maybe their (and the other OEMs) cadence is just slower for laptops.

But even if so (or is just in phones), it should be interesting to see how it plays out.
Click to expand...
You have to understand that Oryon and Snapdragon X roadmaps are different.

Oryon CPUCoresProducts
Gen 1PhoenixX Elite
Gen 2Phoenix-L
Phoenix-M		8 Elite
Ride Elite
Cockpit Elite
Gen 3Pegasus-L
Pegasus-M		8 Elite G2
X Elite G2

Snapdragon X Elite G1Snapdragon X Elite G2
CPUOryon Gen 1Oryon Gen 3
Announcement DateSnapdragon Summit 2023 (October 2023)Snapdragon Summit 2025
(October 2025)
Release DateJune 2024January 2026 (?)
 
Well I do blame them for how they’ve released gen 1 and gen 2. Only several months apart with such a difference in performance. Not good.

If your chart comes to fruition then at least they will have finally synced up their laptop and phone core generations.

And then according to that, then Dell will be the one that is releasing on a slower cadence, to me it wouldn’t make sense to wait eight or nine months after v3 comes out.
 
Last edited:
FlameTail said:
Unreal Engine 5 Nanite didn't work on Arc Alchemist iirc.

Does Arc Battlemage fix this?
Click to expand...
Well, that's one aspect where Adreno 8 matches Battlemage.
SNAPDRAGON-8-ELITE-1-1200x668.jpg
For the Snapdragon 8 Elite's Adreno 830 GPU, Qualcomm explicitly declares support for Unreal Engine 5 and it's features such as Nanite and Chaos.

Seems like Adreno 8 might be a decent GPU architecture for PCs. The Snapdragon X Elite Gen 2, which will debut next year, will use it.

chipsandcheese.com

The Snapdragon X Elite’s Adreno iGPU

Qualcomm is no stranger to integrated graphics.
chipsandcheese.com chipsandcheese.com

Adreno 7 was comparable to Nvidia Maxwell.

Maxwell -> Pascal -> Turing -> Ampere -> Ada Lovelace -> Blackwell

If so, Adreno 8 must be around Pascal/Turing level.
 
FlameTail said:
Well, that's one aspect where Adreno 8 matches Battlemage.
View attachment 113091
For the Snapdragon 8 Elite's Adreno 830 GPU, Qualcomm explicitly declares support for Unreal Engine 5 and it's features such as Nanite and Chaos.

Seems like Adreno 8 might be a decent GPU architecture for PCs. The Snapdragon X Elite Gen 2, which will debut next year, will use it.

chipsandcheese.com

The Snapdragon X Elite’s Adreno iGPU

Qualcomm is no stranger to integrated graphics.
chipsandcheese.com chipsandcheese.com

Adreno 7 was comparable to Nvidia Maxwell.

Maxwell -> Pascal -> Turing -> Ampere -> Ada Lovelace -> Blackwell

If so, Adreno 8 must be around Pascal/Turing level.
Click to expand...
Alchemist can run UE5 just "fine". It's slower than Battlemage because it doesn't natively support execute indirect for example.
 
Tup3x said:
Alchemist can run UE5 just "fine". It's slower than Battlemage because it doesn't natively support execute indirect for example.
Click to expand...
- UE5 needs Execute Indirect.
- Adreno 8 supports UE5
- Therefore, Adreno 8 has Execute Indirect?

Adreno 8 might be architecturally comparable to Intel Xe2, minus Xe2's superior RT and XMX cores.

We won't really know until someone does an architectural deep dive into Adreno 8, but that's probably not going to happen anytime soon.
 
FlameTail said:
- UE5 needs Execute Indirect.
- Adreno 8 supports UE5
- Therefore, Adreno 8 has Execute Indirect?
Click to expand...
Support for certain API extensions is also possible through software rather than µArch.

This was actually how ARM Mali supported tesselation (and geometry?) shaders earlier in their design evolution before the µArch supported the required features natively.
 
Hesperax said:
The ARM vs Qualcomm Trial starts on Monday (Jury Selection was today). Will we see a settlement over the weekend? I think the Apple vs Qualcomm had its settlement the morning of the trial.
Click to expand...
I hope for everybody’s sake there is. The clock is ticking!
 
Hesperax said:
The ARM vs Qualcomm Trial starts on Monday (Jury Selection was today). Will we see a settlement over the weekend? I think the Apple vs Qualcomm had its settlement the morning of the trial.
Click to expand...
How long will this trial take? Will it be done in a few days or drag out over years?
 
FlameTail said:
How long will this trial take? Will it be done in a few days or drag out over years?
Click to expand...
It's a five day trial (Monday to Friday). I am not sure if Jury deliberations will be postponed till after the Holiday or not. If they are not, I would expect the Jury deliberations to be very quick.

Whatever the decision, I would expect either side to Appeal which could take years to resolve definately.
 
Hesperax said:
Rumors for 8 Gen 4 Clock Frequency: Meeco

Not sure if I have seen the GPU Clock yet.

CPU ClockGPU Clock
8650AB (Snapdragon 8 Gen 3)3.3 GHz903 MHz
8650AC (Snapdragon 8 Gen 3 FG)3.4 GHz1000 MHz
8750AB (Snapdragon 8 Gen 4)4.37 GHz1150 MHz
8750AC (Snapdragon 8 Gen 4 FG)4.47 GHz1250 MHz
Click to expand...
8750AB's GPU clock speed in retail devices is 1100 MHz, which is 50 MHz lower than the leaked number. So that implies that 8750AC will also be 50 MHz lower, making it 1200 MHz.

Would it be possible to confirm by taking a look at the Geekbench 6 JSON? Galaxy S25 listings with 8750AC have been popping up;

samsung SM-S936B - Geekbench

Benchmark results for a samsung SM-S936B with an ARM ARMv8 processor.
browser.geekbench.com
 
Hitman928 said:
Intel’s new CEO throwing shade at QC over their high Windows laptops return rate.

"I mean, if you look at the return rate for Arm PCs, you go talk to any retailer, their number one concern is 'I get a large percentage of these back,' because you go to set them up and the things that we just expect do not work, right,"
Click to expand...

I don't think Intel is currently in any condition to criticise other for software not working right on their hardware.
 
Was reading this blogpost:

devblogs.microsoft.com

D3D12 Preview: Mesh Nodes in Work Graphs - DirectX Developer Blog

Hi everyone, Today we’re excited to announce the newest Preview Agility SDK v1.715.0, download here. Since this is a preview SDK, the new features are not in final state and thus only available via Developer Mode within Windows (see the prerequisites section for more). This preview serves as a...
devblogs.microsoft.com devblogs.microsoft.com
Qualcomm: Future support is planned.
Click to expand...
Qualcomm plans to support Mesh Nodes and Work Graphs in a future Adreno architecture.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top