Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
It has eight cores in total, including two prime cores and six performance cores, and it is able to reach peak CPU speeds of 4.32 GHz.

The prime cores can reach 4.32Ghz, but the performance cores max out at 3.5Ghz.

I don’t think I would call this a mobile chip. It’s just a totally different configuration - there do not appear to be any efficiency cores, for instance.

Meanwhile, the A18 Pro, being an actual mobile chip, has 4 efficiency cores and only 2 performance cores. It’s mostly efficiency cores.

So this is a bizarre comparison and doesn’t really work out in Qualcomm’s favour. You would expect their maximum performance/zero efficiency chip to compare favourably in performance terms to Apple’s efficiency-focussed chip. Anything else would be a catastrophic embarrassment.
 
  • Like
Reactions: NetMage
Smartphone chips are still not fully utilized; I can only see these mega chips (which will no noubt ask for a premium) being useful to the very hard-core mobile gamers.

My fairly old and battered iPhone 13 can still perform every task I ask it to, very fast.
 
  • Like
Reactions: applepotato666
I would love for these companies to double down on efficiency. 45% improved performance is nice and all but I think I'm not alone in saying 45% less power usage would be much more appreciated.
I agree it’d be appreciated but it’d likely age pretty badly. You have to make sure that chip would be capable enough to run the software for no less than 6 years and any performance tradeoff just limits you from competing in software.
 
I have stated repeatedly that chip "design" is mostly marketing gimmick and has no impact on the final chip's performance. Everything is based on the fabbing process, which is both more important and intellectually harder than designing a chip, which is on par intellectually with ordering a pizza from Domino's.

QC may be using the N3E process, but it is an improved N3E process that TSMC adapted from fabbing the A18 pro. Of course this chip will outperform the A18 pro since the mfg process is more refined.

Apple fanatics are in denial and this is just further proof that Apple "design" is unimportant junk.

P.S. The displays on the iPhones are technology developed by Samsung/LG. Apple just does high-level specs like screen size, shape, resolution, color profile. Apple does low intelligence task. The real engineering magic is done by Samsung/LG, NOT Apple. Apple can't engineer themselves out of a plastic bag. Apple is only good at marketing and sales volume (Which they're slowly losing).
You have indeed repeatedly stated this. In that vein, I'm actually really interested in hearing your explanation as to why the Lunar Lake CPU on the exact same N3B node* as the Apple M3 gets worse performance at higher energy draws? On CB 24 Lunar Lake is about 14% slower while using double or more the energy in single core and only matches performance in MT at again double the power and is otherwise much slower, by about 17%. You can see similar stats for GB 6 and most other modern benchmarks. If you'd like to confirm these numbers for yourself, you can check the CPU analysis by NotebookCheck reviews. If everything is simply due to the foundry node, how is this possible?

*Actually, according to you it should be an improved N3B since TSMC would improved things for Lunar Lake after the M3 went into production.
 
Last edited:
  • Like
Reactions: NetMage
You have indeed repeatedly stated this. In that vein, I'm actually really interested in hearing your explanation as to why the Lunar Lake CPU on the exact same N3B node* as the Apple M3 gets worse performance at higher energy draws? On CB 24 Lunar Lake is about 14% slower while using double or more the energy in single core and only matches performance in MT at again double the power and is otherwise much slower, by about 17%. You can see similar stats for GB 6 and most other modern benchmarks. If you'd like to confirm these numbers for yourself, you can check the CPU analysis by NotebookCheck reviews. If everything is simply due to the foundry node, how is this possible?

*Actually, according to you it should be an improved N3B since TSMC would improved things for Lunar Lake after the M3 went into production.

That will be Intel's crappy microarchitecture and poorly designed ISA! At least they're talking to AMD now to try and work together to do something about it.

The killer on the Apple CPUs is the absolute immense ROB and the ridiculous instruction compression available on ARM (and the compiler isn't *******, because Apple spent a good decade making sure it wasn't).
 
If all of this is true, we knew it was a matter of time that the competition would catch up. Even Intel is making great strides with moving x86 closer to ARM (although they're still quite a ways off). Apple Silicon truly revolutionized the industry, and Apple should get a lot of credit for it. Although I'm quite critical of Apple in a lot of aspects, the M3 15" MacBook Air is the best laptop I've ever owned, and in my opinion the best laptop ever built.
They haven’t caught up, though. Apple still has the performance/efficiency crown. Everyone else including Qualcomm increases performance by producing a more power hungry processor.
 
So, will this chip be in the S25 Ultra? As it stands, the Samsung S24 Ultra (the best performing phone from the Android sales leader in the States) has the equivalent performance of the 12 Pro Max on the single core score and 14 Pro on the multi core. Seems fair to say they are are least a couple years behind. You could argue 4 years behind.
How can they be 4 years behind when this new chip is the fastest on the market.
Apple where at least 2 years ahead about 2 years ago it was obvious chip competitors where going to catch up
 
  • Haha
Reactions: NetMage
All the evidence you need to know that the iPhone 15 Pro series had thermal issues is to look at how much time Apple spent in their keynote, touting the "totally re-engineered" thermal system of the iPhone 16 Pro series.

It is par for the course for Apple to ignore an issue, gaslight their users, and fix it in the next rev. Antennagate, jelly scroll, touch disease, you name it.
It is not evidence that the iPhone 15 Pro has thermal issues because they changed their thermal system on a redesigned phone. Apple, and every single other manufacturer in the world make changes to designs. It doesn’t mean it came from a bad product.

Apple didn’t ignore antenna gate, they addressed it.
I'll give you Jelly Scrolling but let’s be honest, it’s an exaggerated issue at worst.
Touch disease? iPhone 6 is dropped and on some occasions the IC connector dislodges. Apple are suddenly responsible for people dropping phones?

Altering facts doesn’t make it true.

I pretty much ignore anyone who claims things have "thermal issues". People think a device getting warm is overheating when it's just natural operation because they don't realize how intensive some mundane tasks actually are.
Totally agree.
 
Being Qualcomm they are probably lying.

As per https://github.com/usefulsensors/qc_npu_benchmark they were seeing 1.3% of the quoted NPU performance on the Qualcomm ARM SoC platform 🤣
Seems like a Windows issues as it appears to perform properly on Android and "other platforms". From the link:

I'm hopeful there will be software changes, either at the application, framework, or driver level, that will improve these results in the future, since I've seen the underlying hardware perform very effectively on other platforms like Android.
 
Where did you get that information from? My iPhone 15 Pro seems to be doing fine?
Apparently it had a startup issue that caused it to heat up. Apple addressed it within a month and fixed it. It’s a non issue. Mine is fine and if the poster had one, that would be fine as well.
 
  • Like
Reactions: NetMage
The
The prime cores can reach 4.32Ghz, but the performance cores max out at 3.5Ghz.

I don’t think I would call this a mobile chip. It’s just a totally different configuration - there do not appear to be any efficiency cores, for instance.

Meanwhile, the A18 Pro, being an actual mobile chip, has 4 efficiency cores and only 2 performance cores. It’s mostly efficiency cores.

So this is a bizarre comparison and doesn’t really work out in Qualcomm’s favour. You would expect their maximum performance/zero efficiency chip to compare favourably in performance terms to Apple’s efficiency-focussed chip. Anything else would be a catastrophic embarrassment.
The performance core is also much smaller than the prime core, although I am not sure what this means for peformance/efficiency. In a way the performance core is the efficiency core.
 
Yes, but that software...
I'm not sure when the last time was that you owned an Android phone, but Android is more stable now than iOS is. Hence, my phone isn't restarting out of the blue constantly, among other bugs/issues I've read about and/or experienced first hand in iOS (I have an iPhone for my work phone). It is good for one to look outside the bubble every now and then. :)
 
  • Haha
Reactions: NetMage
The

The performance core is also much smaller than the prime core, although I am not sure what this means for peformance/efficiency. In a way the performance core is the efficiency core.
I think Qualcomm is following in Mediateks footsteps by dropping the efficiency cores because the performance cores ran at lower clocks are about as efficient and are more performant.

 
I think Qualcomm is following in Mediateks footsteps by dropping the efficiency cores because the performance cores ran at lower clocks are about as efficient and are more performant.


Qualcomm doesn't have good in-house efficiency cores, on the other hand Oryon is designed to perform well across a wide ranges of wattages. Sounds like a reasonable thing to do in their situation.
 
My big concern is how much power this Snapdragon 8 Elite will take. I wouldn't be surprised that Samsung had to go to a 5,400 mAh battery for the upcoming Galaxy S25 Ultra.
 
The prime cores can reach 4.32Ghz, but the performance cores max out at 3.5Ghz.

I don’t think I would call this a mobile chip. It’s just a totally different configuration - there do not appear to be any efficiency cores, for instance.

Meanwhile, the A18 Pro, being an actual mobile chip, has 4 efficiency cores and only 2 performance cores. It’s mostly efficiency cores.

So this is a bizarre comparison and doesn’t really work out in Qualcomm’s favour. You would expect their maximum performance/zero efficiency chip to compare favourably in performance terms to Apple’s efficiency-focussed chip. Anything else would be a catastrophic embarrassment.
I think they overclocked it to be able to claim superiority, but when running in Smartphones it will be clocked at least 0.5 GHz lower. But I am used to everything from Qualcomm to be a hype, that doesn't deliver in the end. They probably want the headlines of beating Apple and nobody cares later, when actual independent tests are ready.
 
I think they overclocked it to be able to claim superiority, but when running in Smartphones it will be clocked at least 0.5 GHz lower. But I am used to everything from Qualcomm to be a hype, that doesn't deliver in the end. They probably want the headlines of beating Apple and nobody cares later, when actual independent tests are ready.
Is that not what Apple do with there chips anyway.
 
  • Disagree
  • Like
Reactions: UliBaer and NetMage
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.