Qualcomm Announces Snapdragon 865+: Breaking the 3GHz Threshold (anandtech.com) 38
Today, Qualcomm is announcing an update to its extremely successful Snapdragon 865 SoC: the new Snapdragon 865+. AnandTech reports: The new Snapdragon 865+ is a new binned variant of the [Snapdragon 865] with higher peak frequencies on the part of the "prime" CPU as well as the GPU, promising +10% performance on both aspects. Whilst in relative terms the new chipset's +10% clock improvement isn't all that earth-shattering, in absolute terms it finally allows the new Snapdragon 865+ to be the first mobile SoC to break past the 3GHz threshold, slightly exceeding that mark at a peak 3.1GHz frequency. Ever since the Cortex-A75 generation we've seen Arm make claims about their CPU microarchitectures achieving such high clock frequencies -- however in all those years actual silicon products by vendors never really managed to quite get that close in commercial mass-production designs.
We've had a chat with Qualcomm's SVP and GM of mobile business Alex Katouzian, about how Qualcomm achieved this, and fundamentally it's a combination of aggressive physical design of the product as well as improving manufacturing yields during the product's lifecycle. Katouzian explained that they would have been able to achieve these frequencies on the vanilla Snapdragon 865 -- but they would have had a lower quantity of products being able to meet this mark due to manufacturing variations. Yield improvements during the lifecycle of the Snapdragon 865 means that the company is able to offer this higher frequency variant now. [...] There will be a power increase to reach the higher frequencies, however this will only be linear with the increased clock speed, meaning energy efficiency of the new SoC will maintain the same excellent levels of that of the Snapdragon 865, so battery life will not be affected. [...] Amongst other new novelties of the Snapdragon 865+ platform is the ability for vendors to bundle with the new FastConnect 6900 Wi-Fi chips from Qualcomm, the company's new Wi-Fi 6 chipsets with 6GHz band capability (Wi-Fi 6E).
We've had a chat with Qualcomm's SVP and GM of mobile business Alex Katouzian, about how Qualcomm achieved this, and fundamentally it's a combination of aggressive physical design of the product as well as improving manufacturing yields during the product's lifecycle. Katouzian explained that they would have been able to achieve these frequencies on the vanilla Snapdragon 865 -- but they would have had a lower quantity of products being able to meet this mark due to manufacturing variations. Yield improvements during the lifecycle of the Snapdragon 865 means that the company is able to offer this higher frequency variant now. [...] There will be a power increase to reach the higher frequencies, however this will only be linear with the increased clock speed, meaning energy efficiency of the new SoC will maintain the same excellent levels of that of the Snapdragon 865, so battery life will not be affected. [...] Amongst other new novelties of the Snapdragon 865+ platform is the ability for vendors to bundle with the new FastConnect 6900 Wi-Fi chips from Qualcomm, the company's new Wi-Fi 6 chipsets with 6GHz band capability (Wi-Fi 6E).
Still (Score:1)
Probably slower than Apple.
Re: (Score:1)
Fails To Impress (Score:2, Insightful)
Re: (Score:2, Troll)
Re: (Score:2)
Mmm... I'd like to wait and see on that one. The 865 came already pretty close to the A13. I don't care too much about synthetic tests, especially with two so different processors, I'll wait and see how they're being put to use.
Re: (Score:3, Insightful)
Re: (Score:3)
Not a big deal. The main problem with the A13 is that you have to buy an apple product to get it. No thanks.
Why is this being down voted? I agree completely, and it's a completely factual statement.
Re: (Score:2)
Not a big deal. The main problem with the A13 is that you have to buy an apple product to get it. No thanks.
Why is this being down voted?
Because the Cult of Mac will tolerate no blasphemy.
Re: (Score:2)
Re:Fails To Impress (Score:4, Informative)
It's way faster than the A13 Bionic. The Apple chip only has two high performance cores, this one has four.
As can be seen in these benchmarks even the old Snapdragon 865 beats the A13 in multithreaded performance: https://nanoreview.net/en/soc-... [nanoreview.net]
In AnTuTu it was 14% faster than the A13. With the clock boost in the 865+ that lead will be extended.
Or did you mean just in specific benchmarks that favour single threaded performance?
Re: (Score:2)
Twice the cores and yet only 14% faster?
Re: (Score:2)
More cores has advantages in properly multitasking operating systems like Android. So it is better to have more lower performance cores instead of two slightly higher performance ones.
There are also power advantages too. Even Apple has 4 low power cores.
Raspberry Pi in 202x? (Score:2)
So the current Pi 4 has a 1.5GHz CPU. How many years before it gets the equivalent of this one? And what would the price be if they built it around one of these now?
Re: (Score:2, Informative)
You can always buy a faster arm board. There are dozens out there.
Re: (Score:3)
Apple's chips are faster yet clock slower (Score:5, Interesting)
This is a great example of selling what you've got. Apple's chips are an order of magnitude more performant at half the clock speed of the qualcomm chips, yet we have an article showing how great the snapdragon is.
The fact is, Qualcomm sucks donkey balls when it comes to improving the snapdragon chipset. If I was an android OEM I'd be happy that they're the only game in town because everyone in the phone market sucks equally.
In fact, the only differentiating factor across android is OS support. And that consists of "does this phone get to upgrade at all?" or "does this get security updates?"
Re: (Score:2)
Re: (Score:3)
As does the New Oxford American Dictionary, among others. The term has been around for decades.
Re: (Score:2)
Re: (Score:2)
Performant isnt a word
The New Oxford American Dictionary disagrees with you. The term has been around since the ‘80s and dictionaries started including it a few years ago, just like they started including words like “blog”, “automobile”, and “computer” at some point. The language is a moving target, whether we like it or not.
You cant buy Apples chips and pair them with your own coprocessors and you cant buy Apples coprocessors and pair them with a snapdragon.
And if I was Google or Samsung or HTC, that’d be a valid point, but most of us are not Google or Samsung or HTC. End users care about what you can do with a pr
Re: (Score:3)
Google already makes their own chips in other fields and is looking to make their own chips here as well because Qualcomm hasnâ(TM)t been competitive enough. Samsung has the expertise and may go that route eventually too.
Eh? Samsung has ALREADY gone that route. Outside the US (where CDMA support is not needed), Samsung use their own Exynos chips, not Snapdragon.
Re: (Score:2)
I clearly haven’t kept up very well. I thought they had backed off from Exynos and was suggesting it might be time for a return. My apologies for the mistake.
Re: (Score:2)
I clearly havenâ(TM)t kept up very well. I thought they had backed off from Exynos and was suggesting it might be time for a return. My apologies for the mistake.
LOL OK I think i may have kept up even less well than you! I didn't realise they backed off Exynos!
Re: (Score:2)
You had me convinced I was wrong about them backing off on Exynos, but now I'm just confused.
*goes to check*
Annnnnnd...Wikipedia is no help. Looks like they're still releasing Exynos-based devices, just like you suggested, but I don't recognize the model numbers, so I'm not sure if any of these are actually noteworthy devices (maybe they're just European models, like you suggested?). It also looks like they laid off their Austin-based R&D team working on Exynos related stuff, so I have no idea what to m
Re: (Score:2)
Yeah WTF is going on?
Sounds like they might be sticking with Exynos chips but they might be abandoning their "Mongoose" custom core in favour of stock ARM ones which have been improving. On the other hand it's Samsung who's strategy is to basically pick every strategy at once then randomly select one after a period of time.
Given they have the capability I would guess that there are really significant advantages to Exynos over Snapdragon since they can do the integration of precisely everything they want, in
Re: (Score:2)
You're an order of magnitude away from being half right.
Re: Apple's chips are faster yet clock slower (Score:2)
The 865 is comparable to the A13 when using synthetic benchmarks. If you squint and use a Qualcomm reference device it will even beat the A13 on some benchmarks. The 865+ will probably handily beat it on most actual OEM devices.
The A13, on top of iOS will certainly beat the 865, and probably the 865+, in real world benchmarks but it wonâ(TM)t be an order of magnitude better by any means.
Re: (Score:2)
Agreed. OP vastly overstated Apple’s lead. Apple does still have a lead, but it’s shrunk quite a bit recently, though that doesn’t excuse Qualcomm apologists trying to claim the 865 is better than the A13 by holding up the reference device you mentioned, which had performance that was never even close to being matched by the most popular 865-based production devices.
Re: (Score:3)
Re: (Score:2)
Literally never. What I always say is that I wish it had a faster GPU. But then, I say that about pretty much everything.
Re: (Score:2)
An order of magnitude more performant at half the clock speed? You mean 14% slower at 6% lower clock speed, right?
https://nanoreview.net/en/soc-... [nanoreview.net]
Snapdragon 865 vs A13 Bionic
2840 MHz vs 2660 MHz
Geekbench 5 (Multi-Core) 3504 vs 3501
AnTuTu 8 551863 vs 483094
And that's the old one, not the 865+ which we don't have benchmarks for yet.
Re: (Score:2)
Apple's chips are an order of magnitude more performant at half the clock speed of the qualcomm chips
What, they're 10 times faster running at 1.5 GHz? But Apple's recent chips run at ~2.6 GHz...and they're *not* 10 times faster. You've made that up.
Threshold (Score:2)
Well, thank goodness they didn't call it a "barrier". It is a pet peeve of mine when some new record or milestone is surpassed. "In new Guinness record, man break largest ball-of-string barrier!" No: that is not a barrier, in the way the sound barrier or a coulomb barrier is a real, physical obstacle that needs to be overcome, on the other side of which is something truly different. Another qualifier for the term barrier, in my opinion, is that it is independent of units an