How are the new Mediatek chips in comparison? The "dimensity" 1000 (dumb name for an interesting SoC, IMO) has promising specs, and, thanks to Huawei's difficulties, is apparently made on 7 nm by TSMC. Don't know about their availability on devboards, though.
Why are iPhone users iSlaves in your mind? My 4 year old iPhone still receives iOS updates how many 4 year old Android devices are getting updates to the latest version of Android? The last version of Android the Samsung Galaxy S7 supports is Android 8, which is two versions old.
Well that's just incorrect. You can easily setup sideloading even without jailbreaking using various methods. And if you are jailbroken, the iPhone jailbreak app customization scene is far more advanced than the android one. I have an android phone and tablet, as well as an iPhone. I definitely have more customization options on the iphone. Sure, the tweaks cost $$ unless you pirate them, but they're very cheap and well worth it, and is what allows the scene to flourish.
Not just iSlaves, also iIdiots. Do you need updates? Did updates make your iPhones better? Do you have choices regarding which OS version you want to have (compared to Android)? Do you have an option to downgrade? Do you feel superior for paying 3 to 4 times more than the actual manufacturing cost of an iPhone?... The list goes on but iIdiots would never understand.
Manufacturing cost of a product has absolutely nothing to do with its retail price. The most costly part of any product is R&D and marketing. You don't think the top Android manufacturers also charge several times more than the cost of production?
Yeah sure buddy lol, you seem to not know what the market is actually like right now. A 256GB iPhone 11 Pro Max costs $100 less than a 128GB Note20 Ultra... The idiot seems to be you but the miserable thing is that you don’t even know it. And I’m having a hard time to imagine for who not being downgrade the OS would be a dealbreaker... Android fanboy.
Pray tell, how? Huawei makes their own SoCs, so does samsung. Mediateks are getting better, and qualcomm is still pushing leading performance improvements every year.
Maybe you'd hav e a point if qualcomm did an intel and was sitll selling 835 performance as the 865, but that isnt the case. The current 660 series outperfoms 3 year old 800 series chips.
Indeed. This is a contestable market - meaning that even though it looks like a monopoly, it's a very precarious monopoly. Qualcomm knows that if they sit on their laurels for even a moment, another company will overtake them. So they continually release new, faster products.
More importantly, its not locked down like X86 is. Anyone can get a license and make their own ARM SoCs. If a monopoly did occur there would be more competitors popping out of the woodwork and take the opportunity.
It would be nice, but you probably won't for a long long time if ever. Rockchip, Amlogic and Allwinner are losers in advancement. They never had anything beyond Cortex-A55, because everything after A55 can't support 32-bit OS. And they are refusing/delaying to upgrade to Android 10 because of the 64-bit requirement. Can you even believe Amlogic is rolling out S922X, their premium SoC with big cores in a A53+A73 configuration, THIS YEAR, just to avoid going 64-bit.
Those companies will get to the newest stuff when they get to it. And it will be a lot cheaper than the flagship smartphones. For higher end, I'll be watching MediaTek Dimensity. There isn't a Qualcomm monopoly.
If people are still bored by ~20% gains, they will have to wait for the monolithic 3D ARM SoCs. Hopefully those land within 5 years and not 2030.
Samsung has honestly never made a competitive low-power ARM SoC. That's why they threw in the towel on mobile Exynos.
It's really a shame Apple has seemingly poached a lot of engineering talent throughout the industry to make a mobile SoC that is at the same time superior and proprietary.
It's a shame when these devices get EOLed after 2 years tops due to sealed battery designs. Why not all the phone companies create an open battery standard to give consumer free choice. Even in parallel universe it won't happen, sad fate of this industry forever leeching the consumer and making them dumbed down to consume the product every damn year for $1000.
Desktop machines or replaceable components laptops like MXM and rPGA/LGA live for more than a decade, thanks to the modularity and the mature Windows / Linux x86 ecosystem.
Modern batteries have improved dramatically, I have a 5-year old Galaxy S6 which has a very healthy battery despite being charged every other day. Even if your battery dies in less than 5 years (unlikely), there are plenty of shops which can replace it. So there is no need to make batteries user-replaceable.
Batteries aren't replaceable anymore because consumers preferred thin and waterproof designs. There's a tradeoff between making the device compact and sealed, versus making it easily modified by the average end-user.
I dont prefer the thin style. I was complaining just yesterday about how fragile my phone is compared to that of about 6 years ago.
The phone I had then was dropped every few weeks and never cracked, barely scratched and screen was flawless. The fact the back plate and battery sprung from it when this happened just aided in dissipating the energy away from important components.
I miss my WP7.5 phone, I bet not many would say the same these days.
Unless you "rich" (don't feel any attachment to your toys), or have perfect and non-varying control of your body, I don't understand why someone wouldn't keep a non-ruggedized in a protective shell.
The Galaxy S20 is sealed, has no physical buttons on the front, no 3.5mm jack, and is 7.9mm thick.
Meanwhile, the S5 Active was only 1mm thicker at 8.9mm (as in "not enough you'd notice"), allowed you to easily replace the battery, had actual buttons on the front, a 3.5mm jack....and was STILL IP67 rated.
The S7/Edge has the same IP68 rating as the S20, and still manages to have a 3.5mm jack.
Conclusion: Sealed batteries and removing a standard audio connector are meant to reduce longevity or hinder easy upkeep, and force you to purchase expensive "wireless" accessories for simple things.
You forgot the fact that the S20 has a far bigger battery than both the S5 Active and S7/Edge. Sure, advancements in battery tech (though minor) are partly to thank for that, but being sealed and not having the headphone jack means more space for the battery as well. Weren't people complaining a few years back that they would rather have thicker phones with bigger batteries?
"3.5mm jacks were removed to make the phones waterproof. " Along with "passing along savings"(of a few cents, if passed along at all), "additional space for batteries"(100mah as Samsung admitted), why are these laughable lies repeated? What do you people have to gain from the jack being castrated? There's negligible benefit to the consumer and massive greed in this maneuver.
No, only a handful of devices use pull tabs against glue for example, which has nothing to do with thickness, and glue isn't even all that necessary for a battery in a sealed device.
Sounds amazing, but no idea how you managed to charge only every other day, my S6E, backup device is almost glued to the charger, and if unplugged, not used for anything except the occasional text or call, and still the battery lasts a day at best, half a day at worst and the device lags.
Stop buying expensive garbage. My moto z play was $400 5 years ago, and sported insanely long battery life. Plenty of people reviewing it said the same thing: "why would you ever need a battery life that long?"
This is why: 5 years later and deapite loosing some capacity this phone still provides 10 hours of SoT on a single charge. Most modern $1K flagships have no hope of getting that far. Meanwhile you can now buy a $250 moto g power with almost twice the battery life.
I agree that phones shouldnt be sealed, but at the same time that modularity you talk about takes up a TON of room you just dont have in a phone.
My pockets aren't large enough for my desktop or even my laptop. Come back when you can mention any positive, for consumers, effects that "sealed" designs offer.
What the heck are you talking about? Batteries are replaceable. Go to any service centre and they'll do it for a fee, or even free depending on the circumstances. Removable batteries are risky especially with superfast charge that goes up to 65W. If you want removable battery, you'd be stuck with no more than 10W and a smaller battery. You'd be insane to want that just so you can avoid paying $50 (plus cost of battery which is what you're going to pay regardless) after 1.5-2 years of use.
I am curious. Why would making a battery removable limit recharging power to 10W? In other words, why would it be unfeasible to have a battery that can be fast-charged with 65W _and_ is removable?
>As to what this means for the power and efficiency of the new Snapdragon 865+: 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.**
One more time now? A linear power increase is still a power increase.
2 W to 2 GHz 3 W to 3 GHz
Power draw goes up, battery life goes down. Unless this paragraph is missing key qualifiers, e.g.,, "and Qualcomm has increased efficiency elsewhere, so 865+ will maintain the same power draw as 865 even with an increased peak clock and increased peak CPU power draw."
It may be a *smaller* increase in power draw, but any increase in power draw not offset somewhere else = lower expected battery life of the 865+ versus the 865.
If perf/Watt doesn't change as stated, the total energy for any given task remains the same. So it will use more power but it will also finish earlier. For typical use it means it becomes more responsive with the same battery life. Battery life may actually improve a bit if it sleeps for longer (due to finishing quicker) or use a lower voltage at other frequencies.
As Wilco explained, power draw != energy usage. The 865+ will use more power than the 865, however the energy usage for a fixed workload task will be the same. Battery life on a phone will not change as long as you're doing the same tasks.
This is true, but from the battery's POV, discharge rate matters. Higher the discharge current, the greater the reduction of battery capacity, meaning the battery provides more mAH at lower current than at higher current. I'm sure you already know this. Not a battery expert but my understanding is that internal resistance of the battery plays a role. Does it matter in normal usecases? Probably not enough to notice in my use case but I am a light user. It's also possible that the phone will start throttling for thermal reasons such that this scenario (sustained 3GHz) does not occur in practice.
iPhones suffer a ~11% performance drop at low battery charge states, even on brand-new batteries, because the CPU asks for such a high voltage and/or current draw that the battery can't provide.
That is, ostensibly partially exacerbated by the CPU's high current draw, the battery degrades so much that it can't provide the power the CPU demands. But hopefully w/ just a single core demanding higher current in the 865+, it won't stress batteries as much.
This has nothing to do with unexpected shutdowns: those are a much more severe symptom that has long affected any ho-hum $10 device.
These performance variations, however, are only exhibiting on high-performance, high-consumption ARM-based flagship CPUs. The results are clearly showing that batteries have held back SoC performance.
SoC improvements have easily outpaced the rather slow improvement in Li-on battery output.
@ ikjadoon: That's part of it, yes. But in general, 1AH battery could provide 100mA for 10hrs but if you draw 1A, it may not last even 1hr. And yes, as you get to lower battery charge, it may be unable to provide sufficient voltage at the current draw required.
Then there's long-term battery life and the effect of current draw on that. A sealed degrading battery that costs >$100 to replace is a nice way of planned obsolescence.
One thing is definitely the cost, the batteries themselves cost a few bucks, if you know where to look. If you buy from service centers they'll rip you off for sure.
Good point. If since this is linear, the increase should be, at most, 8-9%. Where does does the 865 fall within the designed discharge rates of li-ion chemistry?
Ah, I see now. Of course, in the end, the power draw must be reduced somewhere else to maintain the same energy usage (i.e., battery life). Here, it's not another part, but simply running the CPU for a shorter time.
That is, a race to idle is the other side here: use slightly more peak power and then sleep for much longer. 12 W for 5 seconds (60 W-s consumed) versus 10 W for 6 seconds (60 W-s).
Faster performance at higher power draw with equal energy usage / battery life.
Though not my initial confusion, deltaFx2's does note higher discharge rates bring an amplified loss of capacity: I can only think of Apple, but hopefully Android OEMs won't fall into the same trap.
Also, the efficiency of 865+ has increased over the 865. If you try and run an 865 at 3.1GHz, it is possible, as long as the thermal limits aren't reached and if the energy required is supplied. However, whilst the early 865 chipset would have to slowly and exponentially increase the Voltage to sustain that frequency... whereas the 865+ seems to only require a linear increase in voltage.
This is the "lowest-hanging fruit" way of increasing performance, because you can simply hit higher frequencies more stably, thanks to having a more accurate silicon. This happens as yields of nodes improve during production. In the PC Gaming space they call this "binning". Regardless, this "linear increase" is better than the "exponential increase", so it is still an efficiency increase.
So you might be wondering if it's even possible for the standard 865 to overclock that high? And to show you it is, an easy example is trying to overclock the Tegra X1 processor found in the Nintendo Switch... that's what we observe on the Nvidia Shield TV, which runs with a higher power-draw and a larger cooling solution.
In order to have the power increase to be "linear with the increased clock speed", we need the cpu to be running at similar voltage range right? How is it possible to run 2.84GHz and 3.1GHz at similar voltage under the same process node and same design? Voltage has to be increased aggressively to get to 3.1GHz
If it will run at 3.1 Ghz at a given voltage then 2.84 is definitely possible at that same voltage. It's more efficient to lower the voltage at the lower speed but the cpu will work fine if you don't.
I thought some said there is no 865plus for this year ?
Why bother with such a release besides marketing hype for qualcomm and self proclaimed “gaming” phones ? Are there actualy people who would wait for this instead of a regullar dragon ?
Because the marketing hype has minimal cost? Nobody would wait for this but some models were waiting for this, e.g. ROGP3, which if anywhere near ROGP2 is a solid specialized gaming device, not "self-proclaimed". The likes of Razer, OTOH is self proclaimed.
Shouldn't those both be doubled? That's how wikichip describes them, LPDDR4X-4266 and LPDDR5-5500, and that would match the usual norms for how DRAM is described wouldn't it?
There are many MANY degrees of doubling involved. Which is why it's all a matter of convention now, pretty much.
But the convention appears to be that when I say LPDDR4X-4266, what I mean is that I will get 4.266 Gbits per sec from one chip, hence 34GB/s from eight chips. The precise details of exactly how many transactions of what width and length that translates into, and issues of "data" transactions vs "command" transactions are all mixed with the issue of DDR vs DDR2 vs DDR3 vs ..., which is precisely why we adopt a convention that focuses on the (PHY, guaranteed not to exceed) bandwidth, not how we got there.
I wonder how much performance one would gain by simply using (faster) LPDDR5 at otherwise unchanged top speed for the fast core? Which phones with SD 865 or 865+ are currently using the newer generation memory?
Why doesn't TSMC grab the Apples, the AMDs, the NVidias, the Qualcomms by their proverbial b@lls, and overnight demand outrageous payment. They'd be the biggest company ever ... for a while anyways.
Yeah, but is Samsung competitive in this area? Why are none of the big players having their silicon produced in Samsung fabs? If all the current tip-of-the-spear chips are flowing through TSMC, that seems to me to be a lot of leverage that could be exploited...
Nvidia is supposedly using Samsung 8nm for at least some of its lineup this year, maybe most of it. Supposedly, Nvidia tried to play hardball with TSMC using Samsung as leverage, but it didn't work. Samsung's 8nm node might be worse than TSMC's 7nm, but it's good enough for the purpose and cheaper.
Even if Samsung continues to trail TSMC by a bit, they are close, they are spending lots of billions to reach 5nm, 3nm, etc. I wouldn't be surprised to see AMD use Samsung for some chips, even if it's just to increase the amount of low-end chips they can sell (they were still selling some 28nm GloFo chips in 2019: A6-9220C and A4-9120C).
Instead of worrying about speed, they should add more L2 cache. All A7x cores should have at least 512KB, and all A5x cores should have at least 256KB.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
81 Comments
Back to Article
Spunjji - Wednesday, July 8, 2020 - link
The eternal tedium of the monopoly market :|abufrejoval - Wednesday, July 8, 2020 - link
I'll admit it has become a bit boring, only having Apple chips not usable unless you're an iSlave or Qualcomm as 'best in a class of one'.But I can't say I am unhappy with what they deliver... only wish I could get a dev board for €99.
eastcoast_pete - Wednesday, July 8, 2020 - link
How are the new Mediatek chips in comparison? The "dimensity" 1000 (dumb name for an interesting SoC, IMO) has promising specs, and, thanks to Huawei's difficulties, is apparently made on 7 nm by TSMC. Don't know about their availability on devboards, though.BedfordTim - Thursday, July 9, 2020 - link
HiSilicon are up there with Qualcomm, but their HiKey boards are reassuringly expensive.29a - Friday, July 10, 2020 - link
Why are iPhone users iSlaves in your mind? My 4 year old iPhone still receives iOS updates how many 4 year old Android devices are getting updates to the latest version of Android? The last version of Android the Samsung Galaxy S7 supports is Android 8, which is two versions old.lmcd - Friday, July 10, 2020 - link
Sideloading doesn't exist and thus the FOSS scene and hobbyist scene absolutely suck.iphonebestgamephone - Saturday, July 11, 2020 - link
Sideloading is possible with jailbreak tho? Or even without it using a workaround.Socius - Friday, July 24, 2020 - link
Well that's just incorrect. You can easily setup sideloading even without jailbreaking using various methods. And if you are jailbroken, the iPhone jailbreak app customization scene is far more advanced than the android one. I have an android phone and tablet, as well as an iPhone. I definitely have more customization options on the iphone. Sure, the tweaks cost $$ unless you pirate them, but they're very cheap and well worth it, and is what allows the scene to flourish.sonny73n - Sunday, July 12, 2020 - link
Not just iSlaves, also iIdiots. Do you need updates? Did updates make your iPhones better? Do you have choices regarding which OS version you want to have (compared to Android)? Do you have an option to downgrade? Do you feel superior for paying 3 to 4 times more than the actual manufacturing cost of an iPhone?... The list goes on but iIdiots would never understand.Socius - Friday, July 24, 2020 - link
Manufacturing cost of a product has absolutely nothing to do with its retail price. The most costly part of any product is R&D and marketing. You don't think the top Android manufacturers also charge several times more than the cost of production?caribbeanblue - Monday, October 26, 2020 - link
Yeah sure buddy lol, you seem to not know what the market is actually like right now. A 256GB iPhone 11 Pro Max costs $100 less than a 128GB Note20 Ultra... The idiot seems to be you but the miserable thing is that you don’t even know it. And I’m having a hard time to imagine for who not being downgrade the OS would be a dealbreaker... Android fanboy.TheinsanegamerN - Wednesday, July 8, 2020 - link
Pray tell, how? Huawei makes their own SoCs, so does samsung. Mediateks are getting better, and qualcomm is still pushing leading performance improvements every year.Maybe you'd hav e a point if qualcomm did an intel and was sitll selling 835 performance as the 865, but that isnt the case. The current 660 series outperfoms 3 year old 800 series chips.
Mikewind Dale - Wednesday, July 8, 2020 - link
Indeed. This is a contestable market - meaning that even though it looks like a monopoly, it's a very precarious monopoly. Qualcomm knows that if they sit on their laurels for even a moment, another company will overtake them. So they continually release new, faster products.TheinsanegamerN - Friday, July 10, 2020 - link
More importantly, its not locked down like X86 is. Anyone can get a license and make their own ARM SoCs. If a monopoly did occur there would be more competitors popping out of the woodwork and take the opportunity.lmcd - Friday, July 10, 2020 - link
Modems are locked down.nandnandnand - Wednesday, July 8, 2020 - link
And if you look more broadly, there is Rockchip, Allwinner, etc. in SBCs.It will be neat to see Cortex-X1 come to SBCs, which should routinely clock above 3 GHz.
dotjaz - Wednesday, July 8, 2020 - link
It would be nice, but you probably won't for a long long time if ever. Rockchip, Amlogic and Allwinner are losers in advancement. They never had anything beyond Cortex-A55, because everything after A55 can't support 32-bit OS. And they are refusing/delaying to upgrade to Android 10 because of the 64-bit requirement.Can you even believe Amlogic is rolling out S922X, their premium SoC with big cores in a A53+A73 configuration, THIS YEAR, just to avoid going 64-bit.
nandnandnand - Wednesday, July 8, 2020 - link
Rockchip RK3588 looks pretty good to me.Those companies will get to the newest stuff when they get to it. And it will be a lot cheaper than the flagship smartphones. For higher end, I'll be watching MediaTek Dimensity. There isn't a Qualcomm monopoly.
If people are still bored by ~20% gains, they will have to wait for the monolithic 3D ARM SoCs. Hopefully those land within 5 years and not 2030.
Samus - Thursday, July 9, 2020 - link
Samsung has honestly never made a competitive low-power ARM SoC. That's why they threw in the towel on mobile Exynos.It's really a shame Apple has seemingly poached a lot of engineering talent throughout the industry to make a mobile SoC that is at the same time superior and proprietary.
kozz_man - Wednesday, July 8, 2020 - link
Any word on the upcoming Samsung Galaxy Note sporting this new proc?Quantumz0d - Wednesday, July 8, 2020 - link
It's a shame when these devices get EOLed after 2 years tops due to sealed battery designs. Why not all the phone companies create an open battery standard to give consumer free choice. Even in parallel universe it won't happen, sad fate of this industry forever leeching the consumer and making them dumbed down to consume the product every damn year for $1000.Desktop machines or replaceable components laptops like MXM and rPGA/LGA live for more than a decade, thanks to the modularity and the mature Windows / Linux x86 ecosystem.
Wilco1 - Wednesday, July 8, 2020 - link
Modern batteries have improved dramatically, I have a 5-year old Galaxy S6 which has a very healthy battery despite being charged every other day. Even if your battery dies in less than 5 years (unlikely), there are plenty of shops which can replace it. So there is no need to make batteries user-replaceable.Sharma_Ji - Wednesday, July 8, 2020 - link
Yeah, and battery replacements are cheap as hell, one from BBK costs under 20$ for 8 pro.Could get one replaced every year.
TheinsanegamerN - Wednesday, July 8, 2020 - link
Good luck removing the old one from your tared and glued phone. Batteries should absolutely be a 30 second job.Mikewind Dale - Wednesday, July 8, 2020 - link
Batteries aren't replaceable anymore because consumers preferred thin and waterproof designs. There's a tradeoff between making the device compact and sealed, versus making it easily modified by the average end-user.jimbo2779 - Wednesday, July 8, 2020 - link
I dont prefer the thin style. I was complaining just yesterday about how fragile my phone is compared to that of about 6 years ago.The phone I had then was dropped every few weeks and never cracked, barely scratched and screen was flawless. The fact the back plate and battery sprung from it when this happened just aided in dissipating the energy away from important components.
I miss my WP7.5 phone, I bet not many would say the same these days.
tuxRoller - Wednesday, July 8, 2020 - link
Unless you "rich" (don't feel any attachment to your toys), or have perfect and non-varying control of your body, I don't understand why someone wouldn't keep a non-ruggedized in a protective shell.FSWKU - Wednesday, July 8, 2020 - link
The Galaxy S20 is sealed, has no physical buttons on the front, no 3.5mm jack, and is 7.9mm thick.Meanwhile, the S5 Active was only 1mm thicker at 8.9mm (as in "not enough you'd notice"), allowed you to easily replace the battery, had actual buttons on the front, a 3.5mm jack....and was STILL IP67 rated.
The S7/Edge has the same IP68 rating as the S20, and still manages to have a 3.5mm jack.
Conclusion: Sealed batteries and removing a standard audio connector are meant to reduce longevity or hinder easy upkeep, and force you to purchase expensive "wireless" accessories for simple things.
Retycint - Wednesday, July 8, 2020 - link
You forgot the fact that the S20 has a far bigger battery than both the S5 Active and S7/Edge. Sure, advancements in battery tech (though minor) are partly to thank for that, but being sealed and not having the headphone jack means more space for the battery as well. Weren't people complaining a few years back that they would rather have thicker phones with bigger batteries?s.yu - Thursday, July 9, 2020 - link
Castrating the jack, as Samsung admitted, adds 100mah to the battery.https://www.notebookcheck.net/Samsung-gives-its-re...
s.yu - Thursday, July 9, 2020 - link
Totally agree!29a - Friday, July 10, 2020 - link
3.5mm jacks were removed to make the phones waterproof.Myrandex - Friday, July 10, 2020 - link
lies, plenty of phones had water resistance with them, see the Galaxy S8, S9, and S10.s.yu - Tuesday, July 14, 2020 - link
"3.5mm jacks were removed to make the phones waterproof. "Along with "passing along savings"(of a few cents, if passed along at all), "additional space for batteries"(100mah as Samsung admitted), why are these laughable lies repeated? What do you people have to gain from the jack being castrated? There's negligible benefit to the consumer and massive greed in this maneuver.
s.yu - Thursday, July 9, 2020 - link
No, only a handful of devices use pull tabs against glue for example, which has nothing to do with thickness, and glue isn't even all that necessary for a battery in a sealed device.dotjaz - Wednesday, July 8, 2020 - link
Nope. We have people doing that for us.iphonebestgamephone - Friday, July 10, 2020 - link
A 5 minute job at the local service center every 2 years is somehow too much?s.yu - Thursday, July 9, 2020 - link
Sounds amazing, but no idea how you managed to charge only every other day, my S6E, backup device is almost glued to the charger, and if unplugged, not used for anything except the occasional text or call, and still the battery lasts a day at best, half a day at worst and the device lags.TheinsanegamerN - Wednesday, July 8, 2020 - link
Stop buying expensive garbage. My moto z play was $400 5 years ago, and sported insanely long battery life. Plenty of people reviewing it said the same thing: "why would you ever need a battery life that long?"This is why: 5 years later and deapite loosing some capacity this phone still provides 10 hours of SoT on a single charge. Most modern $1K flagships have no hope of getting that far. Meanwhile you can now buy a $250 moto g power with almost twice the battery life.
I agree that phones shouldnt be sealed, but at the same time that modularity you talk about takes up a TON of room you just dont have in a phone.
zepi - Wednesday, July 8, 2020 - link
Moto Z play was released in September 2016 (announced a month earlier).That was 3 years 10 months ago.
TheinsanegamerN - Friday, July 10, 2020 - link
You're right, I thought it was 2015. Still it's nearly 4 years old with constant use and is holding up really well.Fritzkier - Wednesday, July 8, 2020 - link
Yeah, I really don't understand the appeal of buying flagship product if what you want is better battery life.Many mid range phones have a huge battery size, especially Xiaomi and Samsung M-series. M41 even include 7000mah battery, and that's pretty huge.
TheinsanegamerN - Friday, July 10, 2020 - link
Samsung M series and Xaiomi are sadly not sold in the US' backwards mobile marketplace.tuxRoller - Wednesday, July 8, 2020 - link
My pockets aren't large enough for my desktop or even my laptop.Come back when you can mention any positive, for consumers, effects that "sealed" designs offer.
dotjaz - Wednesday, July 8, 2020 - link
What the heck are you talking about? Batteries are replaceable. Go to any service centre and they'll do it for a fee, or even free depending on the circumstances. Removable batteries are risky especially with superfast charge that goes up to 65W. If you want removable battery, you'd be stuck with no more than 10W and a smaller battery.You'd be insane to want that just so you can avoid paying $50 (plus cost of battery which is what you're going to pay regardless) after 1.5-2 years of use.
MrVibrato - Thursday, July 9, 2020 - link
I am curious. Why would making a battery removable limit recharging power to 10W?In other words, why would it be unfeasible to have a battery that can be fast-charged with 65W _and_ is removable?
ikjadoon - Wednesday, July 8, 2020 - link
>As to what this means for the power and efficiency of the new Snapdragon 865+: 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.**One more time now? A linear power increase is still a power increase.
2 W to 2 GHz
3 W to 3 GHz
Power draw goes up, battery life goes down. Unless this paragraph is missing key qualifiers, e.g.,, "and Qualcomm has increased efficiency elsewhere, so 865+ will maintain the same power draw as 865 even with an increased peak clock and increased peak CPU power draw."
It may be a *smaller* increase in power draw, but any increase in power draw not offset somewhere else = lower expected battery life of the 865+ versus the 865.
Wilco1 - Wednesday, July 8, 2020 - link
If perf/Watt doesn't change as stated, the total energy for any given task remains the same. So it will use more power but it will also finish earlier. For typical use it means it becomes more responsive with the same battery life. Battery life may actually improve a bit if it sleeps for longer (due to finishing quicker) or use a lower voltage at other frequencies.Andrei Frumusanu - Wednesday, July 8, 2020 - link
As Wilco explained, power draw != energy usage. The 865+ will use more power than the 865, however the energy usage for a fixed workload task will be the same. Battery life on a phone will not change as long as you're doing the same tasks.deltaFx2 - Wednesday, July 8, 2020 - link
This is true, but from the battery's POV, discharge rate matters. Higher the discharge current, the greater the reduction of battery capacity, meaning the battery provides more mAH at lower current than at higher current. I'm sure you already know this. Not a battery expert but my understanding is that internal resistance of the battery plays a role. Does it matter in normal usecases? Probably not enough to notice in my use case but I am a light user. It's also possible that the phone will start throttling for thermal reasons such that this scenario (sustained 3GHz) does not occur in practice.ikjadoon - Wednesday, July 8, 2020 - link
A fair point: we actually have a real-world example, i.e, Apple's battery & CPU throttling.https://www.ifixit.com/News/9472/ios-update-slows-...
iPhones suffer a ~11% performance drop at low battery charge states, even on brand-new batteries, because the CPU asks for such a high voltage and/or current draw that the battery can't provide.
That is, ostensibly partially exacerbated by the CPU's high current draw, the battery degrades so much that it can't provide the power the CPU demands. But hopefully w/ just a single core demanding higher current in the 865+, it won't stress batteries as much.
GC2:CS - Wednesday, July 8, 2020 - link
No matter which phone or battery you use you can discharge it to a level where you will see a “performance drop” called a shutdown.Nobody is complaining about that.
ikjadoon - Wednesday, July 8, 2020 - link
This has nothing to do with unexpected shutdowns: those are a much more severe symptom that has long affected any ho-hum $10 device.These performance variations, however, are only exhibiting on high-performance, high-consumption ARM-based flagship CPUs. The results are clearly showing that batteries have held back SoC performance.
SoC improvements have easily outpaced the rather slow improvement in Li-on battery output.
deltaFx2 - Thursday, July 9, 2020 - link
@ ikjadoon: That's part of it, yes. But in general, 1AH battery could provide 100mA for 10hrs but if you draw 1A, it may not last even 1hr. And yes, as you get to lower battery charge, it may be unable to provide sufficient voltage at the current draw required.https://learn.adafruit.com/all-about-batteries/pow... (first link on a google search, there may be better references) Discharge rate adversely affects the energy delivered by the battery.
Then there's long-term battery life and the effect of current draw on that. A sealed degrading battery that costs >$100 to replace is a nice way of planned obsolescence.
s.yu - Thursday, July 9, 2020 - link
One thing is definitely the cost, the batteries themselves cost a few bucks, if you know where to look. If you buy from service centers they'll rip you off for sure.tuxRoller - Wednesday, July 8, 2020 - link
Good point.If since this is linear, the increase should be, at most, 8-9%.
Where does does the 865 fall within the designed discharge rates of li-ion chemistry?
ikjadoon - Wednesday, July 8, 2020 - link
Ah, I see now. Of course, in the end, the power draw must be reduced somewhere else to maintain the same energy usage (i.e., battery life). Here, it's not another part, but simply running the CPU for a shorter time.That is, a race to idle is the other side here: use slightly more peak power and then sleep for much longer. 12 W for 5 seconds (60 W-s consumed) versus 10 W for 6 seconds (60 W-s).
Faster performance at higher power draw with equal energy usage / battery life.
Though not my initial confusion, deltaFx2's does note higher discharge rates bring an amplified loss of capacity: I can only think of Apple, but hopefully Android OEMs won't fall into the same trap.
Kangal - Thursday, July 9, 2020 - link
Also, the efficiency of 865+ has increased over the 865.If you try and run an 865 at 3.1GHz, it is possible, as long as the thermal limits aren't reached and if the energy required is supplied. However, whilst the early 865 chipset would have to slowly and exponentially increase the Voltage to sustain that frequency... whereas the 865+ seems to only require a linear increase in voltage.
This is the "lowest-hanging fruit" way of increasing performance, because you can simply hit higher frequencies more stably, thanks to having a more accurate silicon. This happens as yields of nodes improve during production. In the PC Gaming space they call this "binning". Regardless, this "linear increase" is better than the "exponential increase", so it is still an efficiency increase.
So you might be wondering if it's even possible for the standard 865 to overclock that high?
And to show you it is, an easy example is trying to overclock the Tegra X1 processor found in the Nintendo Switch... that's what we observe on the Nvidia Shield TV, which runs with a higher power-draw and a larger cooling solution.
iphonebestgamephone - Friday, July 10, 2020 - link
And your source is?helloworld_chip - Friday, July 10, 2020 - link
In order to have the power increase to be "linear with the increased clock speed", we need the cpu to be running at similar voltage range right? How is it possible to run 2.84GHz and 3.1GHz at similar voltage under the same process node and same design? Voltage has to be increased aggressively to get to 3.1GHz29a - Friday, July 10, 2020 - link
If it will run at 3.1 Ghz at a given voltage then 2.84 is definitely possible at that same voltage. It's more efficient to lower the voltage at the lower speed but the cpu will work fine if you don't.helloworld_chip - Friday, July 10, 2020 - link
You are talking about a completely different thing. We are comparing 865 and 865+ here.GC2:CS - Wednesday, July 8, 2020 - link
I thought some said there is no 865plus for this year ?Why bother with such a release besides marketing hype for qualcomm and self proclaimed “gaming” phones ? Are there actualy people who would wait for this instead of a regullar dragon ?
s.yu - Thursday, July 9, 2020 - link
Because the marketing hype has minimal cost? Nobody would wait for this but some models were waiting for this, e.g. ROGP3, which if anywhere near ROGP2 is a solid specialized gaming device, not "self-proclaimed". The likes of Razer, OTOH is self proclaimed.name99 - Wednesday, July 8, 2020 - link
"@ 2133MHz LPDDR4X / 33.4GB/s
or
@ 2750MHz LPDDR5 / 44.0GB/s
"
Shouldn't those both be doubled? That's how wikichip describes them, LPDDR4X-4266 and LPDDR5-5500, and that would match the usual norms for how DRAM is described wouldn't it?
zamadatix - Wednesday, July 8, 2020 - link
DDR = dual data rate so 2133 MHz on the bus means the data rate is 4266.zamadatix - Wednesday, July 8, 2020 - link
double not dual*name99 - Wednesday, July 8, 2020 - link
There are many MANY degrees of doubling involved. Which is why it's all a matter of convention now, pretty much.But the convention appears to be that when I say LPDDR4X-4266, what I mean is that I will get 4.266 Gbits per sec from one chip, hence 34GB/s from eight chips. The precise details of exactly how many transactions of what width and length that translates into, and issues of "data" transactions vs "command" transactions are all mixed with the issue of DDR vs DDR2 vs DDR3 vs ..., which is precisely why we adopt a convention that focuses on the (PHY, guaranteed not to exceed) bandwidth, not how we got there.
eastcoast_pete - Wednesday, July 8, 2020 - link
I wonder how much performance one would gain by simply using (faster) LPDDR5 at otherwise unchanged top speed for the fast core? Which phones with SD 865 or 865+are currently using the newer generation memory?
flyingpants265 - Wednesday, July 8, 2020 - link
Won't matter. All Android phones are insanely laggy, even brand new ones. It's due to extremely poor software design.Retycint - Thursday, July 9, 2020 - link
Ah, here's someone who's never used an Android in their livesDizoja86 - Thursday, July 9, 2020 - link
2010 called, and they'd like their outdated information back.PeterCollier - Thursday, July 9, 2020 - link
Project butter was an abject fail. Bigger than the apple maps fail.Farfolomew - Wednesday, July 8, 2020 - link
Why doesn't TSMC grab the Apples, the AMDs, the NVidias, the Qualcomms by their proverbial b@lls, and overnight demand outrageous payment. They'd be the biggest company ever ... for a while anyways.nandnandnand - Wednesday, July 8, 2020 - link
They can't demand "outrageous" payments when Samsung's offerings are only 5-15% worse.Farfolomew - Thursday, July 9, 2020 - link
Yeah, but is Samsung competitive in this area? Why are none of the big players having their silicon produced in Samsung fabs? If all the current tip-of-the-spear chips are flowing through TSMC, that seems to me to be a lot of leverage that could be exploited...nandnandnand - Friday, July 10, 2020 - link
Nvidia is supposedly using Samsung 8nm for at least some of its lineup this year, maybe most of it. Supposedly, Nvidia tried to play hardball with TSMC using Samsung as leverage, but it didn't work. Samsung's 8nm node might be worse than TSMC's 7nm, but it's good enough for the purpose and cheaper.Even if Samsung continues to trail TSMC by a bit, they are close, they are spending lots of billions to reach 5nm, 3nm, etc. I wouldn't be surprised to see AMD use Samsung for some chips, even if it's just to increase the amount of low-end chips they can sell (they were still selling some 28nm GloFo chips in 2019: A6-9220C and A4-9120C).
29a - Friday, July 10, 2020 - link
They want to continue to be in business.Peskarik - Thursday, July 9, 2020 - link
Does it come with active cooling?LordConrad - Monday, July 13, 2020 - link
Instead of worrying about speed, they should add more L2 cache. All A7x cores should have at least 512KB, and all A5x cores should have at least 256KB.utmode - Sunday, July 19, 2020 - link
I am happy with current performance as I can't afford any more price increase.