iPhone 6 Leaks – Ultra Retina Screen, 2.6GHz A8 Processor, OİS (UPDATED)

[StingBull]

I have an iPod touch 4th gen and you are right it is to small to be used without a case, but on the battery issue I get 4-5 days out of mine and it is over 3 years old (I got it day 1).

I no longer use it for playing games it is not only used as a music player but it is used for at least 1-2 hours a day and it is not charged most nights and it seems to do quite well for itself.

After using mine for over a year I could only get 3-4 hours at the most. But I have used it to play games pretty much 24-7 so I was pretty heavy on mine. Now even if I fully charge it, it will lose half the battery in a couple of hours on standby.

[ivhyn]

Fake.

Apple won't introduce a bigger screen because it would cause a resolution chaos, it will have a 4" display like 5(s,c) with same density. I think the iPhone has perfect dimensions but it's built like shit and needs a bigger battery

[DLM_012]

I love that render. It looks like a cleaner, more low profile N5 with a wedge shape. 

[LAwLz]

1) Multicore is a mythe, we're nothing with it. And saying that more cores save battery life well yeah then you're lost.. You're never ever going to have a PC game like now BF4 on a mobile device unless you consider active cooling which is unacceptable or you break Moore's Law.

Ehh... What? Do you mean "myth"? Are you saying multicore processors do not exist? Yes, multiple cores can save power (assuming the architecture also gets slightly better, or else battery life will stay the same). You get a higher dynamic range. Let's think of it this way. Let's say we have two identical processors, except one is a quad core and one is a dual core. We will load a web page that will require 400 cycles (on a single core) to be fully loaded. Each cycle uses 1 watt of power per core. That means that the quad core can use twice as much power as the dual core if both are maxed out. Now, let's load the web page. The quad core will use 100 cycles to complete the task, and use up a total of 100 * 4 = 400 watts. The dual core will use 200 cycles to complete the task, and use 200 * 2 = 400 watts to complete the task. See how it equals out? This is of course oversimplified but hopefully you will get the point.

 

I don't see why you think we can't get BF4 level games on a mobile device without using active cooling or "breaking Moore's Law". I don't think you understand what Moore's Law is, because that is actually the law which says we will be able to run BF4 level games on mobile devices in the future. If you really think we will stay at roughly the same performance level in phones (and desktops) forever then you need to read some computer history. We will get there some day.

 

 

2) Thought you said that more cores increase your battery life? Car analogy logic again..

I never said a quad core would increase battery life. I said better hardware (as in, more powerful CPUs and so on) might increase your battery life. It's still the same amount of work that has to be done, and the faster the job is done the faster the cores can be turned off and the device can go into a idle state. Having more cores by itself won't increase battery life. Being able to use extra cores to quickly get work done and then go into an idle state will increase battery life. I suggest you read some of the articles regarding "race to idle".

big.LITTLE is a perfect example of how you can use multiple cores to save power.

 

I have no idea why you said "car analogy logic again" because I didn't make a car analogy...

 

 

3) You're nothing with 8 cores if all we do is single threaded meaning you're per core performance reliant. The 5S feels much smoother than any samsungcrap when browsing (http://be.hardware.info/reviews/3234/11/apple-iphone-5-review-vernieuwd-maar-niet-vernieuwend-browserprestaties-beter-dan-de-concurrentie), perfect example why octacores are nothing besides gimmicks. They might be useful for gaming but other than that its a waste.

And how many people only do single threaded tasks? The browser on most Android phones can use more than 4 cores. My video player can do it. Most games can do it. Emulators can. The list goes on. Pretty much all the apps I use are optimized for multiple cores.

The reason why the 5S feels smoother is because it uses tricks. For example it limits the speed at which you can scroll to ensure that you never scroll faster than it can load. Samsung phones does not have this limit. The iPhone also use long animations when doing things to give the CPU time to load. In iOS it also shows an image of the app to fool you into thinking the app is fully loaded, but you're actually just looking at a picture of how the app will look when it is fully loaded.

iOS has a lot of clever tricks into feeling smoother, but that does not mean it is actually fast.

I am not sure why you link to some German hardware running a bunch of browser based benchmarks. Weren't you the one saying benchmarks on mobile devices were useless?

Also, none of the benchmarks in that article is a good indicator of CPU performance since they are browser based. Oh, and none of the devices on that list is an octa core CPU...

 

 

Says the one who is believing mythes.

I have no idea what "mythes" you are referring to, or which "myths" you are referring to. Race to idle is not a myth and CPUs with multiple cores actually exists.

[Paralectic]

Take the iPhone 5s, make it as thick as the iPhone 4s and give us a great battery life > New ultrathin iPhone 6,7,8,9s

[BadgersAreCool]

Take the iPhone 5s, make it as thick as the iPhone 4s and give us a great battery life > New ultrathin iPhone 6,7,8,9s

The problem would be people would call the phone bulky. Right now iphones are pretty thick compared to their competitors. One of the main reasons phones aren't getting thinner is the battery, so they make it smaller with more power to make up for size. If the new iphone is as thin as the leak, the battery capacity will increase, but the retina display will drain it.

[Hawx]

btw, f*ck apple and their stupid huge 4.7 inch screen

'Stupid huge'??? Seriously? 4.7" was the standard size of android phones two years ago. They're still around the same size. Apple screens are piss small.

[Chevaishr]

lol, goniee elife s5.5 ... 5.55mm thin, if apple wanna say they got the thinnest phone or want to say that they made the thinnest phone first, they would be wrong though only by 0.02mm

[Guest]

Apple will always lead the way with industrial design. Battery life will be something to look forward to though - from their phones anyway. My Macbook Pro w/- Retina outlasts all of my friends laptops by a mile.

They lead design, I disagree they sacrifice productivity and power for look

[Faa]

Ehh... What? Do you mean "myth"? Are you saying multicore processors do not exist? Yes, multiple cores can save power (assuming the architecture also gets slightly better, or else battery life will stay the same). You get a higher dynamic range. Let's think of it this way. Let's say we have two identical processors, except one is a quad core and one is a dual core. We will load a web page that will require 400 cycles (on a single core) to be fully loaded. Each cycle uses 1 watt of power per core. That means that the quad core can use twice as much power as the dual core if both are maxed out. Now, let's load the web page. The quad core will use 100 cycles to complete the task, and use up a total of 100 * 4 = 400 watts. The dual core will use 200 cycles to complete the task, and use 200 * 2 = 400 watts to complete the task. See how it equals out? This is of course oversimplified but hopefully you will get the point.

Nope. Have a nice read http://scalibq.wordpress.com/2012/06/01/multi-core-and-multi-threading/

And your example is wrong because you cant split a single thread over a 2nd core, thats not how it works. And you theres no such thing as idle race. Have you ever heard about core parking? You'd enter a sleep state much faster with core parking than your stupid example. It's stupid to claim having a lower power consumption with more cores, thats just flatout wrong. A CPU doesnt only contain cores, using your fiction example the power draw would be higher using more cores.

 

 

I don't see why you think we can't get BF4 level games on a mobile device without using active cooling or "breaking Moore's Law". I don't think you understand what Moore's Law is, because that is actually the law which says we will be able to run BF4 level games on mobile devices in the future. If you really think we will stay at roughly the same performance level in phones (and desktops) forever then you need to read some computer history. We will get there some day.

Seems like you don't since you have no clue what his law is basically about -> the nm wall. At a certain point you cant shrink the transistors anymore and thats the end of your "BF4 on a mobile" dream.

 

 

And how many people only do single threaded tasks? The browser on most Android phones can use more than 4 cores. My video player can do it. Most games can do it. Emulators can. The list goes on. Pretty much all the apps I use are optimized for multiple cores.

 

[What the fuck]How are people doing single threaded tasks?[/What the fuck]

Because you saw some load on some cores it means nothing. A browser usually contains one main thread and all others are extremely minor that can be easily moved to the main threads core without any performance loss. And stop your "apps i use are optimized for multiple cores" crap because you have no clue if they do or not. The majority apps are having one or two main threads and a bunch of minor things. 

Google what a thread (software) is.

 

 

 

The reason why the 5S feels smoother is because it uses tricks. For example it limits the speed at which you can scroll to ensure that you never scroll faster than it can load. Samsung phones does not have this limit. The iPhone also use long animations when doing things to give the CPU time to load. In iOS it also shows an image of the app to fool you into thinking the app is fully loaded, but you're actually just looking at a picture of how the app will look when it is fully loaded.

 

Keep dreaming. Apple isnt stupid, they know what they need -> IPC.

 

 

OS has a lot of clever tricks into feeling smoother, but that does not mean it is actually fast.

 

Keep on.. 

 

 

I have no idea what "mythes" you are referring to, or which "myths" you are referring to. Race to idle is not a myth and CPUs with multiple cores actually exists.

See it this way, lets use your example again, if you run it of 2 cores and 6 cores are being disabled & you get the same performance with all 8 cores being used -> you'd be saving more power with the first one. The other 6 cores frequency will go down and the voltage as well. According to wiki it's if theres a constant voltage, which we dont have.

"Dynamic frequency scaling by itself is rarely worthwhile as a way to conserve switching power. Saving the most power requires dynamic voltage scaling too, because of the V² component and the fact that modern CPUs are strongly optimized for low power idle states. In most constant-voltage cases it is more efficient to run briefly at peak speed and stay in a deep idle state for longer (called "race to idle"), than it is to run at a reduced clock rate for a long time and only stay briefly in a light idle state. However, reducing voltage along with clock rate can change those tradeoffs."

[4Strings]

Anyone notice that it looks like a thin metal Moto X?

[Mr.Duck]

Anyone notice that it looks like a thin metal Moto X?

or a really thin HTC One

[LAwLz]

Nope. Have a nice read http://scalibq.wordpress.com/2012/06/01/multi-core-and-multi-threading/

And your example is wrong because you cant split a single thread over a 2nd core, thats not how it works. And you theres no such thing as idle race. Have you ever heard about core parking? You'd enter a sleep state much faster with core parking than your stupid example. It's stupid to claim having a lower power consumption with more cores, thats just flatout wrong. A CPU doesnt only contain cores, using your fiction example the power draw would be higher using more cores.

Ehh... Did you even read that article, or have looked at a modern program? You will have a hard time finding something that only runs 1 thread. Wanna know how many threads Firefox on my desktop has? 45 individual threads... You will have a hard time finding proper apps on Android and iOS that only use a single thread. Of course, not all threads are equally demanding, but you seem to think that a lot of programs are single threaded which they simply are not. I often see all 4 cores on my Galaxy S 4 light up. It's a very very common thing. I never said anything about running 1 thread on 2 cores.

 

Yes, there is such a thing as "race to idle". I linked you to an article showing it in action and explaining it, and a Wikipedia article explaining it. Yes I have heard about core parking. That's what I was referring to when I said "power gated", and modern SoCs uses power gating as well as multiple cores to save power. You don't "enter a sleep state much faster with core parking", core parking is the sleep state. Cores will be parked faster the faster you can finish a task (by for example having multiple cores or more powerful cores).

Did you read the link to ARM's big.LITTLE concept I linked you? Have you heard about the low power core Nvidia uses in Tegra 3 and Tegra 4? Do you think they are doing this as a joke and it doesn't have any benefits?

 

 

 

Seems like you don't since you have no clue what his law is basically about -> the nm wall. At a certain point you cant shrink the transistors anymore and thats the end of your "BF4 on a mobile" dream.

I do know about that. That's not what moore's law though (which will break once we hit that point). Everything regarding the smallest transistors we can make (about 10nm is the estimation before we run into strange quantum effects) has been based around silicon though. There are materials which would allow us to make even smaller transistors. The smallest transistor ever made was 3nm. Right now we are using 22nm. So even if you don't count architectural improvements (where there are a a lot to be made as well), we still have a long way to go before we run into that issue. I don't really have any issue imaging that we will be able to run BF4 grade games on mobile devices in the future. We already have tablets that can do it (although with active cooling).

 

 

 

[What the fuck]How are people doing single threaded tasks?[/What the fuck]

Because you saw some load on some cores it means nothing. A browser usually contains one main thread and all others are extremely minor that can be easily moved to the main threads core without any performance loss. And stop your "apps i use are optimized for multiple cores" crap because you have no clue if they do or not. The majority apps are having one or two main threads and a bunch of minor things. 

Google what a thread (software) is.

Damn this is going to blow your mind...

Here is my video player showing that it can use all 4 cores (and almost maxes them out as well).

Here is my browser using all 4 cores (the first two being used a lot and the third and fourth firing up every once in a while).

Just because you don't know which apps are or aren't optimized for multiple cores does not mean I don't... Your posts are mostly based on assumptions are very little real knowledge/facts.

 

This is without any multitasking by the way. I can run 2 programs at the same time if I want (having two windows up at the same time). That means that even IF the programs I used only benefited from having 2 cores, I would still benefit from having more than 2.

 

 

 

Keep dreaming. Apple isnt stupid, they know what they need -> IPC.

I have no idea how to respond to this. You didn't address anything I said. Yes IPC is really good to have. That does not mean they won't benefit from having more cores though. You didn't address anything I said about why the iphone feels smooth either.

 

 

See it this way, lets use your example again, if you run it of 2 cores and 6 cores are being disabled & you get the same performance with all 8 cores being used -> you'd be saving more power with the first one. The other 6 cores frequency will go down and the voltage as well. According to wiki it's if theres a constant voltage, which we dont have.

 

"Dynamic frequency scaling by itself is rarely worthwhile as a way to conserve switching power. Saving the most power requires dynamic voltage scaling too, because of the V² component and the fact that modern CPUs are strongly optimized for low power idle states. In most constant-voltage cases it is more efficient to run briefly at peak speed and stay in a deep idle state for longer (called "race to idle"), than it is to run at a reduced clock rate for a long time and only stay briefly in a light idle state. However, reducing voltage along with clock rate can change those tradeoffs."

I think you are misreading the Wikipedia article. Yes we do have dynamic voltage for individual cores in phones, and have had that for quite some time now. No, that does not mean race to idle doesn't work. It is still a better idea to do a task quickly and then go into deep sleep, than it is to drag the task out for a long period of time.

Please note how it says that "dynamic frequency scaling by itself is rarely worthwhile as a way to conserve switching power. Saving the most power requires dynamic voltage scaling too".

It is best to do both, which is what we are currently doing in modern SoCs. That does not mean Race to Idle doesn't work though. It is still better to do a task in a shorter period of time and then fully shut it off, than it is to drag the task out over a long period of time. Race to idle is not exclusive to constant-voltage situations, it's just that it is even more important in those situations.

[raxx]

I am still a fan of my Samsung GS4.

[ThatAngryGnome]

Not gonna happen. When the 5s was coming out, its A7 processor was leaked to be like a quad core 2.3 ghz or something with 2 gigs of RAM. Apple's Apple. Their not gonna change their marketing strategy to fit the market, they'll just force the market into accepting them. Even if the 6 was the same exact phone with maybe a screen bump (thats what I expect), people will still buy the phone. 

[scrid]

My opinion:

good move with the two sizes

could do better on the ppi  :wacko:

A8 is a very good processor for a phone

ultra retina - let's see the battery life first

It will be overpriced as always (with most of the phones not just iphones)

curved back - what am i supposed to do when i use it on a flat surface?

I really believe that they should make it compatible with other devices that are not designed by aplle through NFC and other things.

Generaly I wouldn't buy an iphone because I can simply find phones that are good in more fields and not just the processor or quality in lower prices

[c.bert]

I smell bullshit and 2.6 Ghz sounds to me like a very sweaty palm

[warmmilk]

'Stupid huge'??? Seriously? 4.7" was the standard size of android phones two years ago. They're still around the same size. Apple screens are piss small.

 

and thats the way I like it.  if i need a big screen for something I go and use my surface or my computer, I don't need a huge phone

[Faa]

*

Not going to read this crap honestly. You're running a benchmark in a browser to ramp the cpu loads up to show here that you're somewhat right and with an attitude of "damn this is going to blow your mind" when you're just fooling yourself only.

You're not wanting to understand that humans are bad multitasker, cores are meant to offload to take a thread over a core. When youre texting, thats running off a single core, when youre browsing you do it again, that's probably the most you do with a smartphone. I told you read that article and start to accept that multicores are nothing besides a stupid myth. Your race to idle makes no sense, so we should buy an 4core over a 2core because it will have better battery life time, yeah right.

Instead of moaning about a stupid theory you don't even have a proof about, fix this crap here:

 

 

1) Better hardware in phones is needed. The only reason why people still have desktops is because phones aren't powerful enough. Better hardware won't just benefit performance either, it will also INCREASE battery life because of "race to idle". So if the iPhone gets a quad core you might actually see an increase in battery life, because it will finish tasks quicker and then go down in idle.

That we see an iphone 6 having a better battery life time and more performance has nothing to do with race to idle, your corecount has nothing to do with that either. The fact that you shrink down, allowing you to use more transistors in the same surface makes you only get performance out of this and further reduce the power consumption. Add more transistors to a point that you equal a cpu.

Not going to deny that we need better performing chips, but that you need more cores is just BS.

[Pitou]

 

SOURCE: http://wccftech.com/iphone-6-rumors-ultra-retina/

this is false, everone knows that all apple products are moving closer to looking like a box, minus the pro which is a trash can

[gatecrasherlok]

Hides in a corner at the ready for things to be thrown self.

 

I will most likely be getting the new iphone, I just like the way apple phones work, layout, not complex.

 

Runs

[Misanthrope]

If this is true I can't wait to see how Apple and it's fans will call this innovative and get offended when pointing out the Moto X did it first

[LAwLz]

Not going to read this crap honestly. You're running a benchmark in a browser to ramp the cpu loads up to show here that you're somewhat right and with an attitude of "damn this is going to blow your mind" when you're just fooling yourself only.

Yeah, me posting scientific evidence as proof that what I am saying is correct, as well as posting links to back my claims up with. Yep, that's "crap". I posted that to prove that both my browser, as well as my video player can take advantage of 4 cores. Not only that, but the majority of the programs I use can do so as well. I posted that to prove to you that this part if your post is just flat out wrong and based entirely on bad assumptions you have made:

 

Because you saw some load on some cores it means nothing. A browser usually contains one main thread and all others are extremely minor that can be easily moved to the main threads core without any performance loss. And stop your "apps i use are optimized for multiple cores" crap because you have no clue if they do or not. The majority apps are having one or two main threads and a bunch of minor things.

 

 

You're not wanting to understand that humans are bad multitasker, cores are meant to offload to take a thread over a core. When youre texting, thats running off a single core, when youre browsing you do it again, that's probably the most you do with a smartphone. I told you read that article and start to accept that multicores are nothing besides a stupid myth. Your race to idle makes no sense, so we should buy an 4core over a 2core because it will have better battery life time, yeah right.

You don't seem to understand what the word "myth" means. Myth means a story that is believed by many people, but is not true. When you say multicores is a myth, you are saying "multicore CPUs do not exist", which they most certainly do. Multithreading is not a myth either, as I clearly demonstrated in my post earlier.

Now you have started to straw manning. I never said you should buy a quad core over a dual core because it will give you better battery life. What I said was that a quad core CPU can be more power efficient than a dual core, because it will have higher dynamic range, which is 100% true and has been proven over and over again. There are 0 drawbacks (except for cost) to having a quad core compared to having a dual core thanks to power gating. That was my point. Yes, race to idle does work and I highly suggest you read the links I posted earlier, because you don't seem to understand how and why it works.

 

 

Instead of moaning about a stupid theory you don't even have a proof about, fix this crap here:

I don't have proof!? I linked you to several sites explaining it and showing that it does work. I have proof. I have posted proof. You have not read the proof I have linked, that's the issue here. Calling it a "stupid theory" does not make it untrue.

 

 

 

That we see an iphone 6 having a better battery life time and more performance has nothing to do with race to idle, your corecount has nothing to do with that either. The fact that you shrink down, allowing you to use more transistors in the same surface makes you only get performance out of this and further reduce the power consumption. Add more transistors to a point that you equal a cpu.

Not going to deny that we need better performing chips, but that you need more cores is just BS.

Some of the battery life improvements we have seen can be directly linked to race to idle working. Again, Anand went over this in one of his articles (which I have linked to, you should read it). Yes, making transistors smaller helps power consumption as well, but thanks to race to idle we have seen reduced power consumption even when we have stayed on the same manufacturing process.

I have no idea what you mean by "add more transistors to a point that you equal a cpu".

 

So we don't need more than 2 cores? Really? Sure it is nice to get IPC improvements but it is not that hard to take advantage of 4 cores (see my post above for two examples) so it's a very easy way to double the theoretical performance of a CPU. With big.LITTLE adding more cores can also greatly improve battery life. You keep ignoring big.LITTLE and I suspect you keep ignoring it because you don't actually know what it is or how it works. It's a great idea and we will hopefully soon see it properly implemented in SoCs.

Ignoring my arguments does not make them untrue...

 

If we were talking about for example the octa core Cortex A7 chips that have started popping up then I would be on your side, but we are talking about either quad cores or big.LITTLE implementations and it seems like you are very ill informed regarding both of those.

[sunsetazul]

honestly it doesn't look bad at all but i hate the speakers and how they are at the bottom, always have with phones. they get covered if you hold them sideways, right now i have a Nokia lumia 920 (windows) but i want to switch android as developers barely make apps for windows, my girlfriend on the other hand has a HTC one but i my self would probably go for the HTC one max because i have big hands (I'm 6'5"). will i change to apple? no simply because its awkward for me to hold now...... to narrow. but they are at least taking a step in the right direction in my mind, also i don't find NFC to be that big of a deal to me, i have a wireless charger but i don't know, it seems to charge too slow. i rather plug it in as it would also charge completely. that's all i have to say.

[Faa]

Yeah, me posting scientific evidence as proof that what I am saying is correct, as well as posting links to back my claims up with. Yep, that's "crap". I posted that to prove that both my browser, as well as my video player can take advantage of 4 cores. Not only that, but the majority of the programs I use can do so as well. I posted that to prove to you that this part if your post is just flat out wrong and based entirely on bad assumptions you have made:

 

 

You don't seem to understand what the word "myth" means. Myth means a story that is believed by many people, but is not true. When you say multicores is a myth, you are saying "multicore CPUs do not exist", which they most certainly do. Multithreading is not a myth either, as I clearly demonstrated in my post earlier.

Now you have started to straw manning. I never said you should buy a quad core over a dual core because it will give you better battery life. What I said was that a quad core CPU can be more power efficient than a dual core, because it will have higher dynamic range, which is 100% true and has been proven over and over again. There are 0 drawbacks (except for cost) to having a quad core compared to having a dual core thanks to power gating. That was my point. Yes, race to idle does work and I highly suggest you read the links I posted earlier, because you don't seem to understand how and why it works.

 

 

I don't have proof!? I linked you to several sites explaining it and showing that it does work. I have proof. I have posted proof. You have not read the proof I have linked, that's the issue here. Calling it a "stupid theory" does not make it untrue.

 

 

 

Some of the battery life improvements we have seen can be directly linked to race to idle working. Again, Anand went over this in one of his articles (which I have linked to, you should read it). Yes, making transistors smaller helps power consumption as well, but thanks to race to idle we have seen reduced power consumption even when we have stayed on the same manufacturing process.

I have no idea what you mean by "add more transistors to a point that you equal a cpu".

 

So we don't need more than 2 cores? Really? Sure it is nice to get IPC improvements but it is not that hard to take advantage of 4 cores (see my post above for two examples) so it's a very easy way to double the theoretical performance of a CPU. With big.LITTLE adding more cores can also greatly improve battery life. You keep ignoring big.LITTLE and I suspect you keep ignoring it because you don't actually know what it is or how it works. It's a great idea and we will hopefully soon see it properly implemented in SoCs.

Ignoring my arguments does not make them untrue...

 

If we were talking about for example the octa core Cortex A7 chips that have started popping up then I would be on your side, but we are talking about either quad cores or big.LITTLE implementations and it seems like you are very ill informed regarding both of those.

We haven't seen any proof besides theory.

I'm just ignoring your nonsense. You have complety no clue what multithreading is about, people just suck with multitasking and you don't multitask on a phone. Playing a game in the 2nd half window and browsing on the 1st window isn't in my perspective multitasking. Phones are ideal to multitask and 99% dont even do actively multitasking. I have no ideas why the fuck a video player would torture that cpu so hard, start to use a hardware accelerated video player that would take advantage of the gpu so you have another reason to stop moaning for cores.

The funny thing is, the cores arent turning off there are apps for a reason that allows you to do this. Because you saw 0% on some cores doesnt even mean they are turned off.

All you need is IPC, since you cant compare the instruction per clock rate between different chips so practically you would only be able to compare the end result -> per core performance is all that matters for the majority of us.

Race to idle, race to idle, race to idle, just race to shadap