iPhone 5S impressions

[wco81]

BTW, WSJ reported this week that Apple is working on a 4.5 inch and a 5-inch or greater design for intro later this year.

So they may segment the iPhone product line further by screen size, presuming they keep something like the 5C, for lower price point.

I don't really covet the larger phones that much over my iPhone 5, though I try to minimize reading web sites on it, preferring to use my iPad or a computer. At least for extended reading.

Maybe a 5-inch phone changes that behavior somewhat but that's hard to say. If I didn't have a tablet at all, then it makes sense to have a larger screen on your only mobile device. But I don't ever anticipate being without a tablet. In fact, I'd be more likely to carry around a tablet more frequently, if I do get an iPad Mini or something like a Nexus 7. As it is, my regular sized iPad is with me most of the time during the day.

 

[RudeCurve]

 

[silent_guy]

I hope for some technical explanations not sarcasm!

I don't think CPUs matter all that much these days, and I was living under the impression that both the S4 and 5s were pretty similar in terms of performance. Turns out that I was completely wrong on that one. In the usual crop of standard benchmarks, the 5s destroys the S4. Again, I don't think it matters all that much when doing browsing and stuff, though it may give an advantage when doing complex camera processing?

If you ignore a few choice synthetic benchmarks, the story is no different for the GPU. No surprises, Apple has always been on the forefront there. Does that matter in real life? I don't think may people will notice.

But it really makes me wonder exactly which processor specifications you had in mind when you claim that the S4 destroys the 5s. Given your $20 estimate in the same post, you must have been smoking really good stuff at the time!

In terms of camera, the opinions seem to be divided. Some will point to the 13MP on the S4, others will point out the better performance in the dark of the 8MP of the 5s. It doesn't seem to be a slam dunk one way or the other.

The battery life of the 5s seem to beat the S4 in most use cases, but the S4 has a larger battery in absolute numbers, woohoo!

The display of the 5s is higher quality, the S4 is larger. A matter of taste.

In terms of sensors: you list quintessential stuff like humidity and barometric pressure for the S4, which, I admit, is something I'm sure people are using day in and day out, as opposed to the one that you conveniently forget to mention, the finger print sensor of the 5s, which I see people around me use 30 times per day: every single time they look at their phone.

As for memory size: I take it the S4 has larger memory than the 5s? But don't forget: even the most advanced garbage collection systems require a factor of more memory that those that don't to ensure smooth behavior. In practice, I think we're far beyond the point where the difference matters, to be honest. 1GB, 2GB, 4GB? Are there day-to-day apps where you'd see it in a double blind test? I doubt it.

Of course, it terms of quality of enclosure and finish, the 5s has no competition from the S4, but most people use a case no matter which phone they use, so it's not that big a deal either.

Either way: the whole thing about 'destroys in every possible specification' is the usual bunch of hyperbole that I've come to expect from you.

I think both phones will cost a similar amount to manufacture. The cost of the extra GB RAM and those useful sensors offset by the cost of a larger die on the 5s, the finger print sensor and the precision manufacturing. But Apple will profit more because they can command a higher price and because Samsung spends insane amounts more on advertising.

 

[zed]

On the 6th time, the phone landed perfectly squarely on its screen (It sounded different), and cracked the screen.

It most likely wasnt that you dropped it a different way, the thing is even after that first drop you have all these minute cracks (that you cant see) so you think oh thats great nothing damaged, but thats not true. So you can do a few of these drops until a threshold is reached and then is kaboom, its not cause drop number 6 or 5 or 4 or 7 was higher/different angle than the previous ones is just its tollerance has been exceeded

 

[ltcommander.data]

Is a pure component BOM really reflective of a phone maker's profit margin on the device? For example, a fingerprint sensor might only cost a dollar or two coming back from the fab, but Apple bought a whole company to put that sensor in the iPhone 5S plus additional R&D to get it ready for mass market/mass production. Samsung does make their own SoC, but they are utilizing Cortex CPU designs licensed from ARM, and many Galaxy S4 use off-the-shelf S800 from Qualcomm which is a different investment outlay than architecting your own CPU like Apple does. I'm not sure a BOM is going to accurately capture the cost difference, which includes not only the hardware itself but associated software development like drivers, between using an off-the-shelf component and designing your own.

 

[silent_guy]

Is a pure component BOM really reflective of a phone maker's profit margin on the device? For example, a fingerprint sensor might only cost a dollar or two coming back from the fab, but Apple bought a whole company to put that sensor in the iPhone 5S plus additional R&D to get it ready for mass market/mass production. Samsung does make their own SoC, but they are utilizing Cortex CPU designs licensed from ARM, and many Galaxy S4 use off-the-shelf S800 from Qualcomm which is a different investment outlay than architecting your own CPU like Apple does. I'm not sure a BOM is going to accurately capture the cost difference, which includes not only the hardware itself but associated software development like drivers, between using an off-the-shelf component and designing your own.

Yes, that too. The fact that Apple doesn't have different P&L centers in the company in terms of components doesn't make it any easier.

But it's not as if the phones are radically different. If you want to compare cost, you could look at it would cost for an external company to make a similar product and that would be a good first order estimate.

 

[Ailuros]

If you ignore a few choice synthetic benchmarks, the story is no different for the GPU. No surprises, Apple has always been on the forefront there. Does that matter in real life? I don't think may people will notice.

Depends where you're coming from within the iPhone world and the use case I guess. That one though is pretty damn noticable if you're coming from an iPhone4 f.e: http://forum.beyond3d.com/showpost.php?p=1798807&postcount=2246

And since I recently placed a friend's original mini iPad against the latest one in a store where we did some quick browsing with some heavy pages for fun the difference was immediately noticable.

 

[silent_guy]

Depends where you're coming from within the iPhone world and the use case I guess. That one though is pretty damn noticable if you're coming from an iPhone4

Let me restate then: for the most common use cases such as browsing, the difference will be less noticeable for each successive generation.

There is a huge difference in experience between my old 4 and my current 5, something you could see even after a few minutes of trying out in the store. But if I go to the store now and try a 5s against my 5, it's not obvious at all.

I haven't tried an S4. Maybe the lower performance of its SOC would be significant enough to be noticeable, but I doubt it. (In my case, the Android UI, not the Android features, is revolting enough that it would stand in the way of being able to make an objective assessment anyway.)

 

[Pressure]

Let me restate then: for the most common use cases such as browsing, the difference will be less noticeable for each successive generation.

There is a huge difference in experience between my old 4 and my current 5, something you could see even after a few minutes of trying out in the store. But if I go to the store now and try a 5s against my 5, it's not obvious at all.

I haven't tried an S4. Maybe the lower performance of its SOC would be significant enough to be noticeable, but I doubt it. (In my case, the Android UI, not the Android features, is revolting enough that it would stand in the way of being able to make an objective assessment anyway.)

That's entirely wrong. The A7 is much faster at rendering websites, so it makes a remarkable difference in browser experience, especially compared to A5 and A6.

 

[Pressure]

Let me restate then: for the most common use cases such as browsing, the difference will be less noticeable for each successive generation.

That's entirely wrong. The A7 is much faster at rendering websites, so it makes a remarkable difference in browser experience, especially compared to A5 and A6.

 

[silent_guy]

That's entirely wrong. The A7 is much faster at rendering websites, so it makes a remarkable difference in browser experience, especially compared to A5 and A6.

If you go from rendering a page in 10s to 5s to 2.5s to 1.25s etc, you double the speed for each generation, but that doesn't mean the experience will improve just as much, especially if other parts of the pipeline improve the same way (wireless connection especially.) at one point, it will be fast enough?

Either way, I'd love to be wrong on this. Need some justification for whatever new phone will be released this year...

 

[Entropy]

Is a pure component BOM really reflective of a phone maker's profit margin on the device?

Of course not. The industry uses BOM calculations to estimate market sizes in the subcontractor space.
The software and business sides of consumer computing platforms are completely absent from BOM numbers, as is R&D/fab space reservation/et cetera.

Using BOM numbers to estimate profits is....erm...ill adviced. (*cough*)

 

[Ailuros]

User experience is also relative to a user's experience for which I wouldn't count you under the average user. Either way I'd expect the average user to upgrade every two years or so, which then obviously guarantees a higher difference.

I'd very surprised if you couldn't see a difference in a quick test especially considering the resolution difference we had between the original mini pad & the new retina mini. He won't upgrade either...

 

[silent_guy]

I'd very surprised if you couldn't see a difference in a quick test especially considering the resolution difference we had between the original mini pad & the new retina mini. He won't upgrade either...

I was talking iPhone. The iPad mini was a total no go with that screen. (The CPU was way slower than an iPhone 5 too, I think?). The mini retina entered the house as soon as it was available.

 

[patsu]

It most likely wasnt that you dropped it a different way, the thing is even after that first drop you have all these minute cracks (that you cant see) so you think oh thats great nothing damaged, but thats not true. So you can do a few of these drops until a threshold is reached and then is kaboom, its not cause drop number 6 or 5 or 4 or 7 was higher/different angle than the previous ones is just its tollerance has been exceeded

Yeah you are right too ! But it also depends on the angle of drop/impact. The last one was squarely on the screen. I could see it fall straight down.

Is a pure component BOM really reflective of a phone maker's profit margin on the device? For example, a fingerprint sensor might only cost a dollar or two coming back from the fab, but Apple bought a whole company to put that sensor in the iPhone 5S plus additional R&D to get it ready for mass market/mass production. Samsung does make their own SoC, but they are utilizing Cortex CPU designs licensed from ARM, and many Galaxy S4 use off-the-shelf S800 from Qualcomm which is a different investment outlay than architecting your own CPU like Apple does. I'm not sure a BOM is going to accurately capture the cost difference, which includes not only the hardware itself but associated software development like drivers, between using an off-the-shelf component and designing your own.

BOM, R&D, marketing, channel logistics, online infrastructure.

I hear Apple also make the robots and equipments that make their Macs and devices (not just MacPro).

They are innovating in the art of making and distributing electronics, beyond innovating in the devices.

Steve Jobs say manufacturing is never coming back to US, but I half suspect he lied. They would want end-to-end innovation to happen in 1 place. Initial batches may be done in the western countries (jobs !), and then diffuse to Asian partners in parallel. It should work for heavier products and custom products. Plus probably better ability to keep secrets.

I hope it's true ! The current China-centric manufacturing model works but also has serious flaws.

 

[UniversalTruth]

I don't think CPUs matter all that much these days, and I was living under the impression that both the S4 and 5s were pretty similar in terms of performance. Turns out that I was completely wrong on that one. In the usual crop of standard benchmarks, the 5s destroys the S4. Again, I don't think it matters all that much when doing browsing and stuff, though it may give an advantage when doing complex camera processing?

If you ignore a few choice synthetic benchmarks, the story is no different for the GPU. No surprises, Apple has always been on the forefront there. Does that matter in real life? I don't think may people will notice.

But it really makes me wonder exactly which processor specifications you had in mind when you claim that the S4 destroys the 5s. Given your $20 estimate in the same post, you must have been smoking really good stuff at the time!

In terms of camera, the opinions seem to be divided. Some will point to the 13MP on the S4, others will point out the better performance in the dark of the 8MP of the 5s. It doesn't seem to be a slam dunk one way or the other.

The battery life of the 5s seem to beat the S4 in most use cases, but the S4 has a larger battery in absolute numbers, woohoo!

The display of the 5s is higher quality, the S4 is larger. A matter of taste.

In terms of sensors: you list quintessential stuff like humidity and barometric pressure for the S4, which, I admit, is something I'm sure people are using day in and day out, as opposed to the one that you conveniently forget to mention, the finger print sensor of the 5s, which I see people around me use 30 times per day: every single time they look at their phone.

As for memory size: I take it the S4 has larger memory than the 5s? But don't forget: even the most advanced garbage collection systems require a factor of more memory that those that don't to ensure smooth behavior. In practice, I think we're far beyond the point where the difference matters, to be honest. 1GB, 2GB, 4GB? Are there day-to-day apps where you'd see it in a double blind test? I doubt it.

Of course, it terms of quality of enclosure and finish, the 5s has no competition from the S4, but most people use a case no matter which phone they use, so it's not that big a deal either.

Either way: the whole thing about 'destroys in every possible specification' is the usual bunch of hyperbole that I've come to expect from you.

I think both phones will cost a similar amount to manufacture. The cost of the extra GB RAM and those useful sensors offset by the cost of a larger die on the 5s, the finger print sensor and the precision manufacturing. But Apple will profit more because they can command a higher price and because Samsung spends insane amounts more on advertising.

This is one hell of a good explanation! Thanks!

Just wanted to say that I also don't like the display of S4 being AMOLED, etc, but previously thought that resolution and ppi matter, obviously not that much. And after carefull look through some images of both S4 and 5S, I do agree that the 5S' make looks better quality and definitely more expensive!
About the finger print--- I usually hate all that stuff on my phone and try to keep it as clean as possible, so when you mention finger print and something bad happens in my stomach. Perhaps it is just me

For everything else- thank once again

 

[Grall]

If you go from rendering a page in 10s to 5s to 2.5s to 1.25s etc, you double the speed for each generation, but that doesn't mean the experience will improve just as much, especially if other parts of the pipeline improve the same way (wireless connection especially.) at one point, it will be fast enough?

For browsing, arguably I would say the A7 CPU is already "fast enough". It feels - at least to me - as if it is not the limit when surfing the web. Perhaps there are very script/HTML-heavy web pages where it slows down while rendering, but I've not noticed that in my own travels. The A4 obviously had a LOT of room for improvement in this regard, especially if you multitasked and ran a music player simultaneously and such. A7 however...not so much.

Of course, there's other venues through which to leverage computing power in a mobile device - photography most famously in the A7's case and we'll likely see more of that too in the future.

I was talking iPhone. The iPad mini was a total no go with that screen. (The CPU was way slower than an iPhone 5 too, I think?).

You seem to be correct there. Original Mini uses the A5 CPU (same as in iPad2, although shrunken down to a more recent silicon process it would seem).

 

[Ailuros]

I was talking iPhone. The iPad mini was a total no go with that screen. (The CPU was way slower than an iPhone 5 too, I think?). The mini retina entered the house as soon as it was available.

The difference between A5 & A6 is diametrically smaller than A6 vs A7. You folks are obviously forgetting that comparing 1024*768 to 2048*1536 isn't exactly apples to apples with by 4x times higher resolution.

iPhone5 = Sgx543MP3@325MHz
iPad mini (1) = Sgx543MP2@200MHz

 

[wco81]

It may render web pages fast but the tabs reload a lot.

And iOS 7 on my iPad 3 will reload apps if you switch among 3 or 4 apps, go back to the first one.

That seems to point at lack of RAM, though there are reports that iOS 7.1 handles the reloading of the Safari tabs better.

 

[Laa-Yosh]

Some will point to the 13MP on the S4, others will point out the better performance in the dark of the 8MP of the 5s.

Chasing megapixels is only good for ripping you off. The tiny optics can't give you that much detail even at 5MP so almost anything beyond that is completely useless anyway (I usually downscale my i4S photos to 1/4 size). What it does kill is memory, 16GB phones will run out very fast on 10MP+ pictures and 1080p videos.

The display of the 5s is higher quality, the S4 is larger. A matter of taste.

I decided not to upgrade to 5S partially because I need a bigger display. Typing should get easier with my finger size and reading web pages would be better too, even if the resolution stayed the same.
I'll be in the market for a 4.5"+ iPhone, but if it stays at 4" I'll refuse to upgrade.