Although Anandtech didn't write up anything on it. Broadcom has officially withdraw from the Mobile baseband business. Mediatek will have their own 4G Modem within this year, LG and Huawei too, but all of them are Integrated into their SoC. Which means the market are left with Qualcomm and Intel only. ( Altair are LTE Only )
Cant wait to see your review on MDM9x35, which is very likely to be used by Apple's next iPhone. And I have been wondering why Apple still hasn't bring the baseband into its SoC. If Mediatek can do it i doubt then why not Apple?
Apple traditionally uses Qualcomm modems. AFAIK they license the modem as an IP block to customers, although some of their SoCs integrate modem hardware (not that that is an interesting proposition to Apple). So Apple has two choices: motivate Qualcomm to license the modem as an IP block to them or make their own LTE modems.
Apple has hired baseband engineers. There was a story about this on AnandTech a few months ago. I'm assuming it's too early to see the results of their labors in the iPhone6, but it looks like Apple wants at least a backup plan so they are not at Qualcomm's mercy.
S5 is truly amazing. One of the phones that received high rates based on consumer satisfaction ( take a look at http://www.topreport.org/phones/ for instance...)
Mediatek do have a discrete LTE modem, as do nVidia and Huawei (HiSilicon). Altair, Sequans and GCT have LTE-only modems too. Lots of rumours that LG, Samsung and maybe even Apple are working on their own as well.
While we may now be clearly at the point of seriously limited returns on further PPI increases in mobile phone displays, one notable benefit at least is improvements to Oculus rift products. The V2 is 1920x1080 I believe, split in half - So 960x1080 per eye or about 1megapixel per eye. To go to 2560x1440 would be around 1.8megapixels per eye. Add some really nice AA and you're starting to get a semi-reasonable resolution (I'll assume the Oculus DK3 will have this)
I thought we might be at the point that PPI increases would slow down but I think now we may see continued development. I' wouldn't be surprised to see a full 4k, 5 -> 6" phone within the next 18 months, hopefully OLED of course.
If we want resolution independence at the OS level, it's useful to push resolutions to the point where scaling to an arbitrary degree doesn't create visible aliasing. Apple had to implement 4x OGSSAA to achieve acceptable scaling on the MacBook Pro with Retina display, and even still people complain that the non-integer scaling modes look noticeably less sharp. For most people, that means the sweet spot for phones is somewhere in the 600-1200 ppi range, and something more like 5760x3600 for a 15" laptop. Of course that doesn't account for Vernier acuity, but if you're past the threshold of visual acuity, relatively cheap antialiasing techniques can be used to create images where the aliasing is virtually undetectable.
This is why I think 1080p is the most you need for a smart phone sized screen. Even 720p is plenty in my view. The only reason to offer 1080p to me is so that full HD video doesn't haven't be scaled. As you see with PCs, the more pixels to be pushed, the more stress there is on the hardware.
Just my opinion and people can disagree, but I'd prefer cheaper devices with better battery life or more performance with the same hardware than bragging that my phone has the same resolution as the screen at the movie theater.
I had a Note 2 and now a Note 3. The difference between 720p and 1080p is quite significant so I'd disagree with your "720p is plenty" statement. With that said, the move to 1440p is not as significant and probably an SOC generation or 2 premature until it can not only handle it, but excel with it.
Kudos to Samsung for keeping the battery/performance quite even with the 1080p screen, though.
sure, you're right. 720p Note II vs 1080p Note III very noticeable difference. But the truth is quality of the screen. 720p panel on Note II really poor on color accuracy, color temperatur,brightness and saturation compare to 1080p Note III. That's the key. When u start see a good 720p panel screen, it will be barely noticeable next to 1080p.
It seems, the quality AMOLED Display Samsung got really huge improvement since the Galaxy S4 display.
As Joshua pointed out, these enormous resolutions will make a big difference when you're using the smartphone as a VR device, i.e. using Google Cardboard.
I do think Joshua should have dwelled in this factor a bit more, though. A single reference to VR in the conclusion wasn't enough for most people.
As I thought, minimal battery impact would be seen with higher resolution with their OLED. Would you mind testing for maximum brightness or if possible, your complete display analysis?
Any chance you'll instrument the GS5 LTE-A to measure power consumption directly as Anand did with the Galaxy S 5 (T-Mobile) and GS4 (AT&T)? I'm curious to see the difference in device power at idle with the screen on, to see how much of the battery life parity is due to overall platform power reduction or just the relative efficiency of the screen itself given the resolution.
I would think a resolution increase on an OLED panel wouldn't have as much of an increase on power consumption as a comparable increase in resolution on an LCD due to the emissive nature of the technology. On an LCD with shrinking pixels, it's blocking more light from the backlight, so the backlight needs more power to achieve the same brightness. On an OLED you have more of the smaller pixels that have a similar luminous efficiency as the old larger pixels. I suppose the only power losses would be from higher resistance from smaller wires controlling the pixels and perhaps some other small power increases I'm not considering. Of course there's still the same increase in processing power needed to drive more pixels.
I'd prefer they make improvements to 1080p on a ~5" screen than move to a higher resolution panel. I just don't see the need for higher. The cons far outweigh the pros IMHO. Samsung can build a 1080 AMOLED panel with near-RGB strip arrangement (like in the Note 2). The clarity of the screen would drastically improve there, without needing to strain the internals. Not only does a higher resolution demand more CPU and GPU horsepower, but also bandwidth and more memory (and more internet bandwidth if the services are DPI aware). I'd rather gain considerably more performance and battery life with each iteration of a smartphone than utilizing that improvement in powering gimmicky features with diminishing returns. SoCs have improved so much in the past couple of years yet we slightly felt any difference with each iteration just because those SoCs are simply more strained than former ones. It's such a waste. I can get even more technical about this, but I'm tired of explaining myself.
Really, 1080 is not only excellent, but absolutely amazing and mind blowing on a screen that size. Lets perfect what we have first before moving to the next (unneeded) level, sigh. Apple got away with 640p for more than 4 plus who knows how many more in the future. They took their time to perfect their internals to work with that resolution and DPI to the point that their smartphone is stupidly fluid. I see absolutely no reason why Samsung wouldn't get away with doing the same. Mind you, the panel on the S5 is leaps and bounds better than the on any iPhone.
Trust me people. Standardizing 1080 for smartphones would do a HUGE favor for the industry and the ecosystem. Having less things to worry about would make SoC manufactures concentrate more on what counts, not just optimizing the next big SoC to drive even more pixels...
In what way does retarding the development of one particular component help the industry as a whole move forward? Screen resolution has doubled roughly 3 times in the past 20 years, while mobile SoC GPUs have doubled in performance almost 9 times in 7 years. In many ways, screen resolution has been waiting around for better GPUs; why take the pressure off now that we can finally drive these displays?
Because we've reached a point where more isn't necessarily better in particular components, and improvements need to be made in others. Current mobile GPUs don't even play 1080p games at 60fps with decent efficiency, and only a handful can tell a difference between of text or even video on a 1080p VS 1440p on a 5.x" screen. Why the heck would you improve certain components when you can barely tell the difference? Would you rather have a 1440p smartphone that runs OK or a 1080p smartphone that runs exceptionally well? 1080p on a 5" screen already looks superb, mind you. We're not in dire need of a higher resolution. When hardware can effortlessly and efficiently run EVERYTHING at 1080p and still has change to spare, then think of upgrading, otherwise I believe we're still not at that point yet. There's still lots of room of improvement for 1080p on a mobile small screen.
"In what way does retarding the development of one particular component help the industry as a whole move forward?" I believe "retarding: is an excessive way to put it. You speak as if current 1080p screens are absolutely perfect. They're not. There's lots of room for improvement in various aspects.
We're at 600 ppi right now, after stagnating at 90-150 ppi for over a decade. If we can double twice more, we'll be at 2400 ppi and will have satisfied the Nyquist criterion for human visual acuity. At that point you can draw at any arbitrary resolution you like and scale it with no visible aliasing. If you want to game, great. You can set a minimum frame rate and the game engine will just set the resolution given the capabilities of the SoC. Text and vector art can be composited using another layer at the full screen resolution and will be perfect. All we need is higher resolution screens and more flexible scalers and we can finally achieve resolution independence, which would be an epic achievement.
When the standard sampling rates for digital audio were chosen, they didn't pick 44.1 kHz by accident; they did so because the limit of human hearing was right around 22 kHz. They didn't arbitrarily go with a lower frequency because most people over 30 have an attenuated hearing range. If we want screens where the subpixel arrangement is undetectable, and can accurately reproduce an image with zero aliasing even for young children and individuals with exceptional vision, then we should make an effort to get to at least 2150 ppi. If screens can be produced at such high resolutions affordably and with power requirements comparable to any of the other available displays, there is no reason not to use them in shipping products.
It is a fallacy to state that the current resolution race is adversely affecting any other aspect of mobile device development. The display in the GS5 LTE-A is a perfect example: in addition to being the highest resolution display ever shipped in an Android device, it is also the most accurate, and there was no net adverse effect on battery life. The only metric in which we see this screen lag ever so slightly is max brightness, but it still manages to beat the screen in the regular GS5.
You're missing my point entirely. SoCs WILL get faster and more efficient even without increasing screen resolution. You're point is that higher resolutions don't affect performance, which is completely inaccurate. Your example on audio is relatively non- relevant. Different sampling rates have nowhere near the impact on performance as do different resolutions. You're comparing the GS5 with the GS5 LTE-A. Let me remind you that these are completely different phones with different internals. They just happen to have around the same battery life because the latter happens to be more efficient in driving more pixels, which balances things out. If you want to check if the extra resolution does in fact affect performance and battery life, then you'll have to examine the same device with the same internals against different resolution screens. You have come to the conclusion that the current performance and battery life of today's smartphones are adequate. They're not. A GS5 LTE-A with a 1080p panel will surely run smoother, cooler, and longer on battery than the current one with 1440p. Even if the difference isn't very significant, it's still a difference that the average consumer appreciates. With all do respect, but I believe your priorities should be reevaluated. Those who care about 2000ppi and no visible antialiasing are an absolute minority, especially on a small ~5" screen. 440PPI is by no means the same as the 90-150 ppi we've been "stuck" with for decades. It surely is adequate for today's (and tomorrows) applications. What we "need" now is a faster, cooler performing devices, with 3+ days of battery life and highly optimized software. That's what the industry should be focusing on.
For Amoled screen, with more pixels, each pixel can reduce the luminance to get equal brightness performance, and with a specific brightness the total luminance is the same. That‘s why we see virtually no difference in power consuming. However, there are two basically improvements: - Lower luminance means better durability (which is important with Amoled display) - Theoreotical max brightness is higher, means in the next generation when other power consumption advances are applied, we can see the major bump in brightness (same as S4 to S5)
i may be in the minority but i wish for a 5" small bezel with a 4K sRGB display well calibrated, low power. and 4k in bigger screen sizes too people saying this is good enough is the reason why there are still computers sold with 1366*768 TN panels (ugh)
i got a RAZR HD and i can see the pixels nevermind that my screen is all burned and yellowish now.
im upgrading till s810 comes out and better QuadHD panles come out.
galaxy nexus was the first 720p phone, how was the screen then?
The comparison does not specify smartphones NAND storage size ! it is known that 32GB NAND storage performs significantly faster on sequential writes. Can you confirm same size ( 16GB or 32GB ?? ) of NAND storage for all smartphones under test here ?
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.
32 Comments
Back to Article
iwod - Tuesday, August 5, 2014 - link
Although Anandtech didn't write up anything on it. Broadcom has officially withdraw from the Mobile baseband business. Mediatek will have their own 4G Modem within this year, LG and Huawei too, but all of them are Integrated into their SoC. Which means the market are left with Qualcomm and Intel only. ( Altair are LTE Only )Cant wait to see your review on MDM9x35, which is very likely to be used by Apple's next iPhone. And I have been wondering why Apple still hasn't bring the baseband into its SoC. If Mediatek can do it i doubt then why not Apple?
OreoCookie - Tuesday, August 5, 2014 - link
Apple traditionally uses Qualcomm modems. AFAIK they license the modem as an IP block to customers, although some of their SoCs integrate modem hardware (not that that is an interesting proposition to Apple). So Apple has two choices: motivate Qualcomm to license the modem as an IP block to them or make their own LTE modems.name99 - Tuesday, August 5, 2014 - link
Apple has hired baseband engineers. There was a story about this on AnandTech a few months ago. I'm assuming it's too early to see the results of their labors in the iPhone6, but it looks like Apple wants at least a backup plan so they are not at Qualcomm's mercy.craighamilton - Saturday, December 6, 2014 - link
S5 is truly amazing. One of the phones that received high rates based on consumer satisfaction ( take a look at http://www.topreport.org/phones/ for instance...)adntaylor - Tuesday, August 5, 2014 - link
Mediatek do have a discrete LTE modem, as do nVidia and Huawei (HiSilicon). Altair, Sequans and GCT have LTE-only modems too. Lots of rumours that LG, Samsung and maybe even Apple are working on their own as well.AbRASiON - Tuesday, August 5, 2014 - link
While we may now be clearly at the point of seriously limited returns on further PPI increases in mobile phone displays, one notable benefit at least is improvements to Oculus rift products. The V2 is 1920x1080 I believe, split in half - So 960x1080 per eye or about 1megapixel per eye. To go to 2560x1440 would be around 1.8megapixels per eye. Add some really nice AA and you're starting to get a semi-reasonable resolution (I'll assume the Oculus DK3 will have this)I thought we might be at the point that PPI increases would slow down but I think now we may see continued development. I' wouldn't be surprised to see a full 4k, 5 -> 6" phone within the next 18 months, hopefully OLED of course.
repoman27 - Tuesday, August 5, 2014 - link
If we want resolution independence at the OS level, it's useful to push resolutions to the point where scaling to an arbitrary degree doesn't create visible aliasing. Apple had to implement 4x OGSSAA to achieve acceptable scaling on the MacBook Pro with Retina display, and even still people complain that the non-integer scaling modes look noticeably less sharp. For most people, that means the sweet spot for phones is somewhere in the 600-1200 ppi range, and something more like 5760x3600 for a 15" laptop. Of course that doesn't account for Vernier acuity, but if you're past the threshold of visual acuity, relatively cheap antialiasing techniques can be used to create images where the aliasing is virtually undetectable.damianrobertjones - Tuesday, August 5, 2014 - link
I absolutely LOVE the first picture... Taken with a Nokia 1020! :) If you're going to do it use the best.greyhulk - Tuesday, August 5, 2014 - link
Cool story, bro.MonkeyPaw - Tuesday, August 5, 2014 - link
Sadly, Anandtech never reviewed the 1020.halcyon - Tuesday, August 5, 2014 - link
Why is Samsung Galaxy Note 3 missing from the tables?vortmax2 - Tuesday, August 5, 2014 - link
Yes, I'd like to see where the Note 3 is as well (I'm guessing it's a little below the S5 level of performance in each chart).Death666Angel - Tuesday, August 5, 2014 - link
Both are in bench, so hit it up yourself!FITCamaro - Tuesday, August 5, 2014 - link
This is why I think 1080p is the most you need for a smart phone sized screen. Even 720p is plenty in my view. The only reason to offer 1080p to me is so that full HD video doesn't haven't be scaled. As you see with PCs, the more pixels to be pushed, the more stress there is on the hardware.Just my opinion and people can disagree, but I'd prefer cheaper devices with better battery life or more performance with the same hardware than bragging that my phone has the same resolution as the screen at the movie theater.
vortmax2 - Tuesday, August 5, 2014 - link
I had a Note 2 and now a Note 3. The difference between 720p and 1080p is quite significant so I'd disagree with your "720p is plenty" statement. With that said, the move to 1440p is not as significant and probably an SOC generation or 2 premature until it can not only handle it, but excel with it.Kudos to Samsung for keeping the battery/performance quite even with the 1080p screen, though.
AciMars - Tuesday, August 5, 2014 - link
sure, you're right. 720p Note II vs 1080p Note III very noticeable difference. But the truth is quality of the screen. 720p panel on Note II really poor on color accuracy, color temperatur,brightness and saturation compare to 1080p Note III. That's the key. When u start see a good 720p panel screen, it will be barely noticeable next to 1080p.It seems, the quality AMOLED Display Samsung got really huge improvement since the Galaxy S4 display.
ToTTenTranz - Tuesday, August 5, 2014 - link
As Joshua pointed out, these enormous resolutions will make a big difference when you're using the smartphone as a VR device, i.e. using Google Cardboard.I do think Joshua should have dwelled in this factor a bit more, though. A single reference to VR in the conclusion wasn't enough for most people.
DanNeely - Tuesday, August 5, 2014 - link
Where is this phone going to be available?Assimilator87 - Tuesday, August 5, 2014 - link
The phone's available in South Korea; I don't know what their carriers are. There's already a bunch on eBay as well.zodiacfml - Tuesday, August 5, 2014 - link
As I thought, minimal battery impact would be seen with higher resolution with their OLED. Would you mind testing for maximum brightness or if possible, your complete display analysis?repoman27 - Tuesday, August 5, 2014 - link
Any chance you'll instrument the GS5 LTE-A to measure power consumption directly as Anand did with the Galaxy S 5 (T-Mobile) and GS4 (AT&T)? I'm curious to see the difference in device power at idle with the screen on, to see how much of the battery life parity is due to overall platform power reduction or just the relative efficiency of the screen itself given the resolution.mike55 - Tuesday, August 5, 2014 - link
I would think a resolution increase on an OLED panel wouldn't have as much of an increase on power consumption as a comparable increase in resolution on an LCD due to the emissive nature of the technology. On an LCD with shrinking pixels, it's blocking more light from the backlight, so the backlight needs more power to achieve the same brightness. On an OLED you have more of the smaller pixels that have a similar luminous efficiency as the old larger pixels. I suppose the only power losses would be from higher resistance from smaller wires controlling the pixels and perhaps some other small power increases I'm not considering. Of course there's still the same increase in processing power needed to drive more pixels.lilmoe - Tuesday, August 5, 2014 - link
I'd prefer they make improvements to 1080p on a ~5" screen than move to a higher resolution panel. I just don't see the need for higher. The cons far outweigh the pros IMHO. Samsung can build a 1080 AMOLED panel with near-RGB strip arrangement (like in the Note 2). The clarity of the screen would drastically improve there, without needing to strain the internals. Not only does a higher resolution demand more CPU and GPU horsepower, but also bandwidth and more memory (and more internet bandwidth if the services are DPI aware). I'd rather gain considerably more performance and battery life with each iteration of a smartphone than utilizing that improvement in powering gimmicky features with diminishing returns. SoCs have improved so much in the past couple of years yet we slightly felt any difference with each iteration just because those SoCs are simply more strained than former ones. It's such a waste. I can get even more technical about this, but I'm tired of explaining myself.Really, 1080 is not only excellent, but absolutely amazing and mind blowing on a screen that size. Lets perfect what we have first before moving to the next (unneeded) level, sigh. Apple got away with 640p for more than 4 plus who knows how many more in the future. They took their time to perfect their internals to work with that resolution and DPI to the point that their smartphone is stupidly fluid. I see absolutely no reason why Samsung wouldn't get away with doing the same. Mind you, the panel on the S5 is leaps and bounds better than the on any iPhone.
Trust me people. Standardizing 1080 for smartphones would do a HUGE favor for the industry and the ecosystem. Having less things to worry about would make SoC manufactures concentrate more on what counts, not just optimizing the next big SoC to drive even more pixels...
repoman27 - Tuesday, August 5, 2014 - link
In what way does retarding the development of one particular component help the industry as a whole move forward? Screen resolution has doubled roughly 3 times in the past 20 years, while mobile SoC GPUs have doubled in performance almost 9 times in 7 years. In many ways, screen resolution has been waiting around for better GPUs; why take the pressure off now that we can finally drive these displays?lilmoe - Tuesday, August 5, 2014 - link
Because we've reached a point where more isn't necessarily better in particular components, and improvements need to be made in others. Current mobile GPUs don't even play 1080p games at 60fps with decent efficiency, and only a handful can tell a difference between of text or even video on a 1080p VS 1440p on a 5.x" screen. Why the heck would you improve certain components when you can barely tell the difference? Would you rather have a 1440p smartphone that runs OK or a 1080p smartphone that runs exceptionally well? 1080p on a 5" screen already looks superb, mind you. We're not in dire need of a higher resolution.When hardware can effortlessly and efficiently run EVERYTHING at 1080p and still has change to spare, then think of upgrading, otherwise I believe we're still not at that point yet. There's still lots of room of improvement for 1080p on a mobile small screen.
"In what way does retarding the development of one particular component help the industry as a whole move forward?"
I believe "retarding: is an excessive way to put it. You speak as if current 1080p screens are absolutely perfect. They're not. There's lots of room for improvement in various aspects.
repoman27 - Wednesday, August 6, 2014 - link
We're at 600 ppi right now, after stagnating at 90-150 ppi for over a decade. If we can double twice more, we'll be at 2400 ppi and will have satisfied the Nyquist criterion for human visual acuity. At that point you can draw at any arbitrary resolution you like and scale it with no visible aliasing. If you want to game, great. You can set a minimum frame rate and the game engine will just set the resolution given the capabilities of the SoC. Text and vector art can be composited using another layer at the full screen resolution and will be perfect. All we need is higher resolution screens and more flexible scalers and we can finally achieve resolution independence, which would be an epic achievement.When the standard sampling rates for digital audio were chosen, they didn't pick 44.1 kHz by accident; they did so because the limit of human hearing was right around 22 kHz. They didn't arbitrarily go with a lower frequency because most people over 30 have an attenuated hearing range. If we want screens where the subpixel arrangement is undetectable, and can accurately reproduce an image with zero aliasing even for young children and individuals with exceptional vision, then we should make an effort to get to at least 2150 ppi. If screens can be produced at such high resolutions affordably and with power requirements comparable to any of the other available displays, there is no reason not to use them in shipping products.
It is a fallacy to state that the current resolution race is adversely affecting any other aspect of mobile device development. The display in the GS5 LTE-A is a perfect example: in addition to being the highest resolution display ever shipped in an Android device, it is also the most accurate, and there was no net adverse effect on battery life. The only metric in which we see this screen lag ever so slightly is max brightness, but it still manages to beat the screen in the regular GS5.
lilmoe - Wednesday, August 6, 2014 - link
You're missing my point entirely. SoCs WILL get faster and more efficient even without increasing screen resolution. You're point is that higher resolutions don't affect performance, which is completely inaccurate. Your example on audio is relatively non- relevant. Different sampling rates have nowhere near the impact on performance as do different resolutions.You're comparing the GS5 with the GS5 LTE-A. Let me remind you that these are completely different phones with different internals. They just happen to have around the same battery life because the latter happens to be more efficient in driving more pixels, which balances things out. If you want to check if the extra resolution does in fact affect performance and battery life, then you'll have to examine the same device with the same internals against different resolution screens.
You have come to the conclusion that the current performance and battery life of today's smartphones are adequate. They're not. A GS5 LTE-A with a 1080p panel will surely run smoother, cooler, and longer on battery than the current one with 1440p. Even if the difference isn't very significant, it's still a difference that the average consumer appreciates. With all do respect, but I believe your priorities should be reevaluated. Those who care about 2000ppi and no visible antialiasing are an absolute minority, especially on a small ~5" screen.
440PPI is by no means the same as the 90-150 ppi we've been "stuck" with for decades. It surely is adequate for today's (and tomorrows) applications. What we "need" now is a faster, cooler performing devices, with 3+ days of battery life and highly optimized software. That's what the industry should be focusing on.
hung2900 - Tuesday, August 5, 2014 - link
For Amoled screen, with more pixels, each pixel can reduce the luminance to get equal brightness performance, and with a specific brightness the total luminance is the same. That‘s why we see virtually no difference in power consuming. However, there are two basically improvements:- Lower luminance means better durability (which is important with Amoled display)
- Theoreotical max brightness is higher, means in the next generation when other power consumption advances are applied, we can see the major bump in brightness (same as S4 to S5)
NerdMan - Thursday, August 7, 2014 - link
Why does RGB arrangement matter? Other than for the OCD?Pentile is not noticeable at this resolution. That's why it's 600 DPI.
Pentile is necessary, because blue OLEDs get half the lifespan of green and red OLEDs. So they make them twice as big.
soccerballtux - Tuesday, August 5, 2014 - link
I love the placemats from targetdjvita - Wednesday, August 6, 2014 - link
i may be in the minority but i wish for a 5" small bezel with a 4K sRGB display well calibrated, low power.and 4k in bigger screen sizes too
people saying this is good enough is the reason why there are still computers sold with 1366*768 TN panels (ugh)
i got a RAZR HD and i can see the pixels nevermind that my screen is all burned and yellowish now.
im upgrading till s810 comes out and better QuadHD panles come out.
galaxy nexus was the first 720p phone, how was the screen then?
DBDB - Monday, August 11, 2014 - link
The comparison does not specify smartphones NAND storage size ! it is known that 32GB NAND storage performs significantly faster on sequential writes.Can you confirm same size ( 16GB or 32GB ?? ) of NAND storage for all smartphones under test here ?