iPad Mini: Developers - Is there any truth to the scaling claims (iPad Mini, battery life, etc)?

[macbook123]

People claim that the reason Apple decided not to have a higher resolution on the iPad Mini is that they could have only gone with a full 4 times higher pixel density to maintain a simple scaling of Apps, but that having the same resolution as the big iPad would have required a a better graphics performance and (assuming everybody wants 12 hour battery life) a bigger battery and bigger iPad mini.

Is there any truth to these claims?

Is it actually true that it would be horribly hard for developers to make their Apps scale by a non-integer?

I'm asking this question specifically because I feel the iPad mini's resolution is unacceptably low. It would have been acceptable if it had been, say, 1440x900, without a need to go up all the way to the big brother's resolution, but the present compromise is just too much. My phone has more pixels than this iPad and is easier on my eyes.

[nuckinfutz]

It's about cost


Retina is still too expensive to put in a tablet around $300 and make profit. While their are some higher efficiency panels coming from Sharp and AU Optronics It's doubtful that the production of Retina could sustain both the iPad 4th gen and a mini.

More GPU would be required and RAM which means a hike in battery life as well.

We need to give it a year.

[coelacanth]

Basically, YES.

You'll have to "move" 4 times more pixels if it had the resolution same as iPad 3/4. That means better GPU, more RAM, and larger battery would have been needed to keep the same battery life.

They could have increased the ppi by introducing slightly higher res display (say 1280 or 1600 px wide) and minimize the performance impact, but that means all apps had to be tweaked.

Originally Posted by macbook123

People claim that the reason Apple decided not to have a higher resolution on the iPad Mini is that they could have only gone with a full 4 times higher pixel density to maintain a simple scaling of Apps, but that having the same resolution as the big iPad would have required a a better graphics performance and (assuming everybody wants 12 hour battery life) a bigger battery and bigger iPad mini.

Is there any truth to these claims?

Is it actually true that it would be horribly hard for developers to make their Apps scale by a non-integer?

I'm asking this question specifically because I feel the iPad mini's resolution is unacceptably low. It would have been acceptable if it had been, say, 1440x900, without a need to go up all the way to the big brother's resolution, but the present compromise is just too much. My phone has more pixels than this iPad and is easier on my eyes.

[macbook123]

Originally Posted by nuckinfutz

It's about cost


Retina is still too expensive to put in a tablet around $300 and make profit. While their are some higher efficiency panels coming from Sharp and AU Optronics It's doubtful that the production of Retina could sustain both the iPad 4th gen and a mini.

Why then are the Amazon and Google tablets MUCH cheaper despite having a significantly improved resolution. Also, if you read the original post again you will notice that I (intentionally) did not suggest Apple use a display with the same resolution as the big iPad.

Originally Posted by nuckinfutz

More GPU would be required and RAM which means a hike in battery life as well.

I would gladly take a screen with twice the resolution at half the battery life. Hell, I would even pay $100 extra. Or even better, as suggested earlier, 50% more pixels at a 33 cut in battery life.

The current resolution is simply unacceptably low. I got mine today and it is the saddest piece of electronics I have ever received because it it so fantastic in every other respect and this flaw seems literally like a cruel joke played by Apple.

----------

Originally Posted by coelacanth

They could have increased the ppi by introducing slightly higher res display (say 1280 or 1600 px wide) and minimize the performance impact, but that means all apps had to be tweaked.

I don't think you answered my question, which was: is it really so hard to tweak apps by a non-integer factor? I'm not an app developer but can not imagine that things get significantly more streamlined with a factor 2 than, say, a factor 1.5. If any folks with experience in this field could chime in, that'd be great.

[nuckinfutz]

Originally Posted by macbook123

Why then are the Amazon and Google tablets MUCH cheaper despite having a significantly improved resolution. Also, if you read the original post again you will notice that I (intentionally) did not suggest Apple use a display with the same resolution as the big iPad.



I would gladly take a screen with twice the resolution at half the battery life. Hell, I would even pay $100 extra. Or even better, as suggested earlier, 50% more pixels at a 33 cut in battery life.

The current resolution is simply unacceptably low. I got mine today and it is the saddest piece of electronics I have ever received because it it so fantastic in every other respect and this flaw seems literally like a cruel joke played by Apple.

----------



I don't think you answered my question, which was: is it really so hard to tweak apps by a non-integer factor? I'm not an app developer but can not imagine that things get significantly more streamlined with a factor 2 than, say, a factor 1.5. If any folks with experience in this field could chime in, that'd be great.

I'm guessing Google and Amazon are saving by using plastic and Apple's using Aluminum and better fabrication technique.

Things are much easier with 2x

http://stackoverflow.com/questions/9...d-3-for-assets

Imagine that Apple tells you that you can future proof your program by designing 1x and 2x artwork. So you do this with every single graphic element.

Then all of sudden they pull a fast one and deliver a display on an iOS device that delivers 1.5x resolution. That means as a developer you've got to go back and change every piece of artwork. You are not happy and probably will delay supporting the resolution change with then infuriates users and renders moot the whole reason for increasing the resolution.

Apple has to do it with minimal developer impact because if the developers don't support Retina then it's all for naught.

[seajewel]

Originally Posted by macbook123

I don't think you answered my question, which was: is it really so hard to tweak apps by a non-integer factor? I'm not an app developer but can not imagine that things get significantly more streamlined with a factor 2 than, say, a factor 1.5. If any folks with experience in this field could chime in, that'd be great.

I like how nobody in the thread answered your question. I'm not a developer but my SO is a programmer and scoffed when I asked if it was difficult to do. Hopefully someone will chime in who can help more. I posted in another thread that I think apple will eventually go with some in between resolution as the iPad 3 resolution is overkill on a 7.9" screen, so (as someone else came in responding) it actually sort of makes sense that Apple would make the iPad mini non-retina to "test the waters" so to speak, and let the device get hold while working flawlessly with existing apps, and then eventually introduce a new resolution that developers will have to develop for. no point in putting a pointlessly high resolution in that will suck battery life and processing power if ~250 ppi would be fine.

[coelacanth]

Originally Posted by macbook123

I don't think you answered my question, which was: is it really so hard to tweak apps by a non-integer factor? I'm not an app developer but can not imagine that things get significantly more streamlined with a factor 2 than, say, a factor 1.5. If any folks with experience in this field could chime in, that'd be great.

It really depends on the app, but the problem is that introducing 1.5x version from iPad 1, 2 or mini is not the same as x2 from them which already exists for iPad 3 and now 4. It's not simply x1.5 vs x2 problem. It's new versus existing problem. It is not that hard unless the app is bitmap image intensive (that can be the case for many developers tho), but it is not a "flip of a switch" kind of job.

I am a full-time UX/UI designer.

I don't think Apple's move was largely motivated by not giving developers hard time, but I do think they don't want to have resolution fragmentation on their hand like the Android side.

With better energy management, smaller chip architecture, component price reduction, etc, I think they'll bring the retina display (same pixel count as current iPad) iPad mini in the future rather than introducing 1.5 or whatever new resolution one.

[nuckinfutz]

Originally Posted by seajewel

I like how nobody in the thread answered your question. I'm not a developer but my SO is a programmer and scoffed when I asked if it was difficult to do. Hopefully someone will chime in who can help more. I posted in another thread that I think apple will eventually go with some in between resolution as the iPad 3 resolution is overkill on a 7.9" screen, so (as someone else came in responding) it actually sort of makes sense that Apple would make the iPad mini non-retina to "test the waters" so to speak, and let the device get hold while working flawlessly with existing apps, and then eventually introduce a new resolution that developers will have to develop for. no point in putting a pointlessly high resolution in that will suck battery life and processing power if ~250 ppi would be fine.

Depends on what platform you're targeting and how they draw their UI.

Android has always had many resolutions so their development tools support a wider variety of resolutions.

Apple has had a more conservative approach where pixel perfect UI are developed because the programmer knows exactly how many vertical and horizontal pixels he/she has to work with.

Here's the problem. You'd need to take the panels that are used in iPhone 4 and 4S and start cutting larger 7.85" screens. One problem though the LTPS backlight in iPhone 4 is very expensive at larger sizes.

So Apple needs a more efficient technology to move to (IGZO) which is going to be cheaper in larger sizes. The resolution isn't going to be that much of a problem once manufacturing catches up.

Sharp is working on putting 443 PPI across a 5 inch screen

So given the maturation of their production I don't see where a 2048 x 1536 7.9" display is going to be a problem.

[macbook123]

Originally Posted by coelacanth

It really depends on the app, but the problem is that introducing 1.5x version from iPad 1, 2 or mini is not the same as x2 from them which already exists for iPad 3 and now 4. It's not simply x1.5 vs x2 problem. It's new versus existing problem. It is not that hard unless the app is bitmap image intensive (that can be the case for many developers tho), but it is not a "flip of a switch" kind of job.

I am a full-time UX/UI designer.

I don't think Apple's move was largely motivated by not giving developers hard time, but I do think they don't want to have resolution fragmentation on their hand like the Android side.

With better energy management, smaller chip architecture, component price reduction, etc, I think they'll bring the retina display (same pixel count as current iPad) iPad mini in the future rather than introducing 1.5 or whatever new resolution one.

Very informative answer. Thanks. So how hard do you think it would be for Apple to do resolution fragmentation? Isn't it something that will be natural to do in the long run given increased size fragmentation of iOS devices?

[nuckinfutz]

Originally Posted by macbook123

Very informative answer. Thanks. So how hard do you think it would be for Apple to do resolution fragmentation? Isn't it something that will be natural to do in the long run given increased size fragmentation of iOS devices?

At some level it's already being worked on. There's an API called Auto Layout which allows for my flexible interfaces by allowing you to set constraints on UI elements. It now works across iOS 6 and OS X.

Ideally it would be great for say working with multiple devices that only vary slightly in resolution.

http://www.iphonehacks.com/2012/06/a...ne-screen.html

[sammich]

The other questions are pretty much answered.

One thing I will say is that, by forcing App developers to make a specific UI layout for the iPad and the iPhone separately, and making it obvious to the end user that, 'Hey, this app isn't universal/an iPad version', then it means that every iPad app is tailor made for the iPad (more or less).

Sure, this really only works if the iPad is a success (and from what I hear, it's not really much more effort to create an iPad specific layout. It's more of making it a functionally better-on-a-larger-screen type deal).

The above two contrast with the way Android on tablets has developed. Little compelling market to develop for, and since apps can easily just 'expand' their small screen designed UI layouts, then there's little point to making a high quality, specific larger screen effort.

But as others have mentioned already, iOS6/ML SDK have flexible layouts, so I wonder where that will lead us. (Question for devs: are these flexible layouts still segregated into iPhone/Touch and iPad targets? As in, can you build one UI for both?)

Originally Posted by macbook123

Why then are the Amazon and Google tablets MUCH cheaper despite having a significantly improved resolution.

They sell services. After the retail cut from the Nexus 7 (I'm not sure if the Nexus 4/10 will go to retail), I don't think Google makes much money, if any. The Nexus program also exists to show off Android, and to give OEMs examples of what a quality Android device is. And Amazon's part is pretty sell explanatory.

Apple's game, in contrast, is to provide masses of value in build quality, product experience and in store service. They can't really do that without turning a decent profit on the things they sell.

You also can't forget that Apple has cornered the market on premium computing, and pretty much owns the premium tablet market (for now) and a decent chunk of the premium (not necessarily spec-wise) handset market. Apple's premium brand actually spoils a bit if they drop the price too much. Almost like there must be something missing.

So, in some sort of conclusion: I wish the iPad Mini had a higher resolution. I'd totally love it.

Tell me, honestly, do I sound a little fanboyish?

[macbook123]

Thank you all. I appreciate the responses.

I will still say that I still don't have a sense of *how* difficult it would be to have intermediate resolution jumps, like on Android, and why this is in fact prohibitive enough to justify coming out with an iPad mini with crippled resolution, as opposed to something sensible that's between the resolution of the big iPad and that of this mini, say 1440x900.

I get that people are saying it is difficult, but difficult is a relative word, and I just haven't been convinced that it is difficult enough to stall progress in such a tragic way as has been done with the Mini (it being such a great product overall, with such an obvious flaw pointed out by every reviewer out there, and undoubtedly obvious also to people at Apple).

[-Garry-]

Originally Posted by macbook123

Thank you all. I appreciate the responses.

I will still say that I still don't have a sense of *how* difficult it would be to have intermediate resolution jumps

I think you did get your answer. It depends on the app. If an app is using mostly native Apple controls then not so difficult at all. Take the Apple contacts app for example. That will have been relatively easy to rejig.

On the other hand, if you have an app using mostly custom views you've got to ensure every view scales (or have different views presented, depending on the device.) That can be an awful lot of work... so much in fact that some app developers just wouldn't feel it was worth doing.

One of Apple's biggest strengths is the lack of fragmentation and the availability of high quality apps. Sure, other non-Apple tablets have higher resolutions in a number of different configurations, but the number of quality apps taking advantage that work well on all tablets is much less.

I'm sure a retina iPad mini will come, but not until retina-quality 7.9" screens are cheap enough. In addition you also need a CPU/GPU which will run cool enough to push the pixels without burning up, and a battery to power both of them for a reasonable length of time.

[nuckinfutz]

Originally Posted by macbook123

Thank you all. I appreciate the responses.

I will still say that I still don't have a sense of *how* difficult it would be to have intermediate resolution jumps, like on Android, and why this is in fact prohibitive enough to justify coming out with an iPad mini with crippled resolution, as opposed to something sensible that's between the resolution of the big iPad and that of this mini, say 1440x900.

I get that people are saying it is difficult, but difficult is a relative word, and I just haven't been convinced that it is difficult enough to stall progress in such a tragic way as has been done with the Mini (it being such a great product overall, with such an obvious flaw pointed out by every reviewer out there, and undoubtedly obvious also to people at Apple).

The problem is your point of view. You see the mini display as being "crippled" or 'tragic" yet it's the same resolution that shipped on the first two iPads.

for those sensitive to Retina display, the best option it to wait for the next revision and hope it has Retina. Once you ask developers to make a small incremental change you've asked them to support that change for the lifecycle of that product.

[Pompiliu]

Originally Posted by macbook123

I'm asking this question specifically because I feel the iPad mini's resolution is unacceptably low. It would have been acceptable if it had been, say, 1440x900, without a need to go up all the way to the big brother's resolution, but the present compromise is just too much. My phone has more pixels than this iPad and is easier on my eyes.

1440x900 is not 4:3.
You must be american.

[macbook123]

Originally Posted by -Garry-

On the other hand, if you have an app using mostly custom views you've got to ensure every view scales (or have different views presented, depending on the device.) That can be an awful lot of work... so much in fact that some app developers just wouldn't feel it was worth doing.

But isn't that true for a scaling factor of 2 in the same way it is true for a scaling factor of, say, 1.5? If not, how much more CPU or GPU intensive would the latter be?

I'm not sure this is the same situation, but on my RMBP I don't notice any slow down in performance between the 1440x900 setting and my preferred 1920x1200 setting. When talking about the iPad, are we looking at a similar situation?

[NT1440]

Originally Posted by macbook123

Why then are the Amazon and Google tablets MUCH cheaper despite having a significantly improved resolution. Also, if you read the original post again you will notice that I (intentionally) did not suggest Apple use a display with the same resolution as the big iPad.

.

Because Apple feels the 16:9 and 16:10 aspect ratios are not great for reading or web browsing. Using a wide range of tablets at work, I have to agree.

The new tablets have high PPI screens, but they are awkward for me to use. No one produces retina quality displays in the mini's aspect ratio in nearly high enough volume yet.

IGZO will be a major splash in the tablet market in 2013.

[macbook123]

Originally Posted by NT1440

Because Apple feels the 16:9 and 16:10 aspect ratios are not great for reading or web browsing. Using a wide range of tablets at work, I have to agree.

The new tablets have high PPI screens, but they are awkward for me to use. No one produces retina quality displays in the mini's aspect ratio in nearly high enough volume yet.

IGZO will be a major splash in the tablet market in 2013.

Very interesting point that I was not aware of, and I completely agree with you about the aspect ratio (and was quite surprised about the new choice for the iPhone 5).

[NT1440]

Originally Posted by macbook123

Very interesting point that I was not aware of, and I completely agree with you about the aspect ratio (and was quite surprised about the new choice for the iPhone 5).

I was surprised by the iPhone 5, but using it I really like it. I hate the S3 and basically any "two hander" phone. My phone is supposed to be as simple as slide out of pocket, bang out a text with one hand, and slip it back into my pocket. I don't like phones that are so big that you have to use both hands and focus on the device in full.

That's just my use case though.

[Rodster]

Originally Posted by macbook123

Why then are the Amazon and Google tablets MUCH cheaper despite having a significantly improved resolution.

The answer for Amazon is simple. When you buy a Kindle Fire it opens the device to Amazon's store front which they hope you will use to buy lots of stuff from them. That's why they sell it at cost. I have an Amazon Prime membership so it worked for me.

I suspect it's the same with Google plus they want the tablet market share so devs make apps for them.

[macbook123]

Originally Posted by NT1440

I was surprised by the iPhone 5, but using it I really like it. I hate the S3 and basically any "two hander" phone. My phone is supposed to be as simple as slide out of pocket, bang out a text with one hand, and slip it back into my pocket. I don't like phones that are so big that you have to use both hands and focus on the device in full.

That's just my use case though.

I agree with that as well, however of course people's hand sizes vary about as much as their body heights, i.e, with a standard deviation of about 10%. So whereas in my case a 4.5 inch phone will be a one-hander phone, for another person it would have to be a 3.5 inch phone. For somebody like me taller than average, the 4.5 inch screen is clearly preferable because it fits more information and is more comfortable to read. This was the reason I switched to Android phone-wise, after hoping for years that Apple would become aware of this natural variation in human populations when defining the "right" size for a phone.

I still like the wider aspect ratio better, and would have stuck with an iPhone had the 4" upgrade of the iPhone 5 stuck to the same as that of the iPhone 4, instead of becoming a "skateboard" phone as Steve Jobs put it...;-)

[macbook123]

I had a follow-up question on this thread: what about the fact that Apple deviated with the iPhone 5 in the scaling convention, compared to all previous iPhones for which developers still need to taylor their apps? Why would the same non-integer increase in resolution not be possible for the iPad, except for along both dimensions instead of one?

[zhenya]

Originally Posted by macbook123

Thank you all. I appreciate the responses.

I will still say that I still don't have a sense of *how* difficult it would be to have intermediate resolution jumps, like on Android, and why this is in fact prohibitive enough to justify coming out with an iPad mini with crippled resolution, as opposed to something sensible that's between the resolution of the big iPad and that of this mini, say 1440x900.

I get that people are saying it is difficult, but difficult is a relative word, and I just haven't been convinced that it is difficult enough to stall progress in such a tragic way as has been done with the Mini (it being such a great product overall, with such an obvious flaw pointed out by every reviewer out there, and undoubtedly obvious also to people at Apple).

When you scale elements to a non-exact multiple, there will be visible artifacts.

Here is a simple vector graphic drawn at 1x to 2x illustrating the differences as you bridge the gap between the two.



The only multipliers where it is perfect are at 1x and 2x.

Apple is a company founded on strict control and perfection. It's not surprising they have chosen to only support the method that works perfectly.

Originally Posted by macbook123

But isn't that true for a scaling factor of 2 in the same way it is true for a scaling factor of, say, 1.5? If not, how much more CPU or GPU intensive would the latter be?

I'm not sure this is the same situation, but on my RMBP I don't notice any slow down in performance between the 1440x900 setting and my preferred 1920x1200 setting. When talking about the iPad, are we looking at a similar situation?

There is absolutely a difference in performance on your rMBP when using the scaled resolutions. Read the reviews and you can see exactly how quickly performance drops. It requires a huge amount of graphics horsepower to re-draw the screen to make it look as good as it does.

Originally Posted by macbook123

I had a follow-up question on this thread: what about the fact that Apple deviated with the iPhone 5 in the scaling convention, compared to all previous iPhones for which developers still need to taylor their apps? Why would the same non-integer increase in resolution not be possible for the iPad, except for along both dimensions instead of one?

They did not scale anything with the iPhone 5. Apps that are not updated for the new phone run with black bars at the top and bottom, which, while not ideal, is also not all that noticeable. The bars that would be necessary on an iPad would be huge.

This is an example I made for another thread. The home screen is 1024x768 while the white box is 1536x1152 (exactly 1.5 times the resolution).

[DJinTX]

Originally Posted by zhenya

This is an example I made for another thread. The home screen is 1024x768 while the white box is 1536x1152 (exactly 1.5 times the resolution).

Image

Hey Zhenya,

Could you explain this graphic for me? You obviously have a lot of technical knowledge, but I am having trouble understanding this part. I'm sure you're making a great point related to this discussion, and I just want to be able to follow along.

Thanks~

[Phil A.]

Originally Posted by macbook123

My phone has more pixels than this iPad and is easier on my eyes.

I don't know what phone you have, but the mini has more pixels than any phone Apple have produced