Apple's upgrade dilemma | Hacker News Reader

Apple's upgrade dilemma

virtualpants.com

27 pointsjharrier46 comments

Sloppy, content-free article that is basically half a dollop of pure speculation and little else.

"However, it won’t be long until it becomes an attractive option to increase sales."

Seriously? Um, it's already an attractive option. Always has been, if you are a lazy, sloppy company. Apple's eschewed this option approximately forever, and there's no evidence of anything different.

Agree with most of what you said except "Apple's eschewed this option approximately forever".

Siri for only iPhone 4S when iPhone 4 hardware could support it was one chink in their impecable reputation for supporting older hardware as long as possible.

This is because the iPhone 4 is missing a newer version of the crucial noise reduction chip:

http://www.tuaw.com/2012/02/06/siri-may-be-iphone-4s-only-be...

The fact that Apple supported iOS 6 on the 3GS is proof that they're pretty serious about backwards compatibility.

If this was the case, they should have communicated it.

As an iPhone 4 user, I chose not to upgrade to iPhone 5 when my contract expired because I feel like whatever gets released this coming year (5S/6) might not be compatible with iPhone 5 software wise.

They did.

They did communicate it. They have communicated it for a couple years now.

It wasn't part of the announcement.

That is not true. Siri was available in the App Store from a third party for at least a year prior to its reintroduction in the 4S, and ran on a 3GS just fine. There isn't a technical reason for it not being available on earlier models.

http://techcrunch.com/2011/10/04/the-original-siri-app-gets-...

Siri the APP was available. That's different from what Apple shipped.

And especially given Apple's need to work across multiple languages we don't know if a more sensitive microphone was required.

Define "Just fine." The argument is that Apple decided not to release their version of Siri on the 3GS because a hardware difference meant that the voice recognition accuracy would have been poorer.

You can't really counter that claim with the simple observation that an earlier version of Siri was available for the 4 and 3GS; that's just talking past the point. A more appropriate counter would be a source saying that the voice recognition accuracy of that version of Siri was just fine. Or at least no worse than what ended up being released on the 4S.

Come on, what about VIP in Mail.app and Reading List in Safari being unsupported on older devices. There's no way that the 3GS lacks some required hardware feature. Is there?

VIP is available for 3GS.

It is just the Offline Reading List and it could be related to the memory.

How do you explain the fact that Google Voice Search on iPhone 4 works so good that its recognition is faster then Siri on iPhone 4S?

I'd say that's one where the most painfully obvious answer is likely also the correct one: Google is using better voice recognition technology.

What about when iOS 4 came out and updating rendered the 3G completely unusable? I know forward progress is great, but I felt really screwed over and forced to upgrade.

Absolutely agree. Stopped reading after this:

"However, anecdotally, it seems that people are upgrading their iPhones and iPads in longer cycles."

The entire article is based on personal observation. This is has very little to do with the wider market.

I have seen this argument applied to Dell, Microsoft, Apple and others over the years. It is really a straw man argument. Apple will not kill off old hardware just to boost new sales. Apple keeps their offerings fairly current in order to push the envelope. When Apple dropped the floppy drive they were not trying to screw over consumers, they were looking forward and realizing that documents would not fit on a 1.4mb storage medium for much longer and they were right.

It's a matter of leaning forward or leaning backward. Another strategy Apple will likely pursue is to continue to break into new markets. The TV from Apple is likely the next innovation and area of growth that will boost Apple's outlook and get stock values up again. And that may happen this year. And beyond the TV I think Apple will be exploring other areas. We know that smart cars are coming with iOS integration. That will generate a good deal of revenue for Apple. Products like Nest and Bluetooth LE devices will also become quite handy. Just look at the Nike Fuelband, Jawbone UP and Fitbit for examples of new opportunities for hardware which integrates with the iPhone and gives consumers new gadgets to buy which leverage Apple products. Also keep in mind products like Pebble which are smart watches which integrate with the iPhone and Apple may even create their own smart watch which would surely do well given how popular the idea is with Pebble.

Apple does not need to get people to buy new iPads more frequently. They can simply create more products we are willing to buy.

> When Apple dropped the floppy drive they were not trying to screw over consumers, they were looking forward and realizing that documents would not fit on a 1.4mb storage medium for much longer and they were right.

No, they dropped the floppy disk for aesthetic reasons. Steve was adamant about there being only a trayless CD drive, and USB. The original iMac had a CD tray, but it was switched to trayless in the first revision.

This came back to haunt them, as early CD burners only came with a tray, but aesthetics trumped utility.

Actually when Apple dropped the 1.4mb floppy the Zip Drive was becoming popular which was before the USB thumb drive was popular. And now people rarely use thumb drives and instead use Dropbox, Box.net, Skydrive or Google Drive. Maybe it was partly due to aesthetics, but it was also looking at how the average file for many documents (photos, videos, etc) are generally bigger than 1.4mb. It was a good choice. I use to own the early iMacs. I worked on them quite a bit actually. They were great.

Everyone does this - it's not unique to Apple. How many single core phones run Android 4 from the factory?

It's probably less about upgrade revenue than the new opportunities in the SoC space which, unlike desktop/server CPU's, are increasing in performance/W at a much greater clip.

Software tends to expand to fit hardware capabilities, so as newer hardware comes out, software is written to use, then require the added performance.

If Apple could have shoehorned iOS 6 onto the original 256MB RAM iPad, they probably would have, but it probably performed so badly they avoided it.

> Everyone does this - it's not unique to Apple.

Definitely. Just look at the Android OEMs (HTC, Samsung, etc). They're always at an impasse to whether they should continue to update their devices to the latest Android version (and avoid angering customers) or stop updating to boost more new phone sales. On average though, it seems the assumption they can sell more devices seems to win out for Android OEMs than updating.

Most Android users have no idea what Android version they run (and what the latest version is) and I think the lack of being informed allows the Android OEMs to get away more often with not pushing updates. Android users figure out they need a new device when they realize they can't run a certain app or the apps they use no longer perform as well. Some probably also figure it out by comparing with friends' devices.

Only exception to the above really are Nexus devices since Google has no great vested interest really in selling more of them and benefits more from everyone having the latest Android version. Unfortunately, Google seems too worried if they pressure OEMs to push more updates in an assertive manner, they will suffer backlash from them.

edit: typos

impasse is spelled without the accent ;)

Actually, continually providing upgrades to older iPhones is a wonderful way to push new hardware sales: the upgrades significantly degrade the user experience because the hardware isn't as capable.

I've had issues with iOS upgrades before, but they were all fixable. Usually by an annoying backup and restore. Even iOS 4 on my 3G worked OK.

I'm sure others haven't been so lucky.

The idea that Apple is purposely introducing bugs to degrade user experience is silly. There is no conspiracy.

I wasn't implying they are introducing bugs intentionally. In the normal course of adding capability to new releases of an operating system tailored to newer phones with better hardware, it's only natural that older phones will run those new operating systems slower and the experience will degrade. Nothing conspiratorial about it. But, it works really well for Apple, because they can charge you to upgrade to the new iOS version, and then in 6 months when you're sick of your laggy, unresponsive old-ass iPhone, they can helpfully sell you a new one.

I am completely behind the longer-term upgrading of your phone. But only once you find your match.

I have changed over 8 phones in 2 years and it has been actually rather frustrating. It was at the time when smartphones were coming out. Of course , as a geek, I had to have one. It was nightmare. Displays were unresponsive, phones were freezing etc. I end up using old keyboard-standard nokia for few years.

About a year ago, I have been using Galaxy for few months. The experience was much better then my first phones, but still not perfect. And then I bough my dream phone. Many of you might not know it, but its called XiaoMi (小米). It is chinese phone, but not some chinese garbage it is well thought-trough phone, performance much better then newest galaxy and armed with android and MIUI on top. Peope say that Xiomi company is like apple in china. I have to say I have never been so happy with phone.

XiaoMi now released version two of their phone. The phone has everything that first version is missing, and I am very temped to buy it, but there is a bit of fear if the phone wil let me down the same way my first phones did. And i am so happy with my current phone, why would I change? Maybe in few years....

Apple is already killing off device support, sometimes (seemingly) for marketing reasons. With iOS 6, Apple dropped support for the first-generation iPad, but continued to support the iPhone 3GS (which I cannot imagine being much faster [edit: see Xuzz's reply to this comment]). The obvious reason, from my layman's perspective, is sales: the iPhone 3GS was on the market more recently, so the (newer) iPad got the ax.

Apple also tied app support for the iPhone 5, and future iPhones, to ending support for the iPhone 3G and its ilk; I understand that long build times are far from desirable, but this move seems like a great way to encourage upgrades.

Then again, maybe I'm just a cynic.

Edit: I've not used an iPhone 3GS on a day-to-day basis, but I've seen better performance out of my first gen iPad (and maybe it's just me) just before it was EOL'd than I ever saw with the iPhone 3G/iPod touch 2G family, both of which were truly awful performance-wise before Apple pulled the plug. I'm guessing the real reason for keeping support for the iPhone 3GS but not the iPad 1G was, as coob mentioned, the fact that it was on sale when iOS 6 was released.

I assumed dropping the iPad had more to do with pushing all those extra pixels and low memory problems.

The iPhone 3GS is much faster than the original iPad. On the iPhone 3GS, you have a reasonably slow CPU and GPU, but you also only have 320x480 pixels to deal with. On the iPad, you have a slightly faster CPU and GPU (but still single core), but have to render to 1024x760 — way more pixels. You also don't have any extra RAM on the iPad, but your images and framebuffer still have to cover that extra area.

The net result is that the iPhone 3GS is much faster in practice than the original iPad (or the iPod touch 4G, which has similar issues to a lesser extent — but which they still sell, so can't drop support for). Another similarly underpowered device is the iPad 3, so I'd expect them to drop support for that even before the older iPad 2.

Original iPad is even faster than iPhone 4

http://browser.primatelabs.com/geekbench2/compare/1635302/16...

So it's faster than a 3GS http://browser.primatelabs.com/geekbench2/compare/1634913/16...

Yes, and you will note I said "slightly faster CPU and GPU". The point is that the whole device is much more resource constrained, as what needs to be done with the CPU and GPU on the original iPad is more than on the iPhone 3GS, but the CPU and GPU are not faster enough to create the same effective performance.

Yes, raw performance may be better on the original iPad. But raw performance is irrelevant if you can't get the same effective performance for the user.

I see

I find safari on an original iPad almost unusable; after upgrading from iOS4 to 5, it crashed regularly - I presume due to not having enough free memory. Support stopped at 5.1, just under 2 years after its international release, and only 1 year after it was replaced by ipad2.

Seems like they already have a handle on deprecation.

It's not particularly random. Apple supported the iPad 1, iPhone, and iPhone 3G with 2 major software updates + all minors within those major releases. That pattern is now breaking because Apple has continued to sell the iPad 2 and iPhone 3GS for much longer than those other products were available. A more interesting argument might be: given the technical constraints of the 3GS and the fact that it continued to be sold until September 2012, will Apple support it in iOS 7? It seems unlikely, but maybe we'll all be pleasantly surprised. Certainly their choice to go downmarket by selling the 2 previous models has forced them to support those devices for longer timeframes than they may have originally envisioned.

The 3GS supported iOS6 because Apple were still selling it at the time. They were not selling the original iPad.

In other words, they had to support iOS 6 on 3GS, but for iPad 1 it was optional. Hence they didn't support iPad 1.

It's mostly strange because iOS 6 didn't really add any performance-critical changes, it was mostly changes to iCloud. Unlike iOS 4 that added multitasking, for example. Well, the new Maps.app is likely more demanding than old one, so perhaps they kept back the entire iOS 6 because of that.

The excellent iOS support matrix makes the decision more obvious: http://www.flickr.com/photos/89549358@N02/8454676411/sizes/l... As you can see, the geekbench score for the iPad is indeed higher than the 3GS but the 256MB system memory coupled with the display size likely meant performance wasn't acceptable. Since there's no paging on iOS, Apps are killed when they consume too much memory. Since the iPad already has less system memory available (because more is being used for graphics) it would lead to instability in existing Apps, and be harder to support.

Yeah, it was memory. Apple undershot the onboard DRAM pretty badly on that device. Really it was a design mistake, not a deliberate support termination.

My almost two year-old HTC Thunderbolt was supposed to be the "next big thing" when I bought it. It finally got upgraded to ICS Saturday (yes ... 2/9/2013) morning. It takes an effort to support the hardware you produce, but if you're putting people into two year contracts, it's pretty pathetic that the upgrade finally occurred within two months of my renewal (OTOH, I won't be trading this for another HTC phone as I've learned my lesson).

I'm another person that's perfectly content with my iPad 1, I have a laptop that's working great (with Ubuntu) after 6 years and I tend to wear things out before I upgrade. I was disappointed with Apple's "throw-away" chassis on the new MacBook Pro and it seems like the trend is only getting worse.

Heh...same phone, but my plan upgrade came the day they deployed the update. I learned my lesson too...bought a Samsung.

I'm leaning towards a Samsung phone as well ... which model did you pick and how do you like it (so far)?

I got the galaxy S3. So far it is amazing. It is great in all the technical features and despite the larger screen it is lighter and more comfortable in my pocket. And the battery lasts twice as long as my thunderbolt at its end of life.

Note that the iPhone/smart phone market is in no way saturated - there are many dumb/feature phone users to sell to. We only just passed 50% of the US market being smart phones. There are currently about 6.5 billion mobile phone subscribers worldwide, but only about 1 billion smart phones. That leaves a market five times the size of the existing smart phone base, or in other words the smart phone market is only 16% saturated.

The one thing I can think of that has effectively EOL-ed older iPhones is dropping support for ARM v6 codegen from recent versions of Xcode. It is not possible to build against the iOS 6 SDK (which is needed to access newer APIs) and support the 3G & below.

I'm no expert on compilers, but I find it hard to see how that isn't deliberate, or what other benefits it provides.

Dropping support is usually deliberate but not malicious. In this case, I suspect it was one less toolchain to maintain and run QA on for each release.

It was entirely deliberate. You can't write an app to support both iPhone 5 (with full height support) and iPhone 3G and put it on the AppStore, and use the normal compiler. Apple don't support the large image with < iOS 6 SDK, and they don't support armv6 with the iOS 6 SDK.

Reply on news.ycombinator.com