iPhone or Droid Incredible? In other words, is AT&T really so bad? (2 Viewers)

[DaveF]

 

How does anyone look at the market as it stands and make that statement?

He does, by looking not at number of phones sold but at dollars earned. It's related to the question of what matters more: marketshare or profits?

 

This graph shows the shift in earnings from the launch of the iPhone to June of this year.

http://www.asymco.com/wp-content/uploads/2010/08/Screen-shot-2010-08-13-at-8-13-11.41.20-AM.png

 

It's very dramatic: earnings have been transferred to Apple. He notes that the cellphone market has not grown in size, so Apple's growth from 1% to 48% of market revenue (of analyzed companies) means that it has taken the revenue from the other companies. Currently it's a zero-sum game and Apple's benefit is Samsung and Motorola's loss. Also, RIM' earnings increased.

 

What does this mean? If this captures Android's rise in popularity, then revenues are shrinking even as more Android phones are sold. The extreme interpretation of this is that selling ever more high-price phones at very slim, or even negative, margins and losing money on it.

 

The unanswered question is whether Q2 2010 captures enough of Androids recent growth to reflect what's happening. And so I want to see this analysis again in 6-12 months. If Q1-Q2 are representative of Android's popularity increase, it suggests a world in which usage market share is increasing but revenues are not -- which is both possible and a very bad place to be. If it does not capture enough of the Android's growth, then we'll see a reversal of the earnings shift away from Apple and back to the other companies.

 

And the exclusion of HTC could be a major spoiler, as well.

 

I don't know that this analysis is sufficient, or conclusions correct. But I do know that "marketshare" can be a canard. That's how Apple can have a modest 10% of the "PC" marketshare but be tops in profits.

 

[Hanson]

The issue for Apple, in light of the rumors of a Verizon iPhone 4, is that Apple charges AT&T more money for their phone than other manufacturers charge. And that's because of the exclusivity deal that they signed. The challenge for Apple, if they do dip their toe into the Verizon pool, is the loss of profits in exchange for marketshare. It's a given that a Verizon iPhone would increase their marketshare, but they would lose the AT&T pricing advantages. And Verizon is taking a very public position that they are very happy with Android, either because it's true or it's a negotiation tactic with Apple.

 

Also, all of the analysis dutifully ignores the elephant in the room. RIM is pathologically slow footed, just like he claims the large manufacturers are, and yet it is ignored in the analysis because it doesn't fit in with anything the author is saying.

 

And his suggestion for Google to avoid the large manufacturers and strike out on their own with a small vendor is ludicrous. They have tried that, in effect, with the Nexus One (under contract from HTC), but it tanked. The iPhone has become it's own product, but when they launched it, the hook was that it was the top of the line iPod you could make calls with. It had a lot of other bells and whistles, but Coverflow was the killer app. Android simply has nothing like to make a big splash in the smartphone market.

 

And the reason WinMO failed is because the market shifted overnight from hard keyboard interfaces to touch only with the introduction of the iPhone. It wasn't the vendors with were slow to respond, it was MS itself, who just could not crank out a touch interface worth its salt. HTC tried in vain to make the basic interfaces palatable, but in the end, WinMo was a jury rigged failure that lost not only the hardware manufacturers, but the actual company that made it.

 

[Hanson]

http://today.msnbc.msn.com/id/38635041/ns/technology_and_science-wireless/

 

[DaveF]

No, the biggest reason is that it miscalculated how much a prolonged exclusivity with AT&T would cost.

The article is wonderfully vague. The Verizon iPhone is "too late", the author writes. Too late for what?

 

And I love made-up numbers presented as "daming" facts:

The market research firm iSuppli just put out the most damning numbers: In terms of global smart phone market share...

 

It may be that Apple kept AT&T exclusivity too long to achieve some unspecified optimal results as measured against an unstated merit function.

 

(For my part, I think it's possible Apple mis-judged Android's ascent and may be late in getting an iPhone on other carriers to achieve as much profit as they could have, in the mid-term.)

 

[Sam Posten]

Real world battery tests:

http://blog.laptopmag.com/android-battery-test-reveals-droid-x-lasts-longest-amoled-handsets-trail

 

I'd like to see how the original iPhone, 3g, 3gs and 4 stack up on this list. I'd bet the original would hit about 4 (ie worse than any of these) and the 4 would be close to a 10, ie better than all of them.

 

Battery life isn't the be all consideration of course, but it ranks up there.

 

[DaveF]

It's interesting to see the anecdotal comments about the sub-par HTC Incredible battery life confirmed demonstrated in a controlled way.

 

Apple claims:

 

As they are underestimating battery life for the iPad, this is probably about right for the iPhone 4. But I don't know if the particular form of this test would affect the results.

Internet use: Up to 6 hours on 3G Up to 10 hours on Wi-Fi

 

[Hanson]

I would say that 5.5 hours on a full charge is about what I get on average for my Evo. However...

 

I think the methodology is completely flawed. The screens are on for the entire duration and there is too little browsing going on (1 page per minute is quite atypical use). The way it's set it up, power consumption appears to be almost wholly affected by the efficiency of the screen than anything else. And as has been pointed out in the comments, simply setting to 40% on each device is completely wrong -- it is quite possible that the reason the Motorola phones do so well is that they the 40% brightness level is dimmer than the 40% brightness level in the HTC or Samsung models. So this test isn't even a proper way to stack up the phones against each other.

 

Here's the thing I'd really like to know, if these tests are "real world" -- my battery log shows LCD Screen use to be typically 8-15% of battery use (the OS and cell phone stand-by are the biggest juice hogs). If that number is significantly higher after running these tests (like, greater than 50%) than the test has no bearing on the real world, and any battery life similarities to mine are strictly coincidental.

 

The method they use may be fine for laptops, whose screens may normally be using most of the juice, but it has very, very little to do with "real world" conditions for cell phones. Just think how long the iPhone 4 would last if the screen were on for 7 hours straight (if it lasts that long). Because that's all this test is really measuring.

 

Oh, and in case anyone is shocked that I only get 5.5 hours on a full charge, you have to realize that I multitask on it like crazy. There are many Evo users who reportedly get 16-20 hours, but I guess they barely use theirs. Plus, having a fully charged spare is a wonderful luxury -- I can literally use the phone indefinitely because by the time one battery depletes, I have a fresh one ready to go. On the weekends, I only charge my phone while I'm sleeping (and that's just out of habit since I really don't have to cradle the phone).

 

Oh, and I have the screen set to auto-brightness. It seems like a waste to make the screen dim to squeeze out a few more minutes.

 

[Sam Posten]

I want to see the test redone with flash =)

 

[DaveF]

Originally Posted by Hanson Yoo But to each their own.

 

These are a specific set of tests, that show how much battery life you get solely in web surfing mode. One minute per page sounds like a fair compromise: some pages you breeze through and others you actually spend some time reading. And it's no more or less a measure of how much total time you get using the browser.

 

You're right that if they don't, they need appropriate control of screen brightness, at least to make sure they correspond to "normal" usage on those devices.

 

But while "day in the life" tests are interesting, (a mixed-use case of phone, 3G data, wifi data, email, video, music, etc.) I now prefer single-activity tests. They're easier for me to understand and relate to my normal usage. In a mixed usage test, I have no idea if their proportions of the activities match my usage. Iif they say it gets 16 hours of music playback, then I know how long it will last playing music during a car trip.

 

[Hanson]

What I am saying is that this is essentially a test of screen efficiency since the amount of work that the phone is doing for browsing will have very little impact on battery life (my browser uses 1-3% of the battery even with very heavy use). But most people don't actually do things that require their screens to be on for 5 straight hours. So as a "real world" test, it is total fail. And because there is no consistency between "40% brightness", it's not even a good way to compare between models.

 

[Hanson]

Originally Posted by Sam Posten

I want to see the test redone with flash =)

But then iPhones wouldn't be able to participate :-(

 

[Hanson]

Originally Posted by ahilal

I'm curious how jailbreakers factor into this conversation. I would love to know answers to the following  

1) Is it actually harder to "root" an iPhone, given the huge jailbreaker community and free tools for doing so?

2) Are more people jailbreaking iPhones or "rooting" or otherwise modifying Android phones?

 

I know jailbreaking isn't sanctioned and in fact Apple is trying like hell to stamp it out. But it still goes on, and I wonder if that takes any of the steam out of the conversation. In short: if you want carrier choice and the ability to root your iPhone, you have those choices. No?

Not really, because any sort of rooting/hacking is, by its nature, on borrowed time. Any time you want to apply an update or patch a security flaw (even if its the same security flaw that made it possible to jailbreak your phone) you lose what you have, and reinstall all of your jailbroken apps.

 

But here's the funny thing -- you jailbreak an iPhone for functionality that's available in Android out of the box. When you root an Android phone like the Evo, you can actually install a completely different ROM that can change drivers and base functionality. Many custom ROM users report exceptional battery life, higher quality photo and video recording, and greater overall speed and higher benchmarks. Even if you jailbreak an iPhone, you're not able to truly customize it on as granular a level as you can with Android. If you do want to install custom ROM's, however, get an Evo or Nexus One. The latter is a dev phone, and the former is very similar to the HD2, which was hacked with by tons of homebrews devs. So the Evo is the current homebrew dev favorite.

 

Most Android users don't root because, frankly, they don't have to (outside of free tethering, most of the added functionality from rooting is minor for casual users). The ones who jailbreak iPhones are just trying to get what I would consider basic functionality, ie things that most other smartphone OS'es can do out of the box.

 

[DaveF]

 

Originally Posted by Hanson Yoo

What I am saying is that this is essentially a test of screen efficiency...

It's a test of using the device for nothing but web browsing, which by necessity requires the screen on. No one surfs the web with the screen off. If in that situation, the screen uses most of the battery, then web-usage is limited by screen efficiency.

 

The assumption is that using the device for five hours of browsing continuously has the same power consumption as doing the browsing in 10 minute increments over a longer duration, a total of five hours web usage.

 

In either case, the test is how much web browsing one can do total on this device, exclusive of all other activity.

 

A fair question is whether a 1 min refresh is typical of normal web-browsing usage. It seems reasonable, but I don't know.

 

A similar test would be continuous music playback with screen turned off (because that's the typical usage). While most people won't listen to music non-stop for 16 hours (or whatever), it is informative to know how long a device can play music.

 

But most people don't actually do things that require their screens to be on for 5 straight hours. So as a "real world" test, it is total fail. .

I don't know about you, but there are days when I use my iPhone essentially only web surfing on wifi. And over the course of a day I will do 4-6 hours of browsing. So this sort of test is wholly relevant to me. It's also useful because if my phone gets, say, 10 hours of "web" usage, then I know that 2 hours of web use during the day will leave me at 80% battery, which means there's plenty of battery left for phone calls, or music, or etc.

 

Hopefully this clarifies things.

 

[Hanson]

I think you're missing my point Dave. If my browser takes 1-3% of my battery, then it's not significantly affecting the longevity of the phone on a single charge. They might as well have left the phones on at a white screen with auto shut off disabled -- I doubt the numbers would have been that different. So then, how does any of this equate to the "real world"? Here's my "real world" issue -- cell phone standby will take 40% of the battery. In these tests, the phone is turned off. So their results equate to exactly zilcho in the "real world".

 

If they were playing 3D games or something it would be interesting, but this was essentially a test of how long the phones stayed powered with the screen on. And without any control to make sure the field was level between them, it was a pretty poor test at that.

 

[DaveF]

If your standby mode is using 40% of your battery during a day, you're not using your phone at all, or your phone has terrible power management (The iPhone gets a claimed 300 hours of standby -- almost two weeks). And in that case you need a test of "standby" to compare how long phones last between charges without any use.

 

As for the rest, you're clearly not reading what I've written, as I've noted how it's quite "real world" for me.

 

[DaveF]

Three reasons the iPhone could lose significant marketshare in the future:

1) "Cloud" -- keeping it an appendage to the desktop computer too long

2) "AT&T - remaining shackled to a single carrier too long in the US

3) AppStore Review Process -- hosing enough developers without explanation andin apparent conflict with expectations

 

[ManW_TheUncool]

 

Originally Posted by DaveF -- hosing enough developers without explanation andin apparent conflict with expectations

I didn't look closely at the Briefs app, but sounds like that would be the kind of app (some sort of dev toolkit) that Apple would either not want or would want to have a very tight reign over.

 

Don't know any details about the App Review process, but to be honest, 3 months is probably not quite *THAT* long to wait on something like a dev toolkit, if it's any good, considering the App Review group must review tons and tons of app submissions of all kinds (and presumably do some sort of QA/QC, including malware checks and future malware considerations, for them). The problem probably lies more in the fundamental dilemma of having a strict App Review process at all (and owned by the parent corporate entity) and the likely limited resources that can be allocated to that process. That plus the lack of transparency about the process of course.

 

_Man_

 

[DaveF]

From the rant of the friend, the problem is that the reviewers are simply non-responsive. As he says, being rejected is fine; it's the unexplained limbo that's no good.

 

[ManW_TheUncool]

Things can get probably easily lost and go unexplained for a long while for any number of reasons w/ an app that's probably on the higher end of complexity (and potential problems for the Apple outlook/agenda/etc) for the App Review process. We're not talking about a simple game app afterall.

 

Anyone who's worked long in dev in a corporate environment can probably see that happening from time to time -- and in this case, the developer would be *outside* that corporate entity trying to look in, etc, which would probably be comparable to throwing in the kinds of additional barriers that exist for outsourced dev work.

 

_Man_

 

[ManW_TheUncool]

BTW, is there any potential for trademark rights issue w/ that app/toolkit being named Briefs?

 

I have no idea, but the name (along w/ the mention of an underwear logo) reminds me of the old Brief emacs-like editor for DOS created by a company called Underware (or rather, the developer and stakeholders probably decided to name the company in line w/ the program itself).

 

http://en.wikipedia.org/wiki/Brief_%28text_editor%29

 

_Man_