NEWS: Apple sued over hot iPad shutdowns [Wireless Internet]

Prev: NEWS: Security shortcomings in WPA2 that threaten security of wireless networks
Next: NEWS: Security shortcomings in WPA2 that threaten security ofwireless networks

From: ZnU on 8 Aug 2010 14:31

In article <0vnt569ooulbctshsd5rvjn1177ukrcapb(a)4ax.com>,
Jeff Liebermann <jeffl(a)cruzio.com> wrote:

> On Sun, 08 Aug 2010 11:15:44 -0400, ZnU <znu(a)fake.invalid> wrote:
>
> >But this is precisely my point. Having more models doesn't actually mean
> >the phone an individual user buys is less likely to have an issue.
>
> Actually, it does. If all the Android phones were released
> simultaneously, it's conceivable that they may all have some common
> problem overlooked during the development. However, with the current
> staggered release of various models by various vendors, each
> succeeding model builds upon the lessons learned by the competitors.
> This is the way to make incremental improvements, which are certainly
> beneficial to the customer.

Can you point to any actual examples of this?

> >It just means consumers less likely to _hear_ about issues.
>
> Judging by the iPhone 4 sales, consumers may be listening, but they
> don't seem to care and buy the phone anyway. I guess the US buying
> public has been successfully educated in accepting defects. Polling
> a few friends that have iPhone 4's, the consensus seems to be that
> whatever problems arrise, they're minor and Apple will fix it.

IOW, the whole thing was a huge Internet echo chamber phenomenon that
turns out to be pretty pointless.

Which is what I said it probably was to begin with. This was the largest
blowup to data, but it's not like we haven't been through this a
half-dozen times before.

[snip]

> >There are also lots of clueless
> >Apple fans who demand entirely unreasonable things from Apple and get
> >pissed off when they don't get them. (See any Mac web forum on the day of
> >a major Apple announcement.)
>
> Sure, but the fanatics are also the early adopters and first to buy
> anything new. I keep waiting for a iPod Touch with a GPS, camera, and
> built in microphone. Yawn...

Um... the iPod Touch is _way_ beyond the "early adopter" phase.

> >And the iPhone still has substantially better brand recognition than
> >Android.
>
> Yep, but as I previously suggested, the real brand loyalty is to the
> cellular vendor (AT&T or Verizon) as enforced by a 2 year contract.
> If Verizon magically appears with a CDMA iPhone tomorrow, how many
> current AT&T customers, with iPhone are going to pay $325 early
> termination fee to jump ship?

This appears unrelated to what I was talking about.

> There's another factor at work here. Fear of screwing up. Everyone
> "knows" that the iPhone will do everything, because of the 4 year
> track record. If not, there's an app for it somewhere. The GUM
> (great unwashed masses) are not so sure if the various Android phones
> can do the same. FUD (fear uncertainty doubt) at work in the Android
> market.
>
> >Put these together with the tendency for sensationalist Internet
> >'journalism', you have a recipe for a firestorm -- in a way you just
> >don't with Android devices.
>
> I don't see it. There aren't enough Apple haters, clueless Apple
> fans, and brand loyal fanatics to make much of a difference.

Within the Internet tech community there absolutely are. Claiming
otherwise is ludicrous.

> Incidentally, when Apple finally gets the color matching correct, and
> the white iPhone 4 is released, don't forget to count it as a
> seperate product, like the various Android phones.

Don't be absurd. There are _ substantive differences_ between Android
models.

--
"The game of professional investment is intolerably boring and over-exacting to
anyone who is entirely exempt from the gambling instinct; whilst he who has it
must pay to this propensity the appropriate toll." -- John Maynard Keynes

From: John Navas on 8 Aug 2010 14:39

On Sun, 08 Aug 2010 14:31:40 -0400, in
<znu-CC1361.14314008082010(a)Port80.Individual.NET>, ZnU
<znu(a)fake.invalid> wrote:

>In article <0vnt569ooulbctshsd5rvjn1177ukrcapb(a)4ax.com>,
> Jeff Liebermann <jeffl(a)cruzio.com> wrote:

>> Actually, it does. If all the Android phones were released
>> simultaneously, it's conceivable that they may all have some common
>> problem overlooked during the development. However, with the current
>> staggered release of various models by various vendors, each
>> succeeding model builds upon the lessons learned by the competitors.
>> This is the way to make incremental improvements, which are certainly
>> beneficial to the customer.
>
>Can you point to any actual examples of this?

HTC model history.
Motorola model history.

>> >And the iPhone still has substantially better brand recognition than
>> >Android.
>>
>> Yep, but as I previously suggested, the real brand loyalty is to the
>> cellular vendor (AT&T or Verizon) as enforced by a 2 year contract.
>> If Verizon magically appears with a CDMA iPhone tomorrow, how many
>> current AT&T customers, with iPhone are going to pay $325 early
>> termination fee to jump ship?
>
>This appears unrelated to what I was talking about.

Read what he wrote more carefully.

>> I don't see it. There aren't enough Apple haters, clueless Apple
>> fans, and brand loyal fanatics to make much of a difference.
>
>Within the Internet tech community there absolutely are. Claiming
>otherwise is ludicrous.

Really? What support do you have for that claim (if any)?

>> Incidentally, when Apple finally gets the color matching correct, and
>> the white iPhone 4 is released, don't forget to count it as a
>> seperate product, like the various Android phones.
>
>Don't be absurd. There are _ substantive differences_ between Android
>models.

Just as there are between iPhone models. Is each iPhone in a separate
class in your opinion? :)

--
John

"Facts? We ain't got no facts. We don't need no facts. I don't have
to show you any stinking facts!" [with apologies to John Huston]

From: Jeff Liebermann on 8 Aug 2010 14:41

On Sun, 8 Aug 2010 04:44:51 -0700 (PDT), -hh
<recscuba_google(a)huntzinger.com> wrote:

>Jeff Liebermann <je...(a)cruzio.com> wrote:
>> However, the numbers don't work. �The state of the art for GaAS FET
>> front end designs are fairly uniform in the industry. �At worst, the
>> spread in receiver sensitivities might be perhaps 3dB. �The iPhone 4
>> antenna is also about 3 times as large as the typical cell phone
>> antenna, offering at best another 2dB of gain. �That's 5dB at best,
>> out of the 16dB that can be accounted for by front end sensitivity and
>> antenna gain. �The other 11dB is coming from somewhere. �Where?
>
>I'm not sure either. Simplistically, this is why we need to have all
>the numbers, not just some of them.

Again, you have all the numbers I can provide with what I have to work
with. However, additional numbers are not going to do much when their
effect either cancels (such as my example of adding 100dB to both
sides of a ratio), or has minimal effect insufficient to explain the
HUGE deterioration in sensitivity on the iPhone 4.

>For example, perhaps they're not using GaAS, but perhaps GaN. I've
>been experiencing increased leadtimes on GaN availability, so its
>demand is up, despite its current cost.

If it used Gallium Nitride, it would be front page news in the
semiconductor trade press. I see GaN used for various things, but
receiver front ends and the iPhone 4 are not mentioned. Actually, I'm
guessing it's GaAs. It could easily be SiGe or even ordinary Si. I
have no way to tell from the various teardowns and FCC ID web site.

However, it doesn't really matter. The difference between the best
and the worst technologies might be a few dB improvement in
sensitivity. There's still about 11dB of unexplained signal reduction
that has to be accounted for, which isn't happening.

Incidentally, a really hot receiver in a cell phone is often a bad
idea, as it picks up digital rubbish from the CPU.

>Notionally, consider an antenna which is buried within a thick
>insulating shell or coating. The shell protects it from any
>significant amount of random 'death grip' losses, but the reason why
>is partly because it is continuously in a 'death grip' from the
>coating, so its basic reception level is degraded versus an exposed
>antenna.
>
>How is your test method accounting for this situation?

No accounting. As Apple didn't care when it made video clips showing
the effects of hand holding on various Android models. However, may I
point out that there was one phone (vx8000) in my test, that had a
telescoping antenna. In my test, I did a death grip with both hands,
one of the antenna, the other on the phone.
<http://802.11junk.com/jeffl/cellular/cell-test.htm>
Only -6dB difference. I just found some assorted Motorola phones, all
with telescoping antennas. I'll add them later today.

Incidentally, I accidentally did a field strength sensitivity test in
the above table. Note the "two finger" column numbers. The highest
signal strengths were with the phones with the biggest antenna.

>Understood, with the caveat being 'partially' caused. My point is
>that its a bit risky to claim an overall conclusion without at least
>knowing where 'the rest' is coming from, since we may be completely
>overlooking an entire phenomenology...perhaps as much as 11dB :-)

There's very little that you can add to the puzzle that will account
for about a 10 times difference in either testing method or
uncontrolled factors. You're welcome to try.

Incidentally, I'm being rather conservative with my guesswork and
claims. The only thing I'm sure of is that the iPhone 4 is unusually
sensitive to having the antenna touched, far more than even the worst
phone I tested, and sufficiently to make me wonder if there's a
receiver front end design.

>Obviously, when all of the conditional requirements stack, it will
>manifest itself as a problem...but for the (notionally) vast majority
>of the time, the conditional requirements don't stack, so the consumer
>gets the benefit with zero apparent downside.

Ok. You've successfully identified two possible conditions, the
receiver sensitivity and the antenna size. Got any more than will
explain an 11dB difference? Forgive my ignorance, but I'm at a loss
of additional factors.

>> >Agreed on the math ... but:
>>
>> >> [...]�For example, if I add 100dB to each number:
>>
>> >...but: � my point is that this "100dB" isn't necessarily the exact
>> >same value for the two transceivers, and that's a factor that can
>> >swamp the 16dB differences in attenuation losses into irrelevance.
>>
>> Wrong. �It cannot "swamp", negate, or compensate for anything.
>
>Yet you just agreed that there could be a difference in base
>sensitivity present.

I agreed on a possible difference. If I can get my hands on an iPhone
4 long enough to run a bench test, I can attach some numbers to my
guesswork.

My comments that you cannot "swamp" (your term) or otherwise minimize
the effects was amply demonstrated by what happens when I added 100dB
of who know what to both sides of the equation and ended up with
exactly the same difference between the iPhone and others. Also, as
long as the sensitivities are controlled by the antenna and the front
end, throwing unrelated factors into the equation isn't going to
change much, certainly not 11dB.

>My point is that the assessment at the system's net performance level
>requires insight both into potential attenuation losses as well as its
>basic sensitivity.

Nope. I could attach an audio analyzer to the speaker, re-run the
hand test using audio SNR instead of signal strength, and produce
exactly the same results (as long as there's no additional audio
distortion added). The noise figure of the IF, demodulator, and audio
processor do NOT have any effect on the receiver sensitivity or field
strength sensitivity. You can demonstrate this with any cascade RF NF
calculator. Here's a simple example:
<http://www.minicircuits.com/pages/mcl_nf_calc.html>
As long as the gain of the first stage is greater than about 10dB, the
total Noise Figure at the output will be primarily controlled by the
NF of the first stage. You can mess with the NF of the 2nd stage
quite a bit before the Total NF will change very much. Add a big
noisy IF circuit after the 2nd stage and it will hardly change at
all.

>> Agreed, but it's important. �If AT&t extends the dropout time before
>> the signal gets lost, the effects of the "hand on antenna" problem is
>> effectively hidden and conceivably no longer a real problem. �What
>> would happen is that users would be constantly asking for repeats, and
>> dropouts would be dramatically reduced. �This might also explain why
>> some users claim they don't get any dropouts as a result of the "death
>> grip". �It's easy enough to measure. �I'll give it a try and see what
>> happens.

Oops. I'm on vacation and forgot to bring home some AT&T phones.
Maybe I'll borrow one or run the test next week. The test is simple
enough. I call my home phone from an AT&T cell phone. At some point,
I put the phone inside my high tech shield room (microwave oven),
close the door, and time how long it takes to drop the call. You
might want to do the same to see what happens. Umm... don't turn on
the microwave oven.

>Thanks. It would be nice to be able to ignore this from the holistic
>system design perspective, but it is a real-world factor ... and it
>would be quite interesting IMO if we were to find that AT&T were using
>different values based upon specific makes/models of phones (ie, has
>clearly taken extra steps to accommodate the iPhone4): then, you're
>really be onto something significant.

It takes a long time for AT&T to do something technical and obvious.
For example, the AT&T 3G MicroCell still likes to do a handoff from
the MicroCell to a weak tower signal, and drop the call.
<http://forums.wireless.att.com/t5/AT-T-3G-MicroCell/Microcell-drops-handoff/m-p/1960251>
I'm sure AT&T will figure out the obvious (i.e. disable handoff)
eventually. At least the iPhone 4 seems critical and worthy of an
expedited response. Maybe next year.

--
Jeff Liebermann jeffl(a)cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

From: John Navas on 8 Aug 2010 14:41

On Sun, 08 Aug 2010 11:39:17 -0700, in
<v9ut569kkb5soa46241rg2t2jrrehdjcs7(a)4ax.com>, John Navas
<spamfilter1(a)navasgroup.com> wrote:

>On Sun, 08 Aug 2010 14:31:40 -0400, in
><znu-CC1361.14314008082010(a)Port80.Individual.NET>, ZnU
><znu(a)fake.invalid> wrote:
>
>>In article <0vnt569ooulbctshsd5rvjn1177ukrcapb(a)4ax.com>,
>> Jeff Liebermann <jeffl(a)cruzio.com> wrote:
>
>>> Actually, it does. If all the Android phones were released
>>> simultaneously, it's conceivable that they may all have some common
>>> problem overlooked during the development. However, with the current
>>> staggered release of various models by various vendors, each
>>> succeeding model builds upon the lessons learned by the competitors.
>>> This is the way to make incremental improvements, which are certainly
>>> beneficial to the customer.
>>
>>Can you point to any actual examples of this?
>
>HTC model history.
>Motorola model history.
>
>>> >And the iPhone still has substantially better brand recognition than
>>> >Android.
>>>
>>> Yep, but as I previously suggested, the real brand loyalty is to the
>>> cellular vendor (AT&T or Verizon) as enforced by a 2 year contract.
>>> If Verizon magically appears with a CDMA iPhone tomorrow, how many
>>> current AT&T customers, with iPhone are going to pay $325 early
>>> termination fee to jump ship?
>>
>>This appears unrelated to what I was talking about.
>
>Read what he wrote more carefully.
>
>>> I don't see it. There aren't enough Apple haters, clueless Apple
>>> fans, and brand loyal fanatics to make much of a difference.
>>
>>Within the Internet tech community there absolutely are. Claiming
>>otherwise is ludicrous.
>
>Really? What support do you have for that claim (if any)?
>
>>> Incidentally, when Apple finally gets the color matching correct, and
>>> the white iPhone 4 is released, don't forget to count it as a
>>> seperate product, like the various Android phones.
>>
>>Don't be absurd. There are _ substantive differences_ between Android
>>models.
>
>Just as there are between iPhone models. Is each iPhone in a separate
>class in your opinion? :)

p.s. Still waiting for your list of specific Android models with
significant market share that don't measure up to the class of the
iPhone (any version). Or can't you come up with any?

--
John

"Assumption is the mother of all screw ups."
[Wethern�s Law of Suspended Judgement]

From: nospam on 8 Aug 2010 14:44

In article <znu-CC1361.14314008082010(a)Port80.Individual.NET>, ZnU
<znu(a)fake.invalid> wrote:

> > Sure, but the fanatics are also the early adopters and first to buy
> > anything new. I keep waiting for a iPod Touch with a GPS, camera, and
> > built in microphone. Yawn...
>
> Um... the iPod Touch is _way_ beyond the "early adopter" phase.

yes it is, but it would be nice if the ipod touch had a few more
features found in the iphone. a lot of people want an ios device but
not the required contract, regardless of carrier.

based on current rumours, the next ipod touch will be getting a front
facing camera and probably a microphone too since facetime would be
stupid without voice. it might also get a rear facing camera and it's
very likely it will get a retina display. it's highly likely it will
get a gyroscope given that it's marketed as a 'fun' ipod suitable for
gaming. the gps is less certain since the gps is part of the cellular
chipset, which is why only the 3g ipad has gps, not the wifi ipad.

traditionally ipods are announced in september, but there is some
indication that it might be later this month.