Coverage of Apple's iPhone 4 Press Conference

[jeff33702]

Here's another thing - will holding it for watching video or using apps cause the 3G signal to degrade?

It causes the 3G signal to STOP in certain areas - see sig.

"Q: Why does this happen with just a single finger? It doesn't require a full grip.
A: Bob: Your body is a pretty effective signal absorber. So when you touch it, you attenuate the signal. But when you grip it, you can attenuate it even more."

They are sticking to the LIE that the body is absorbing the signal. The truth is that it only happens when the two edges are bridged together...if you ask me this isn't the huge technical/industry problem that Jobs would like you to think it is. It's simply an error in engineering. If a case fixes it, guess what - Apple f'ed up... They will quietly apply a fix during manufacturing, then proclaim that they are the only cell phone that DOESN"T have this problem. What a joke.

I returned mine on Friday. Yes, I'm still here because I'm bitter... there isn't a smartphone out there with these features and I've grown to love them on my previous iphones.. I just need to be able to make calls...pretty simple really.

 

[PerfSeeker]

Order cancelled. Thanks Apple!

 

[kalsta]

343 * 3600 * 0.00062137 = 767/1
1234800/3600 * 3600/1 * 767/1234800 = 767/1
767/1 * 1234800/1234800 * 3600/3600 = 767/1
767/1 = 767/1

Now, when you group them yourself, you can see that s/s (bolded) is not in fact equal to 1.

Oh for the love of Apple… You have bolded s/h not s/s! Please, take some time to understand what I've done before responding with another argument. Of course there are 3600 seconds to 1 hour. That was never in question. In the next line down I have grouped the same variables, exactly as I did in your iPhone equation. If you wanted to bold s/s you should have done it here:

343 * 3600 * 0.00062137 = 767/1
1234800/3600 * 3600/1 * 767/1234800 = 767/1
767/1 * 1234800/1234800 * 3600/3600 = 767/1
767/1 = 767/1

Again, you are just wrong. It's not arbitrary at all. The measured speed is 343 meters per 1 second. By definition, there are 3600 seconds in one hour.

343/1 is the same as 686/2 which is the same as 1234800/3600. When you make the s in m/s equal to 1, but the s in s/h equal to 3600 (which is what you're trying to do), you're choosing those values arbitrarily and your equation is internally inconsistent. If you still don't get this I really don't know how else to explain it.

You seem to think that I am arguing that if you rearrange the equation to group like-terms you won't get the same answer. Of course you will; I never said you wouldn't.

Here's exactly what you said, word for word:

Your simplification is valid only in terms of units, not for the values. For example, the c in the denominator of a/c is not the same value as the c in the numerator of c/d, so you can't use the same variable in both places. Let me demonstrate with a conversion of the speed of sound from meters per second to mph:

343 m/s * 3600 s/h * 0.00062137 miles/m = 767 mph.

In the first term, the value of seconds is 1 (the denominator); in the second term, the value of seconds is 3600 (the numerator). You aren't able to use the same variable to express both of these values in the equation.

A simple, 'Sorry, I made a mistake' will more than suffice. C'mon, show Apple how it's done!

 

[Master Chief]

It causes the 3G signal to STOP in certain areas - see sig.

"Q: Why does this happen with just a single finger? It doesn't require a full grip.
A: Bob: Your body is a pretty effective signal absorber. So when you touch it, you attenuate the signal. But when you grip it, you can attenuate it even more."

They are sticking to the LIE that the body is absorbing the signal. The truth is that it only happens when the two edges are bridged together...if you ask me this isn't the huge technical/industry problem that Jobs would like you to think it is. It's simply an error in engineering. If a case fixes it, guess what - Apple f'ed up... They will quietly apply a fix during manufacturing, then proclaim that they are the only cell phone that DOESN"T have this problem. What a joke.

I returned mine on Friday. Yes, I'm still here because I'm bitter... there isn't a smartphone out there with these features and I've grown to love them on my previous iphones.. I just need to be able to make calls...pretty simple really.

I'm not sure I should even reply to you, but let's ignore the fact that you are a new forum member.

Now. I can show footage where my iPhone 4 has no issue, whatsoever. Won't even drop a call when I double death grip it, and five minutes later, in a harbor... it is gone instantly. Just by touching it with a single finger. Pretty interesting engineering error. Don't you agree?

I say this is due to bad coverage. AT&T sucks!

 

[BobVB]

Pretty interesting engineering error. Don't you agree?

Well an easily understood one anyway. Some locations the 3G signal is thousands of times the 5 bar level, even with the 20-24 dBm hit touching the antenna causes still leaves more than enough signal strength to display 5 bars. In other locations where the signal is just barely at the 4-5 bar level that same dBm hit is enough to cause the call to drop.

In my office picking up the naked phone will kill the call. 4 miles to the north the signal is so strong you can put the phone in a can and it stills shows 5 bars.

It ain't rocket science.

 

[Eso]

EDIT: it's become clear to me that you either 1) confused the units from my original iPhone formula for values or 2) don't understand the difference between units and values. I get that from your reply here about terms canceling out, and this reply where you simplify the formula by replacing my units with a, b, c, and d, and later call 'a' a value. I demonstrate the difference to great detail with the following.

Oh for the love of Apple… You have bolded s/h not s/s! Please, take some time to understand what I've done before responding with another argument.

If you had included units, maybe it would have made sense.

Let me go through this step by step:

343 m / 1 s * 3600 s / 1 h * 0.00062137 miles / 1 m = 767 miles/hour​

Rearrange the denominators using the commutative property:

343 m / 1 m * 3600 s / 1 s * 0.00062137 miles / 1 h = 767 miles/h​

Since all the operations are multiplication, you can multiply (or divide) them in any order. Move the denominators so values are grouped by units. Cancel units:

343 m / 1 m * 3600 s / 1 s * 0.00062137 miles / 1 h= 767 miles/h​

You'll notice that the term that includes seconds is 3600/1, not 3600/3600. Now combine the unitless terms:

1234800 * 0.00062137 miles / 1 h = 767 miles/h​

Finally, carry out the operation:

767 miles / 1 h = 767 miles/h​

If you recall how you "simplified" my original equation:

That's the same as writing:
a/b * c/c * d/d = a/b

Now c/c = 1, and d/d = 1, so your formula just becomes:
a/b = a/b

When you did this, you improperly assigned variables to the units (i.e. seconds = c). The key difference is that while two different values can have the same units, two different values can not be expressed by the same variable.

At some point, you confused a, b, c, and d for values, not units. Or perhaps you didn't realize that in my original iPhone equation, "# calls", "iPhone with issue", etc. represented units, not values?

When c is a unit, then 1 is also considered a unit in the equation c/c = 1. A unit of 1 is called unitless. That does not indicate that the actual value that carries that particular unit is 1.

That's why I stated this in response to your simplified equation many posts ago:

Your simplification is valid only in terms of units, not for the values.

If you recall my original iPhone equation:

X ("iPhone with issue"/"users who report issue") * Y ("users who report issue"/"# calls") * Z ("# calls"/"total iPhones") = G ("iPhone with issue"/"total iPhones")

...X, Y, and Z are the values. Everything in quotes within the parenthesis are units. To properly simplify this equation by replacing the units with a, b, c, and d is to do like so:

X (a/c) * Y (c/d) * Z (d/b) = G (a/b)​

...where X, Y, Z, and G are the values and a, b, c, d are the units. To properly group the terms with the same units, you first have to divide terms by 1 so you can show the units on the bottom:

X (a) /1 (c) * Y (c) /1 (d) * Z (d)/1 (b) = G (a/b)​

Now you can properly group terms with the same units by moving the denominators and you will see what happens:

X (a) * Y (c) /1 (c) * Z (d)/1 (d) * 1 / 1 (b)= G (a/b)​

We no longer have to explicitly account for the 1 in the denominators, so we can remove them:

X (a/b) * Y (c/c) * Z (d/d) = G (a/b)​

At this point, you made the mistake of confusing 'c' and 'd' as values to state that the middle terms cancel out, leaving you only with a/b = a/b. Instead, the fact that c/c = 1 indicates that the middle terms are unitless. As you can see, the values of these terms are still indicated by the variables Y and Z.

At this point we can drop the units which have canceled and we are left with:

X (a/b) * Y * Z = G (a/b)​

Finally, you can combine terms so you end up with:

X*Y*Z (a/b) = G (a/b)​

Now, this is same conclusion you reached when you stated that a/b = a/b. As you say,

Not very useful for determining the unknown value of 'a'!

'a' is not a value, however. It's the unit of the value we are interested in. The actual unknown value is G, so the final equation does become useful in determining it as long as we know the values for X, Y, and Z.

The heart is that Apple used statistics in a misleading manner by reporting Z as if it were G, when in fact G is the product of XYZ, meaning that their figure for "G" is off by a factor of XY.

 

[AppleFan360]

Well an easily understood one anyway. Some locations the 3G signal is thousands of times the 5 bar level, even with the 20-24 dBm hit touching the antenna causes still leaves more than enough signal strength to display 5 bars. In other locations where the signal is just barely at the 4-5 bar level that same dBm hit is enough to cause the call to drop.

In my office picking up the naked phone will kill the call. 4 miles to the north the signal is so strong you can put the phone in a can and it stills shows 5 bars.

It ain't rocket science.

Same here. At work I can put one finger on the phone and the signal drops. At home I tried several different grips and the bars don't drop at all.

This whole thing has been so overblown I often wonder if a license should be required to use the internet.

 

[kalsta]

it's become clear to me that you either 1) confused the units from my original iPhone formula for values …

True enough. It wasn't obvious to me that you had defined your own unit called 'iPhone with issue / users who report issue' (you should have called it one unit, not two, but more on that below), nor where you hoped to get the associated value from. That's why I broke it down, to show you that 'iPhone with issue' was an unsolvable variable. Nothing wrong with that. All you had to do was explain that you were making up the overall value out of thin air in order to do a simple unit conversion.

or 2) don't understand the difference between units and values. I get that from your reply here about terms canceling out, and this reply where you simplify the formula by replacing my units with a, b, c, and d, and later call 'a' a value.
…
At some point, you confused a, b, c, and d for values, not units. Or perhaps you didn't realize that in my original iPhone equation, "# calls", "iPhone with issue", etc. represented units, not values?

I think it is you who doesn't completely understand units and values. 'Metres per second' is a single unit. It's a derived unit defined by distance in metres divided by time in seconds. If you want to break it down into the units of 'metres' and 'seconds' then you are now concerned with the individual values of those units. Furthermore, you can't talk about a quantity of units without assigning a value. That's just silly.

Imagine I was trying to sell sacks of potatoes at the markets. I'm not interested in selling individual potatoes, only sacks. What I have done is define a unit called 'sack of potatoes'. But you, the customer, come up to me and ask, 'How many potatoes in a sack?' In your mind, the unit of interest is 'potato', not 'sack of potatoes'. Now imagine I told you, 'We can't talk about the number of potatoes. "Potato" is just a unit of "sack of potatoes", not a value.' You would think I was a complete nutter and take your business elsewhere.

If you want to define your own derived unit called 'iPhone with issue / users who report issue', you're entitled to do that. But if you want to talk about the separate units of 'iPhone with issue' and 'users who report issue', each of those units will have an associated value.

Your assertion that this can't be done without having different values for the same variable throughout a solvable equation is just ludicrous, and I proved it to you using the very equation you sought to discredit my argument with. Why can't you just concede that and move on? Why keep spinning your wheels in the mud?

 

[Master Chief]

Well an easily understood one anyway. Some locations the 3G signal is thousands of times the 5 bar level, even with the 20-24 dBm hit touching the antenna causes still leaves more than enough signal strength to display 5 bars. In other locations where the signal is just barely at the 4-5 bar level that same dBm hit is enough to cause the call to drop.

In my office picking up the naked phone will kill the call. 4 miles to the north the signal is so strong you can put the phone in a can and it stills shows 5 bars.

It ain't rocket science.

Exactly. And thus is this new signal indicator – even with the new formula – just as bad as before, because it still doesn't show you (at all times) what it should do; tell you that you are in a bad spot with a weak coverage. Exposing you to higher SAR, and since we still don't know the long-term affects of it...

Let's hope that all operators and cell phone manufacturers will agree to use my new signal indicator – developed for the iPhone – because it will instantly solve this problem; the signal indicator colors: green, orange and red are used to inform the user about both the signal strength (next to the five bars) and SAR. Red simply means that you should use a headset or some other hands free tool.

Note: Each bar has tree different db ranges, based on the signal strength / color. Effectively giving us a total of 15 bars, instead of just 5 bars.

 

[BobVB]

Same here. At work I can put one finger on the phone and the signal drops. At home I tried several different grips and the bars don't drop at all.

This whole thing has been so overblown I often wonder if a license should be required to use the internet.

See I would go the other way, since we know:
That if you are in a good reception are the iP4 has less problems,
And yet
The ip4 is showing 1% more than the previous 3GS dropped call rate,
And let's say 10% of the people are in areas with problems,

They would be dropping calls 10% more than they were before, and with each drop in % affected this goes up even more.

Having even 5% of your customers dropping 20% more calls isn't over blown - that's a problem that needs to be fixed, really fixed.

 

[-aggie-]

See I would go the other way, since we know:
That if you are in a good reception are the iP4 has less problems,
And yet
The ip4 is showing 1% more than the previous 3GS dropped call rate,
And let's say 10% of the people are in areas with problems,

They would be dropping calls 10% more than they were before, and with each drop in % affected this goes up even more.

Having even 5% of your customers dropping 20% more calls isn't over blown - that's a problem that needs to be fixed, really fixed.

You’ve got some messed up math and logic there. If you were in an area where previous phones had a 10% dropped call rate, then the iPhone 4 would have a 11% dropped call rate. Here, the issue would be the network, not the phone.

 

[Master Chief]

See I would go the other way, since we know:

Fine. Let's do this.

That if you are in a good reception are the iP4 has less problems,

I would say; no problems at all.

And yet
The ip4 is showing 1% more than the previous 3GS dropped call rate,

No. Steve only gave us the delta. He said (from the top of my head) less than 1 more per 100 calls. Not 1% more per 100 calls, or whatever.

And let's say 10% of the people are in areas with problems,

It is probably even simpler, by following plain old logic. I mean. If a call wasn't dropped by the proximity bug, then you must be in a weak spot. Like 100% Right? Or am I missing something?

 

[BobVB]

Fine. Let's do this.
I would say; no problems at all.

then that means the increase is all really in those that are having problems, right? It is actually 'Deceit with Statistics 101' to average a population that is having problems with those that aren't, right?

No. Steve only gave us the delta. He said (from the top of my head) less than 1 more per 100 calls. Not 1% more per 100 calls, or whatever.

oh please, he said these were the statistics from AT&T, they were the increase over the 3GS dropped calls. Nothing in that dog and pony show was 'off the top' the presentation were carefully calculated to try and minimize the perceived problem.

If he could have said a less than one in 200 call dropping increase, he would have, since he didn't we know it's at least 0.5% but less than 1% increase in dropped calls. This doesn't even count all the 'straight to voice mail' that those having problems report. So logically, if most are having reduced problems over the 3GS then those who are are having significantly worse, that +1% for everyone is condensed down to them, e,g. if only 10% are having the problem they are dropping an additional 10%, if only 5% then +20% and in the ridiculous nonsense spouted by some of 1% that would mean they were having 100+% dropped calls.

Again, putting a spot that effectively neutralizes the antenna in a place of normal holding makes as much sense as putting the camera lense there.

It is probably even simpler, by following plain old logic. I mean. If a call wasn't dropped by the proximity bug, then you must be in a weak spot. Like 100% Right? Or am I missing something?

only if you are calling anything less than 5 bars a 'weak spot'. I dropped with 5 bars with 4.0, now I drop with 4 bars. It fixed nothing.

You’ve got some messed up math and logic there.

You have that backwards - if an additional call failure rate averages out to 1 more call per 100% and only 10% are really having problems, that means they are dropping an additional 10% of their calls (that would average to 1 per 100% right?). If it only involved 5% than that would involve a 20% increase in their drops, if only 2% a 50% increase, or the silly '1%' claim would mean they were dropping 100% of their calls to have it average out to 1 per the entire population.

Again, taking a smaller group's problems and averaging them in with a larger group without that problem is 'Deceit with Statistics 101' - anyone who does that is deliberately trying to deceive you.

 

[Master Chief]

then that means the increase is all really in those that are having problems, right? It is actually 'Deceit with Statistics 101' to average a population that is having problems with those that aren't, right?

LOL I don't need to answer this, do I?

oh please, he said these were the statistics from AT&T, they were the increase over the 3GS dropped calls. Nothing in that dog and pony show was 'off the top' the presentation were carefully calculated to try and minimize the perceived problem.

Sorry. The "(off the top of my head)" clause was mine, and mine only, simply because I haven't checked the sheets when I replied to you.

If he could have said a less than one in 200 call dropping increase, he would have, since he didn't we know it's at least 0.5% but less than 1% increase in dropped calls. This doesn't even count all the 'straight to voice mail' that those having problems report. So logically, if most are having reduced problems over the 3GS then those who are are having significantly worse, that +1% for everyone is condensed down to them, e,g. if only 10% are having the problem they are dropping an additional 10%, if only 5% then +20% and in the ridiculous nonsense spouted by some of 1% that would mean they were having 100+% dropped calls.

Please. Don't get me wrong. I have made comments here, in this forum, about Apple's use of: "< 1 more per 100 calls". Basically shielding off hard evidence, due to so called "AT&T confidentiality". I couldn't care less about anything.

I mean what if the call drop rate of an 3GS is 10 per 100 calls? That would mean 10% but hey... what do I know without the AT&T data. It can be even higher. Or lower of course.

What I do know is that people are reporting dropped calls, due to the proximity (sensor) bug, which may effectively make things look bad (even more?) for Apple and/or AT&T. Maybe that is why they refuse to show us this data.

Again, putting a spot that effectively neutralizes the antenna in a place of normal holding makes as much sense as putting the camera lense there.

only if you are calling anything less than 5 bars a 'weak spot'. I dropped with 5 bars with 4.0, now I drop with 4 bars. It fixed nothing.

Well. The iOS 4.0.1 update did fix a software flaw – in the signal strength indicator – but this particular update is not the one you are waiting for. That to me is very clear.

But if I was you, in a weak AT&T area, then I would have called AT&T and demand a free MicroCell. This despite the fact that I don't really like the idea, but I do so much more with my iPhone. Don't you?

Wait. What about using a free bumper or a case?

/me ducks

 

[Pommy]

There's a difference between a 1% increase in dropped calls between the 3GS and iPhone 4 and an increase of less than one call per each 100 between the two.

What Steve says doesn't tell us anything about the percent increase in dropped calls without the absolute number of dropped calls per 100 calls for either the 3GS or the iPhone 4.

For example:

Say the 3GS dropped 5 out of every 100 calls. Then according to "< 1 call / 100" the iPhone 4 drops no more than 6 out of every 100 calls. That means looking at the upper bounds (for simplicity), there is at most a 6-5/5 = 1/5 = 20% increase in dropped calls, which is pretty huge.

But if you take an extreme example in the other direction -- say the 3GS dropped 50 out of every 100 calls. Then the iPhone 4 would drop no more than 51 out of every 100 calls, which is at most a 51-50/50 = 1/50 = 2% increase.

In short, the delta in "calls dropped per 100 calls" doesn't really say anything meaningful for how many percent "worse" the iPhone 4 is without knowing the absolute ratios of "calls dropped per 100 calls" for the 3GS.

Which I think is what BobVB is saying <_<


Also Eso you can only do a unit conversion if each of the intermediate terms are identities (equal to "one"), as in your units of time example, so for your equation

X ("iPhone with issue"/"users who report issue") * Y ("users who report issue"/"# calls") * Z ("# calls"/"total iPhones") = G ("iPhone with issue"/"total iPhones")


If you're solving for the ratio of iPhones with issues to total iPhones, you need the ratio of (users who report issue / # calls) to be "1" and the ratio of (# calls / total iPhones) to be "1". Otherwise the equality doesn't hold.

Edit: Sorry, put "1" in quotes because it's confusing here. Basically you need to have "x # of users who report an issue" to be an equivalent value to "y # of calls" (how you would quantify that I don't know... doesn't make sense).

 

[kalsta]

I mean what if the call drop rate of an 3GS is 10 per 100 calls? That would mean 10% but hey... what do I know without the AT&T data. It can be even higher. Or lower of course.

Say the 3GS dropped 5 out of every 100 calls. Then according to "< 1 call / 100" the iPhone 4 drops no more than 6 out of every 100 calls. That means looking at the upper bounds (for simplicity), there is at most a 6-5/5 = 1/5 = 20% increase in dropped calls, which is pretty huge.

But if you take an extreme example in the other direction -- say the 3GS dropped 50 out of every 100 calls. Then the iPhone 4 would drop no more than 51 out of every 100 calls, which is at most a 51-50/50 = 1/50 = 2% increase.

You guys are confusing BobVB's words and talking about something different even though he has explained it quite well. He's not trying to express the average increase in dropped calls for all users as a ratio. He's talking about the potential increase in the rate of dropped calls for someone who might be in a weaker signal area, given that the drop in signal, though relatively significant, doesn't result in a dropped call when signal strength is above a certain threshold. You might not see that as a relevant statistic, but it sure as heck is for the person experiencing it!

 

[kalsta]

if an additional call failure rate averages out to 1 more call per 100% and only 10% are really having problems, that means they are dropping an additional 10% of their calls (that would average to 1 per 100% right?).

Well, that's assuming a few things. Firstly it assumes everyone makes the same number of calls. If the people having problems made less calls, their percentage of dropped calls would be even higher. Secondly, it assumes that the rate of dropped calls for the 3GS was evenly distributed. That's being a little selective with your application of 'Deceit with Statistics 101' isn't it?

 

[Eso]

'Metres per second' is a single unit. It's a derived unit defined by distance in metres divided by time in seconds. If you want to break it down into the units of 'metres' and 'seconds' then you are now concerned with the individual values of those units.

Actually, m/s is a compound unit, but it doesn't matter. Even derived units (like Energy) can be expressed in terms of their base SI units interchangeably.

If you are given 343 m/s, you can always express it as 343 m / 1 s. If you wanted to, it would be just as valid to express it as 1 m / 0.00292 s.

If you want to break it down into the units of 'metres' and 'seconds' then you are now concerned with the individual values of those units.

It sounds like you are saying that instead of using:

X ("iPhone with issue"/"users who report issue") * Y ("users who report issue"/"# calls") * Z ("# calls"/"total iPhones") = G ("iPhone with issue"/"total iPhones")​

...you would use:

A ("iPhone with issue") / B ("users who report issue") * C ("users who report issue") / D( "# calls") * E ("# calls") / F ("total iPhones") = G ("iPhone with issue") / H ("total iPhones")​

That is certainly valid, however it is necessary to realize that X = A/B, Y = C/D and Z=E/F. The two equations are not independent, so it really doesn't matter which one you use. It' not necessary to use the second equation since we are more concerned with the percentage of iPhones that are affected, not the absolute number of affected iPhones.

It's also extremely important to use different variables for terms that have the same units because they do not necessarily have the same value.

I proved it to you using the very equation you sought to discredit my argument with. Why can't you just concede that and move on? Why keep spinning your wheels in the mud?

There's nothing to concede. I already pointed the fundamental error you made when you used a, b, c, and d as units and then started treating them as values. When you did that, you erroneously implied that terms with the same units had to have the same value. That error lead to you believe that c/c = 1 indicated the value of the term was 1, not that it was unitless. Here's what your error looks like in mathematical terms:

A ("iPhone with issue") / C ("users who report issue") * C ("users who report issue") / D( "# calls") * D ("# calls") / B ("total iPhones") = A ("iPhone with issue") / B ("total iPhones")​

If you notice, this is the same structure as your original "simplified" equation with a, b, c, and d. This equation is different from the equation above and it is incorrect.

The same mistake is evident when you claim that, in the speed of sound example, grouping the terms with the units of seconds results in 3600s / 3600s. I demonstrated step-by-step that this is false.

Also Eso you can only do a unit conversion if each of the intermediate terms are identities (equal to "one"), as in your units of time example, so for your equation

X ("iPhone with issue"/"users who report issue") * Y ("users who report issue"/"# calls") * Z ("# calls"/"total iPhones") = G ("iPhone with issue"/"total iPhones")


If you're solving for the ratio of iPhones with issues to total iPhones, you need the ratio of (users who report issue / # calls) to be "1" and the ratio of (# calls / total iPhones) to be "1". Otherwise the equality doesn't hold.

While I used a unit conversion as an example, I never said that my iPhone equation was merely a unit conversion. Actually, I have argued the opposite from the beginning - the figure Apple reported doesn't have the correct units to legitimately estimate the percentage of affected iPhones.

 

[Master Chief]

@kalsta: Like I said in post #1115: "Steve only gave us the delta." and thus we don't know jack. How they got to this "< 1 more per 100 calls" is a mystery to me, and without knowing at least one other parameter – being the number of people experiencing dropped calls – it's all speculation.

 

[kalsta]

Even derived units (like Energy) can be expressed in terms of their base SI units interchangeably.

If you are given 343 m/s, you can always express it as 343 m / 1 s. If you wanted to, it would be just as valid to express it as 1 m / 0.00292 s.

What's that you say? It's possible to express it in terms of it's base units, each unit having a separate value? Mate, that's what I've been telling you all along!!

If you wanted to, it would be just as valid to express it as 1 m / 0.00292 s.

What's that you say? 'If you wanted to'? So in other words, the choice to express it as '343 m / 1 s' is arbitrary? Sound familiar??

FYI 'm/s' is a derived unit. Look it up. It happens to be written as a compound unit since it has no special name, but that's really irrelevant.

 

[kalsta]

@kalsta: Like I said in post #1115: "Steve only gave us the delta." and thus we don't know jack. How they got to this "< 1 more per 100 calls" is a mystery to me, and without knowing at least one other parameter – being the number of people experiencing dropped calls – it's all speculation.

That's true. We know very little because Apple has given us very little. I thought Bob's point was an interesting one though, and it didn't seem like other people were getting it. It is quite plausible (mathematically) that those who are affected are seeing a much larger increase in their dropped call percentage than the <1% Apple quoted. That seems to fit with the anecdotal user data coming in too — when the phone works (in strong signal area, or held without bridging the antenna gap) it works really well. When it fails (in low to medium signal area while bridging the antenna gap) it fails spectacularly.

 

[JaredJL]

Woohoo!!! It's gonna be interesting!