or Connect
AppleInsider › Forums › Mobile › iPhone › Apple's iPhone 5s sensor woes may be linked to new accelerometer supplier
New Posts  All Forums:Forum Nav:

Apple's iPhone 5s sensor woes may be linked to new accelerometer supplier

post #1 of 41
Thread Starter 
Owners of Apple's latest flagship iPhone 5s have been experiencing issues seemingly related to the handset's on-board sensors, and a new report claims the use of a new accelerometer supplier may be to blame.

Accelerometer


Following up on a report from early October, Gizmodo reported on Wednesday that a switch in Apple's accelerometer suppliers may likely be the cause of misreadings seen in the iOS 7 Compass app.

According to a teardown by Chipworks, the iPhone 5s now uses an accelerometer made by Bosch Sensortech, while older models of the smartphone employ STMicroelectronics silicon. Interestingly, STM still makes the three-axis gyroscope in the 5s.

Because component specifications differ from manufacturer to manufacturer, the new part does not behave in the same manner as those found in previous iPhones, meaning apps accessing its raw data will output incorrect readings. In our own anecdotal testing, we found the Compass app's level to be off by some three to five percent against a physical tubular spirit measure, even after recalibrating multiple times.

A more precise assessment of the apparent discrepancy was shared by RealityCap, a company specializing in sensor-based real-time 3D location software for iOS. As CEO Eagle Jones explained in a blog post, an accelerometer's accuracy relies on variance, or the consistency in readings, and bias, or constant inaccuracy due to manufacturing flaws.

This is where we find the problem: the typical bias for the ST part is +/- 20mg, while the Bosch part lists +/-95mg. This almost 5x greater offset range is confirmed by our measurements, and is absolutely consistent with the failures being reported by users and the media. Specifically, a +/- 20mg offset range would translate to around a +/-1 degree accuracy range in tilt detection, and a +/-95mg offset translates to +/-5 degrees in tilt.


According to Jones, Apple would be able to remedy the issue by building in a recalibrated firmware bias at the factory. Alternatively, manual calibration can theoretically be accomplished on an app-by-app basis, a solution RealityCap is currently working on.

For its part, Apple remains mum on the matter.
post #2 of 41
So is this something that they just need to recalibrate the software parameters (ie software update) or does it require a swap out for new hardware?
post #3 of 41

Accelerometergate?

"Apple should pull the plug on the iPhone."

John C. Dvorak, 2007
Reply

"Apple should pull the plug on the iPhone."

John C. Dvorak, 2007
Reply
post #4 of 41
Quote:
Originally Posted by pkabir View Post

So is this something that they just need to recalibrate the software parameters (ie software update) or does it require a swap out for new hardware?

That would work if they're all off by the same margin, if not it needs to be user adjusted.
"Few things are harder to put up with than the annoyance of a good example" Mark Twain
"Just because something is deemed the law doesn't make it just" - SolipsismX
Reply
"Few things are harder to put up with than the annoyance of a good example" Mark Twain
"Just because something is deemed the law doesn't make it just" - SolipsismX
Reply
post #5 of 41

I have a 5S but I have no way to know if the level is accurate or not - there's not a single thing in my home that's level or plumb. No walls, no floors, no doors... everything is crooked or tilted. The joys of living in a 100+ year old house hehehe. :D 

 

Anyhow, if I do want to put up a shelf or hang a picture, I think I'll use a $5 hardware store level, and maybe not precariously balance my $700 phone while doing DIY home projects.

post #6 of 41

As an Apple fanboy, I must say, if true, this is looking like a botch release, from Apple's normal standards. Not entirely Apple's doing, but indirectly still tarnishes the brand. 

 

Definitely get the AppleCare this go around! 

post #7 of 41
Quote:
Originally Posted by Atashi View Post

I have a 5S but I have no way to know if the level is accurate or not - there's not a single thing in my home that's level or plumb. No walls, no floors, no doors... everything is crooked or tilted. The joys of living in a 100+ year old house hehehe. 1biggrin.gif  

Anyhow, if I do want to put up a shelf or hang a picture, I think I'll use a $5 hardware store level, and maybe not precariously balance my $700 phone while doing DIY home projects.
Well if nothing is level in your house the 5s is perfect for you, the offset error might work in your favor.
post #8 of 41
Quote:
Originally Posted by AppleInsider View Post
 

This is where we find the problem: the typical bias for the ST part is +/- 20mg, while the Bosch part lists +/-95mg.

I wonder how much cheaper the Bosch part is? :rolleyes:

Shut up and go away, you useless, pathetic FUDmonger - Tallest Skil
Reply
Shut up and go away, you useless, pathetic FUDmonger - Tallest Skil
Reply
post #9 of 41
Quote:
Originally Posted by Richard Getz View Post
 

As an Apple fanboy, I must say, if true, this is looking like a botch release, from Apple's normal standards. Not entirely Apple's doing, but indirectly still tarnishes the brand. 

 

Definitely get the AppleCare this go around! 

 

To be fair, there are parts of The Matrix where the gravitational field is not entirely stable.

"Apple should pull the plug on the iPhone."

John C. Dvorak, 2007
Reply

"Apple should pull the plug on the iPhone."

John C. Dvorak, 2007
Reply
post #10 of 41
Quote:
Originally Posted by hentaiboy View Post

I wonder how much cheaper the Bosch part is? 1rolleyes.gif

May not be cheaper at all. Apple is trying to have more than one source for some components to insure shipments or needs multiple sources for capacity... Both are smart moves.
"That (the) world is moving so quickly that iOS is already amongst the older mobile operating systems in active development today." — The Verge
Reply
"That (the) world is moving so quickly that iOS is already amongst the older mobile operating systems in active development today." — The Verge
Reply
post #11 of 41
Quote:
Originally Posted by hentaiboy View Post

I wonder how much cheaper the Bosch part is? 1rolleyes.gif
5%, same as error difference :-D
post #12 of 41
i'm heavily involved in GPS and accelerometer development. The new sensor doesn't explain the compass readings being off by 11-13 degrees -- which is alarming for me and anyone who relies on compass readings. in our testing, the 5S is always off by 11-13 degrees, while the 5 and the 4S are consistent. the magnetometer is definitely off, much more so than the accelerometer. i suspect the problem might be in the translation between the 64 bit architecture and 32 bit software, or perhaps the software translating the sensor equipment readings is defective. i truly hope it's not a hardware problem because that would imply a fix that won't reach millions of our customers.
post #13 of 41
Quote:
Originally Posted by Richard Getz View Post

 

Definitely get the AppleCare this go around! 

 

Getting the AppleCare Protection Plan just to fix this problem is maybe a bit over the top, since Apple already offers a 1 year limited warranty by default.

post #14 of 41
Quote:
Originally Posted by peter zulu View Post

i'm heavily involved in GPS and accelerometer development. The new sensor doesn't explain the compass readings being off by 11-13 degrees -- which is alarming for me and anyone who relies on compass readings. in our testing, the 5S is always off by 11-13 degrees, while the 5 and the 4S are consistent. the magnetometer is definitely off, much more so than the accelerometer. i suspect the problem might be in the translation between the 64 bit architecture and 32 bit software, or perhaps the software translating the sensor equipment readings is defective. i truly hope it's not a hardware problem because that would imply a fix that won't reach millions of our customers.

and or the M7?

 

I'm finding the software for using the compass and accelerometer  on the 4s under 7.0.2 ( compass and inclinometer ) seems a bit flaky.  compass seems to 'lockup' at crazy declinations for minutes and then oscillating then locking up.   my iphone 5 under iOS 6 was not nearly as crazy.  

 

now under 7 both phones seems to be crazy sensitive accelerometers compared to the prior iOS release.   So I doubt it's HW, and more a SW issue.

post #15 of 41
Quote:
Originally Posted by Macky the Macky View Post


May not be cheaper at all. Apple is trying to have more than one source for some components to insure shipments or needs multiple sources for capacity... Both are smart moves.

Doubt that.  the 5s isn't dual sourcing it's accelerometers.

 

It's also been discussed that the power requirements, and the net size of the Bosch chip are smaller, with size being the likely differentiator.   I strongly doubt this is dual sourcing.

post #16 of 41
Quote:
Originally Posted by Atashi View Post
 

I have a 5S but I have no way to know if the level is accurate or not - there's not a single thing in my home that's level or plumb. No walls, no floors, no doors... everything is crooked or tilted. The joys of living in a 100+ year old house hehehe. :D 

 

Anyhow, if I do want to put up a shelf or hang a picture, I think I'll use a $5 hardware store level, and maybe not precariously balance my $700 phone while doing DIY home projects.

get inclinometer... it has a calibration mode that takes 'flat but unlevel' under account.

 

but for the most part, the $5 level is a better tool for the job.  (I do like the iPhone for leveling furniture, however).

 

-ToG (in a 120yo house.... It's amazing how much .2deg slope makes a difference for 8'x6' bookshelves;-)... or worse... when new work is level against floors and ceilings that are not;)

post #17 of 41
Quote:
Originally Posted by TheOtherGeoff View Post
 

and or the M7?

 

I'm finding the software for using the compass and accelerometer  on the 4s under 7.0.2 ( compass and inclinometer ) seems a bit flaky.  compass seems to 'lockup' at crazy declinations for minutes and then oscillating then locking up.   my iphone 5 under iOS 6 was not nearly as crazy.  

 

now under 7 both phones seems to be crazy sensitive accelerometers compared to the prior iOS release.   So I doubt it's HW, and more a SW issue.

The Maps app does that on my iPhone 5 sometimes.  When I'm in an area with little or no data coverage and weak GPS the maps app will oscillate back and forth especially if I come to a stop.  It will even do it if I'm driving down the road.  I'm starting to think this may be a combination of iOS7 and hardware issues.

Just say no to MacMall.  They don't honor their promotions and won't respond to customer inquiries.  There are better retailers out there.
Reply
Just say no to MacMall.  They don't honor their promotions and won't respond to customer inquiries.  There are better retailers out there.
Reply
post #18 of 41

Just another example of people focusing on things that do not mater. For those who are relaying on an iphone to make your house level or keep you from getting lost in the woods, you deserve to live in a crooked house and be lost in the woods.

 

There is a saying which is true in this case, use the right tool for the right job. Remember the fact the phone can do these things was never its intended purpose, therefore, The fact the phone can do these thing makes it good enough. Otherwise, if you want precision then you need to buy the right tool. 

post #19 of 41
Quote:
Originally Posted by razorpit View Post
 

The Maps app does that on my iPhone 5 sometimes.  When I'm in an area with little or no data coverage and weak GPS the maps app will oscillate back and forth especially if I come to a stop.  It will even do it if I'm driving down the road.  I'm starting to think this may be a combination of iOS7 and hardware issues.

No, that just typical behavior for an A-GPS device (such as phones) when you have no data connection and weak GPS signal. In such a situation the error margin for calculating the location grows much bigger and as such it will appear as if you might be in multiple places due to the inaccuracy. I've seen such effects happen on Android phones with Google Maps as well and on iOS 6 and earlier in the same circumstance.

post #20 of 41
Quote:
Originally Posted by Suddenly Newton View Post
 

Accelerometergate?

 

No...Sensorgate (it's inclusive)

post #21 of 41
Quote:
Originally Posted by razorpit View Post
 

The Maps app does that on my iPhone 5 sometimes.  When I'm in an area with little or no data coverage and weak GPS the maps app will oscillate back and forth especially if I come to a stop.  It will even do it if I'm driving down the road.  I'm starting to think this may be a combination of iOS7 and hardware issues.

Actually the phone is using 1 or 3 method to figure out where the phone is, first is simple cell tower locations, if are connected to the tower and that has a known location the phone can approximate your locations, next it will use a WiFi signal since those may have a location based on a database that apple and other companies have collected, Lastly is using the GPS which is going to triangulate your exact location with in a few feet. Apple most likely has an algorithm which used all three of these methods depending on what is going on at any particular time.

 

To prove this to yourself, turn off wifi, then turn off your phone for a period of time, then go in a structure like a house (shielded form a GPS signal) turn the phone back on, and bring up maps, and your location will be based on the cell tower location. Then turn on Wifi and your location will get a little better, then walk outside and then your location will become even better.

post #22 of 41
Quote:
Originally Posted by MikeJones View Post
 

No, that just typical behavior for an A-GPS device (such as phones) when you have no data connection and weak GPS signal. In such a situation the error margin for calculating the location grows much bigger and as such it will appear as if you might be in multiple places due to the inaccuracy. I've seen such effects happen on Android phones with Google Maps as well and on iOS 6 and earlier in the same circumstance.

 

 

Quote:
Originally Posted by Maestro64 View Post
 

Actually the phone is using 1 or 3 method to figure out where the phone is, first is simple cell tower locations, if are connected to the tower and that has a known location the phone can approximate your locations, next it will use a WiFi signal since those may have a location based on a database that apple and other companies have collected, Lastly is using the GPS which is going to triangulate your exact location with in a few feet. Apple most likely has an algorithm which used all three of these methods depending on what is going on at any particular time.

 

To prove this to yourself, turn off wifi, then turn off your phone for a period of time, then go in a structure like a house (shielded form a GPS signal) turn the phone back on, and bring up maps, and your location will be based on the cell tower location. Then turn on Wifi and your location will get a little better, then walk outside and then your location will become even better.

 

I understand what you're both saying but I've never seen it move like this before.  Its not that I'm "bouncing" from place to place, it's the direction the phone thinks I'm facing is incorrect.  It definitely appears this is a compass issue and not a "positional" issue.  I probably didn't explain it the best but even as I travel the pointer shows me to be in the correct position on the map, but the map is twitching back and forth as if it can't tell the direction I am facing.  In other words it looks as if the phone thinks I'm "side stepping" as I travel.  Hard to explain but having used GPS since 2000 I've never seen something appear quite like that.

 

I'm not too worried, it doesn't happen often, I figure it'll be fixed in a later release.  I'm more concerned about an OH address appearing in MN in the maps app  I reported months ago...

Just say no to MacMall.  They don't honor their promotions and won't respond to customer inquiries.  There are better retailers out there.
Reply
Just say no to MacMall.  They don't honor their promotions and won't respond to customer inquiries.  There are better retailers out there.
Reply
post #23 of 41
Originally Posted by Richard Getz View Post
As an Apple fanboy

 

 

Come on, man; you know better.

 
…if true, this is looking like a botch release, from Apple’s normal standards.

 

Not really. There’s always something to whine about.

Originally Posted by Marvin

The only thing more insecure than Android’s OS is its userbase.
Reply

Originally Posted by Marvin

The only thing more insecure than Android’s OS is its userbase.
Reply
post #24 of 41
Quote:
Originally Posted by razorpit View Post

 

I understand what you're both saying but I've never seen it move like this before.  Its not that I'm "bouncing" from place to place, it's the direction the phone thinks I'm facing is incorrect. It definitely appears this is a compass issue and not a "positional" issue.

No, I don't think you do. Both the bouncing and wrong heading are a side effect of the increased margin of error for both location and heading calculation in the situation you've described. I've seen it happen in Google Maps on Android and on iOS 6 and earlier with apps like Waze. You don't seem to understand how much error is introduced when using an A-GPS in such situations. A-GPS is only reasonably accurate with good satellite fixes and a data connection to triangulate. I was seeing exactly what you describe just a few days ago on my iPhone 5 in an area with poor data connection and it was very overcast that day which means that it was having a hard time getting good satellite fixes.

post #25 of 41
Quote:
Originally Posted by MikeJones View Post
 
Quote:
Originally Posted by razorpit View Post

 

I understand what you're both saying but I've never seen it move like this before.  Its not that I'm "bouncing" from place to place, it's the direction the phone thinks I'm facing is incorrect. It definitely appears this is a compass issue and not a "positional" issue.

No, I don't think you do. Both the bouncing and wrong heading are a side effect of the increased margin of error for both location and heading calculation in the situation you've described. I've seen it happen in Google Maps on Android and on iOS 6 and earlier with apps like Waze. You don't seem to understand how much error is introduced when using an A-GPS in such situations. A-GPS is only reasonably accurate with good satellite fixes and a data connection to triangulate. I was seeing exactly what you describe just a few days ago on my iPhone 5 in an area with poor data connection and it was very overcast that day which means that it was having a hard time getting good satellite fixes.

 

Actually you may be misunderstanding assisted GPS, which simply refers to the use of other positional data (wifi, cell towers etc.) to speed up the initial GPS location fix.  Once that position is roughly acquired then those additional data sources are disregarded, primarily because they are very inaccurate by comparison to the actual GPS data.

 

In determining direction, when stationary or at low speed, I'm not sure how the navigational apps weight the derivative GPS data versus the compass data, or even whether they all use the same weighting.

post #26 of 41
Quote:
Originally Posted by MikeJones View Post
 

No, I don't think you do. Both the bouncing and wrong heading are a side effect of the increased margin of error for both location and heading calculation in the situation you've described. I've seen it happen in Google Maps on Android and on iOS 6 and earlier with apps like Waze. You don't seem to understand how much error is introduced when using an A-GPS in such situations. A-GPS is only reasonably accurate with good satellite fixes and a data connection to triangulate. I was seeing exactly what you describe just a few days ago on my iPhone 5 in an area with poor data connection and it was very overcast that day which means that it was having a hard time getting good satellite fixes.

 

I spend a good amount of time in areas with little or no cellular connection and this is the first time I've ever seen these specific events happen under these specific conditions.  I've seen them happen with a cold startup on a GPS but never after acquiring a lock and then losing it temporarily.  Apparently I haven't been paying attention before upgrading to iOS7.

Just say no to MacMall.  They don't honor their promotions and won't respond to customer inquiries.  There are better retailers out there.
Reply
Just say no to MacMall.  They don't honor their promotions and won't respond to customer inquiries.  There are better retailers out there.
Reply
post #27 of 41
Quote:
Originally Posted by razorpit View Post
 
Quote:
Originally Posted by MikeJones View Post
 

No, I don't think you do. Both the bouncing and wrong heading are a side effect of the increased margin of error for both location and heading calculation in the situation you've described. I've seen it happen in Google Maps on Android and on iOS 6 and earlier with apps like Waze. You don't seem to understand how much error is introduced when using an A-GPS in such situations. A-GPS is only reasonably accurate with good satellite fixes and a data connection to triangulate. I was seeing exactly what you describe just a few days ago on my iPhone 5 in an area with poor data connection and it was very overcast that day which means that it was having a hard time getting good satellite fixes.

 

I spend a good amount of time in areas with little or no cellular connection and this is the first time I've ever seen these specific events happen under these specific conditions.  I've seen them happen with a cold startup on a GPS but never after acquiring a lock and then losing it temporarily.  Apparently I haven't been paying attention before upgrading to iOS7.

 

Cellular reception has no effect except on cold start times.  Also cloud cover (or most other tropospheric conditions) has no significant effect on GPS signal strength.

post #28 of 41
Quote:
Originally Posted by razorpit View Post
 

 

 

 

I understand what you're both saying but I've never seen it move like this before.  Its not that I'm "bouncing" from place to place, it's the direction the phone thinks I'm facing is incorrect.  It definitely appears this is a compass issue and not a "positional" issue.  I probably didn't explain it the best but even as I travel the pointer shows me to be in the correct position on the map, but the map is twitching back and forth as if it can't tell the direction I am facing.  In other words it looks as if the phone thinks I'm "side stepping" as I travel.  Hard to explain but having used GPS since 2000 I've never seen something appear quite like that.

 

I'm not too worried, it doesn't happen often, I figure it'll be fixed in a later release.  I'm more concerned about an OH address appearing in MN in the maps app  I reported months ago...

To your point, I too have been using GPS systems for a long time, my firsts dates back to the 90's. What you describe I have seen on a stand alone GPS units as well as on an Andriod phone of various models, it does not happen all the time but for some reason the system gets bad or corrupt data and tries to triangulate on that data. Depending on the system they respond to this in different ways. I have seen the location jump from one point to another or spin around unclear what direction I am heading. The newer systems seem to do better at knowing direction of travel, but you still need to be moving to know exact what direction you are heading even with a compass feature.  I believe the newer systems just do a better job of filtering unlikely data our and estimating where you are since it does not requirement it to be exactly right it just good enough. I do not think this is a specific Apple like of issue but if the compass has more error to it then in the past their algorithm may not be good enough and shows more errors.

post #29 of 41
Quote:
Originally Posted by nikmanshah View Post

Quote:
Originally Posted by Richard Getz View Post

Definitely get the AppleCare this go around! 

Getting the AppleCare Protection Plan just to fix this problem is maybe a bit over the top, since Apple already offers a 1 year limited warranty by default.

Indeed, standard warranty is enough. I got a new phone (5) last month only because I told them the power button didn't always work. They didn't even try it out; replaced with a brand new (64GB) one, no questions asked. Well, still in warranty.

While waiting for the Genius I had 3 people with a broken 5 at my table; all got a new phone. Two power button issues, one with a broken screen. Don't know if the broken screen was a) the only problem and b) if she had AppleCare, still, all replaced.
I’d rather have a better product than a better price.
Reply
I’d rather have a better product than a better price.
Reply
post #30 of 41
Quote:
Originally Posted by PhilBoogie View Post


Indeed, standard warranty is enough. I got a new phone (5) last month only because I told them the power button didn't always work. They didn't even try it out; replaced with a brand new (64GB) one, no questions asked. Well, still in warranty.

While waiting for the Genius I had 3 people with a broken 5 at my table; all got a new phone. Two power button issues, one with a broken screen. Don't know if the broken screen was a) the only problem and b) if she had AppleCare, still, all replaced.

 

Quote:
Originally Posted by nikmanshah View Post
 

 

Getting the AppleCare Protection Plan just to fix this problem is maybe a bit over the top, since Apple already offers a 1 year limited warranty by default.

 

Probably, but why take the chance? 

 
Quote:
Originally Posted by Tallest Skil View Post
 

Come on, man; you know better.

 

Not really. There’s always something to whine about.

 

 

I know, and yes, I'm whining a bit, partially because my 5s is still "preparing to ship" :grumble: . However, this does seem like a more troubled rollout than usual. 

post #31 of 41
Originally Posted by Richard Getz View Post

I know, and yes, I'm whining a bit, partially because my 5s is still "preparing to ship" :grumble: . However, this does seem like a more troubled rollout than usual. 

 

We really need to compile a chart of every product release and associated info, you know? See how everything compares…

Originally Posted by Marvin

The only thing more insecure than Android’s OS is its userbase.
Reply

Originally Posted by Marvin

The only thing more insecure than Android’s OS is its userbase.
Reply
post #32 of 41
Quote:
Originally Posted by Maestro64 View Post
 

To your point, I too have been using GPS systems for a long time, my firsts dates back to the 90's. What you describe I have seen on a stand alone GPS units as well as on an Andriod phone of various models, it does not happen all the time but for some reason the system gets bad or corrupt data and tries to triangulate on that data. Depending on the system they respond to this in different ways. I have seen the location jump from one point to another or spin around unclear what direction I am heading. The newer systems seem to do better at knowing direction of travel, but you still need to be moving to know exact what direction you are heading even with a compass feature.  I believe the newer systems just do a better job of filtering unlikely data our and estimating where you are since it does not requirement it to be exactly right it just good enough. I do not think this is a specific Apple like of issue but if the compass has more error to it then in the past their algorithm may not be good enough and shows more errors.

I think you may be correct on this.  I think there might have been some changes made to the the algorithms in the compass and maps apps in iOS 7 of which Apple needs to address.  Not criticizing them, they did a lot over the last year and bugs are to be expected, but I think some things can be improved slightly.

 

What's a little odd is that I also use Navigon when I get in to really remote locations and Maps is unable to grab data and I've yet to experience the same issues.

Just say no to MacMall.  They don't honor their promotions and won't respond to customer inquiries.  There are better retailers out there.
Reply
Just say no to MacMall.  They don't honor their promotions and won't respond to customer inquiries.  There are better retailers out there.
Reply
post #33 of 41
Quote:
Originally Posted by Tallest Skil View Post

We really need to compile a chart of every product release and associated info, you know? See how everything compares…

But who to report on Samsung, without kdarling around?

Sorry to hear about your poor delivery time Rich. Apple is releasing a gazillion countries next Friday, so it would be fair for you to already have received yours by then. Hang in there...
I’d rather have a better product than a better price.
Reply
I’d rather have a better product than a better price.
Reply
post #34 of 41
Quote:
Originally Posted by Tallest Skil View Post


Come on, man; you know better.

Not really. There’s always something to whine about.
I
Code:

Maybe Apple will blame it on the user holding the phone incorrectly. They just need to tilt everything a few degrees.
post #35 of 41
Quote:
Originally Posted by dasanman69 View Post


That would work if they're all off by the same margin, if not it needs to be user adjusted.

That's not true. I understand you're trying to apply common sense. But consider that, if you have never really used and calibrated accelerometers before, it might be best not to inject an unqualified opinion. After all, you don't want to mislead people, do you?

post #36 of 41
Quote:
Originally Posted by muppetry View Post
 

 

Actually you may be misunderstanding assisted GPS, which simply refers to the use of other positional data (wifi, cell towers etc.) to speed up the initial GPS location fix.  Once that position is roughly acquired then those additional data sources are disregarded, primarily because they are very inaccurate by comparison to the actual GPS data.

Which means that in absence of those other positional data and a weak GPS signal that you get wider margins of error in location and heading calculations. I've seen this happen on a variety of different phones. It's why in the apps that show the area where they think you are gets much, much larger when you have no data connection along with obviously a weak GPS signal. Both things that person brought up can be seen on a wide variety of GPS devices if they are having trouble getting very accurate positional and heading data. If this were simply a iPhone 5s and iOS 7 issue then I would not have been able to see it on Android phones I've owned and on my iPhone 5 running iOS 6. And even with accurate positioning I've seen apps like Waze have my vehicle pointing in the wrong direction then the way I'm going on more than a few occasions.

 

 
post #37 of 41
My master thesis was about self calibrating accelerometers and magnetometers in attitude determination applications. It focused in the potential use of "worse" sensors for some non-critical processes, with good enough direction accuracy.

The figures the expert said are good, from my experience, bad sensors give alignment errors of about 5-10 degrees. The self calibration method I developed brought those down to 1-2 degrees! It's really easy to implement this feature via firmware update, and correct this bug.

What rubs me the wrong way is apple going for cheap (and worse) silicon on a premium priced device...
post #38 of 41
I have the 5c and it has the same issues. I even compared it to my friend's 5s. Is it really the hardware or is it just the calibration? Or does the 5c have the same accelerometer as the 5s? Either way, MotionX GPS Drive seems to read correctly. So does Theodolite. So I'm not too sure if its hardware.
post #39 of 41
To check a level. Place it on a flat surface, look at the vial, note the location of the bubble. Turn the level 180 degrees, and look at the vial. If the bubble is in the same place, the level is level.
post #40 of 41
I hope they fix this soon. Just three days ago I nicked a wingtip on the roof a uranium refinement facility while flying my F22 on a recon mission over North Korea. I was using my 5s as a HUD coz the Raptor's dash has too many blinky lights.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: iPhone
  • Apple's iPhone 5s sensor woes may be linked to new accelerometer supplier
AppleInsider › Forums › Mobile › iPhone › Apple's iPhone 5s sensor woes may be linked to new accelerometer supplier