Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
iPhone 5s accelerometer/gyroscope/compass values incorrect
- Thread starter tharepairguy
- Start date
- Sort by reaction score
Restore Results
I wiped my iphone 5s and downgrade the firmware to 7.0.1 (wish there were a 7.0 out there) and setup the iphone as new and then checked the level out and it was off 2-3 degrees but when I updated the firmware and restored my iphone back with my itunes backup, it's now off by 4-5 degrees. Strange and this points to a software related issue.
I wiped my iphone 5s and downgrade the firmware to 7.0.1 (wish there were a 7.0 out there) and setup the iphone as new and then checked the level out and it was off 2-3 degrees but when I updated the firmware and restored my iphone back with my itunes backup, it's now off by 4-5 degrees. Strange and this points to a software related issue.
Actually, it was the iPhone 4 that had the proximity sensor issue. Mine didn't have it, UNTIL the software update that "fixed" it. Oh I was so mad.
iPhone 5s accelerometer/gyroscope/compass values incorrect
I don't think the difference is constant. When I tested my iPhone 5S earlier it was getting 0 on the right side and bottom and 4 on the left and tool. When I tested it later it was around 2 on the bottom., 1 on the right and 3 on the left.
Lying on it's back is pretty much consistently off by approximately 3 and the direction always leans from top left of phone to bottom right, no matter how the phone is oriented.
I find it odd that there's been no official statement from Apple on this despite some support people telling people it will be fixed in the next software release (while others say an exchange is needed). That leads me to believe there's no official doctrine from Apple on what support should say, meaning a quick fix is unlikely.
I don't think the difference is constant. When I tested my iPhone 5S earlier it was getting 0 on the right side and bottom and 4 on the left and tool. When I tested it later it was around 2 on the bottom., 1 on the right and 3 on the left.
Lying on it's back is pretty much consistently off by approximately 3 and the direction always leans from top left of phone to bottom right, no matter how the phone is oriented.
I find it odd that there's been no official statement from Apple on this despite some support people telling people it will be fixed in the next software release (while others say an exchange is needed). That leads me to believe there's no official doctrine from Apple on what support should say, meaning a quick fix is unlikely.
On a flat surface, my iPhone 4S was getting 0 on side and 0 on back.
I was getting -2 on side and -3 on back of iPhone 5S 16GB Gold from AT&T.
I went AT&T store in Mountain View, CA and told the gyroscope was defective and wanted to return it.
The manager Corey came out and told me it was normal and showed that his black iPhone 5S was getting -6 on the side. Then, He told me I should take my phone to Apple to see if they could fix it. I told him I wanted to return it and he said he would charge me a $35 restocking fee because the tilt was minimal.
I paid the restocking fee and returned it. I called AT&T afterwards and they waived the fee. Next time, I'm buying directly from Apple for 30 days exchange/refund and no restocking fee. Once it is confirmed fixed from Apple, I will order another iPhone 5S from Apple.
I bet Apple will fix the calibration issue or hardware issue in the factory. It would be too expensive for them to recall all the defective units.
Then provide a software calibration workaround for the initial buyers.
I was getting -2 on side and -3 on back of iPhone 5S 16GB Gold from AT&T.
I went AT&T store in Mountain View, CA and told the gyroscope was defective and wanted to return it.
The manager Corey came out and told me it was normal and showed that his black iPhone 5S was getting -6 on the side. Then, He told me I should take my phone to Apple to see if they could fix it. I told him I wanted to return it and he said he would charge me a $35 restocking fee because the tilt was minimal.
I paid the restocking fee and returned it. I called AT&T afterwards and they waived the fee. Next time, I'm buying directly from Apple for 30 days exchange/refund and no restocking fee. Once it is confirmed fixed from Apple, I will order another iPhone 5S from Apple.
I bet Apple will fix the calibration issue or hardware issue in the factory. It would be too expensive for them to recall all the defective units.
Then provide a software calibration workaround for the initial buyers.
Thanks for correcting, I couldn't remember if it was the 4 or 3g, for I owned both and never had the issue myself. Cheers!!
----------
I've been testing this for about a week now and the the values are constant for me except when I downgraded the firmware and set the iphone up as new.
Yup that's him and he reported some of us for using the T word and got the thread closed. He also pulled the same thing on the cultofmac article. Told some dr he needs to prescribe more OCD meds. His opinions are useless.
Last edited:
I hope this is true. But one tech saying what we want to hear, doesn't make it true. He could very well be blowing smoke up our butt. Most reps are well trained experts at doing that.
I can also confirm a senior support member emailing me that engineering got back to them with an update- but wanted to discuss over the phone with me on Monday what engineering relayed.
Leaves me a bit worries though that it needs to be discussed over a phone call. If it was going in the direction of a software update wouldn't they just say that in an email ?
From past experiences, I totally agree with you. However, he told me to keep his direct phone number and if the issue isn't resolved after a couple firmware updates, give him a call and not to go through an Apple store for the employees there can be annoying at times. He stated that the problem was wide spread and that it was a high priority and it's most likely software. I'll take his word for now but if it doesn't get fixed soon, he'll be getting a call.
I read on other sites that IOS 7.0.3 is due out next week with build 11B491, I hope there's a fix in it. If not, I guess it's not a priority as I was told. Cheers!!
Yel, does sound a little strange. But so does the original email. They normally don't email at all because they don't want anything documented. I was told yesterday that it's most likely software and to all the tests I've done and seen, it seems like it can be fixed by software. If not, this will not be good for Apple. They first of all got on stage and told the world about how great this M7 chip is and now we all learn it doesn't work properly. This is a wide spread problem and if it's hardware, it's not going to be good for Apple. Didn't they test / check all this stuff before bragging about it and releasing to millions of users. Gosh, someone should be fired!!
Right. One senior engineer writes an email and the whole issue spreads to all tech sites and explodes.
The thing is: There might only be a slim chance to provide a software fix.
If the sensor is off, you can only compensate by providing data with less jitter, causing less accuracy.
Just imagine everyone sends back the whole first batch of 5s.
I personally think the issue will be fixed with a new hardware revision that will not be announced at all. Keep in mind that the M7 only collects and aggregates motion data, the sensor hardware is not part of that.
Apple probably switched suppliers to switch to cheaper or more integrated components.
I'm sure we will never here from it again, because it's only 10 month until the iPhone 6 hits the markets.
There's no way in hell Apple will change the current design just because a sensor is not that accurate if it's a major redesign with regards to the logicboard or the M7 coprocessor.
The thing is: There might only be a slim chance to provide a software fix.
If the sensor is off, you can only compensate by providing data with less jitter, causing less accuracy.
Just imagine everyone sends back the whole first batch of 5s.
I personally think the issue will be fixed with a new hardware revision that will not be announced at all. Keep in mind that the M7 only collects and aggregates motion data, the sensor hardware is not part of that.
Apple probably switched suppliers to switch to cheaper or more integrated components.
I'm sure we will never here from it again, because it's only 10 month until the iPhone 6 hits the markets.
There's no way in hell Apple will change the current design just because a sensor is not that accurate if it's a major redesign with regards to the logicboard or the M7 coprocessor.
Seems like a lot of people at apple think that plugging in the phone for 90 mins will calibrate it. Maybe this was something that existed pre ios7 or maybe it exists now just not working or enabled in the 5s for some reason. Maybe this is all that has to be fixed. I'm tempted to test this with my ipad at an angle and see if it changes at all. Thoughts?
Same Here
I spoke to them a couple nights ago and they told me to backup my iphone in iTunes, reset to factory, setup as new, connect power adapter, turn on Do Not Disturb, place the iphone on a flat surface facing up and then wait 90 minutes. The results were exactly what I had expected, no change. The Tech supervisor told me it was a wide spread issue and it appears to be a software issue and can expect a fix very soon since it's a high priority. I hope so.
Is there anyone still on the original IOS 7 and didn't upgrade? If so, do you have the same issues? I downgraded my iphone to 7.0.1 and set it up as new and my error was 2-3 degrees but when I updated to 7.0.2 and restored my backup, it's now 4-5 degrees out and it's been constant.
Just wondering if this problem was created in the IOS 7.0.1. I can't understand IOS 7 having this issue too and at the WWDC they were so proud to introduce the M7. Someone didn't do their homework.
I spoke to them a couple nights ago and they told me to backup my iphone in iTunes, reset to factory, setup as new, connect power adapter, turn on Do Not Disturb, place the iphone on a flat surface facing up and then wait 90 minutes. The results were exactly what I had expected, no change. The Tech supervisor told me it was a wide spread issue and it appears to be a software issue and can expect a fix very soon since it's a high priority. I hope so.
Is there anyone still on the original IOS 7 and didn't upgrade? If so, do you have the same issues? I downgraded my iphone to 7.0.1 and set it up as new and my error was 2-3 degrees but when I updated to 7.0.2 and restored my backup, it's now 4-5 degrees out and it's been constant.
Just wondering if this problem was created in the IOS 7.0.1. I can't understand IOS 7 having this issue too and at the WWDC they were so proud to introduce the M7. Someone didn't do their homework.
software patch will just massage the readings, they will never be accurate, will be fixed quietly when the 6 is out if the 5s is not discontinued. they have sold too many, they will not do a recall nor will the change they hardware as that will lead to millions demanding replacements. really bad for what is the most expensive phone on the market. just live with it guys and enjoy all the other features. if u haven't bought one yet just buy the smallest capacity (or a 5/5c) as it's the cheapest, this is not a phone to keep for long and maybe the iphone with the worst resale value.
Apple hasn't even official acknowledged the problem exists, let alone stated whether it's a hardware or software problem.
Contrary to what people have been reporting more recently from AppleCare I was told on Thursday that it was almost certainly a hardware issue that will be covered by the warranty.
I really hope it turns out to be software related but wouldn't bet on it.
Apple should really own up to this one and at least fix it in the factory so replacements actually work!
This works for me i.e. the level in the Apple compass app shows -1 but once 0 is calibrated in iHandy level it shows 0 degrees even when my 5s is turned around 180 degrees. So if Apple could add a system wide calibration that worked the same way as the iHandy level calibration surely this would solve the problem?
The problem could have any number of causes:
1. Faulty or uncalibrated sensors - if sensors are same in 5C and 5S, this seems unlikely.
2. Flaw in the M7 chip.
3. Bug in iOS 7 reading the chip (rounding error or other bug with 64-bit code)
If it's #1 there's no way to blanket fix it through software as each sensor would need to be calibrated individually on each phone.
If it's #2 a fix could be possible, depending on what it is. For example the original Intel Pentium chip had a floating point math error in it, that was fixed via a firmware update.
Obviously if it's #3 it can be fixed in software.
1. Faulty or uncalibrated sensors - if sensors are same in 5C and 5S, this seems unlikely.
2. Flaw in the M7 chip.
3. Bug in iOS 7 reading the chip (rounding error or other bug with 64-bit code)
If it's #1 there's no way to blanket fix it through software as each sensor would need to be calibrated individually on each phone.
If it's #2 a fix could be possible, depending on what it is. For example the original Intel Pentium chip had a floating point math error in it, that was fixed via a firmware update.
Obviously if it's #3 it can be fixed in software.
Surely if you're an Apple engineer (which is the only way you'd know this for sure) you shouldn't be posting like this on a public forum? On the other hand, if you're not an Apple engineer (which I strongly suspect) maybe you should just stop spreading FUD and wait with the rest of us for a proper statement on what the problem is and what's going to be done to fix it?
1 degree for you 4 or 5 for many other people. You're adding a lot of value to this thread. Sarcasm.