Tests suggest some iPhone 5s motion sensors may be miscalibrated
Apple's iPhone line is known for its build quality and reliability, but a recent examination appears to show that the sensors in at least some models of the iPhone 5s are noticeably miscalibrated.
Developers of many apps have taken advantage of the sensors Apple builds into its bestselling handsets, using the technology for games, measurements, navigation, and more. Now a series of tests conducted by Gizmodo on two separate iPhone 5s units show those units giving inaccurate measures for the device's level, accelerometer, compass, and gyroscope.
The tests compared the iPhone 5s' results against those of the iPhone 5 and a number of dedicated physical measuring tools. In testing the device's leveling calibration, the iPhone 5s' readout was 2-3 degrees off when compared against a Stanley spirit level. Other users reported that the 5s' readout was off by 4-6 degrees. The iPhone 5, though, gave a reading nearly identical to the physical level, indicating that hardware could be the issue.
The gyroscopic function in the iPhone 5s ? which relies on the same hardware as the inclinometer ? fared little better. The iPhone 5 showed the expected 0-degree readout when set on a level table. The iPhone 5s, however, showed -3 degrees off level.
In additional tests, the compass app on the iPhone 5 showed a more accurate reading for magnetic north when compared against a real compass than it did on the iPhone 5s. The iPhone 5 and 5s were consistently measured at 8-10 degrees apart. The accelerometer on the newer iPhone also reliably registered more latent motion than did the one in the iPhone 5.
Gizmodo notes that its testing is far from conclusive, but that it is in line with what has been seen in other tests of the device's sensors. It is unclear to what extent the flawed readings stem from hardware, but the report expresses confidence that at least some of the problem could be mitigated with a software update. A number of users have reportedly taken their devices back to exchange them, and some of those report that the problem has disappeared.
Developers of many apps have taken advantage of the sensors Apple builds into its bestselling handsets, using the technology for games, measurements, navigation, and more. Now a series of tests conducted by Gizmodo on two separate iPhone 5s units show those units giving inaccurate measures for the device's level, accelerometer, compass, and gyroscope.
The tests compared the iPhone 5s' results against those of the iPhone 5 and a number of dedicated physical measuring tools. In testing the device's leveling calibration, the iPhone 5s' readout was 2-3 degrees off when compared against a Stanley spirit level. Other users reported that the 5s' readout was off by 4-6 degrees. The iPhone 5, though, gave a reading nearly identical to the physical level, indicating that hardware could be the issue.
The gyroscopic function in the iPhone 5s ? which relies on the same hardware as the inclinometer ? fared little better. The iPhone 5 showed the expected 0-degree readout when set on a level table. The iPhone 5s, however, showed -3 degrees off level.
In additional tests, the compass app on the iPhone 5 showed a more accurate reading for magnetic north when compared against a real compass than it did on the iPhone 5s. The iPhone 5 and 5s were consistently measured at 8-10 degrees apart. The accelerometer on the newer iPhone also reliably registered more latent motion than did the one in the iPhone 5.
Gizmodo notes that its testing is far from conclusive, but that it is in line with what has been seen in other tests of the device's sensors. It is unclear to what extent the flawed readings stem from hardware, but the report expresses confidence that at least some of the problem could be mitigated with a software update. A number of users have reportedly taken their devices back to exchange them, and some of those report that the problem has disappeared.
Comments
/s
For the inclinometer, if you tap the screen, it will reset whatever orientation the iPhone is at to 0 deg. While the default may be off, perhaps just resetting will fix some of the errant readings. Seems like another tempest in a teapot.
Apple is doomed and double-doomed. Let me use my boiler-plate text: This would never have happened if Steve Jobs were alive.
/s
Or at least such repetitive, meaningless posts wouldn't occur. zzzz
Have you no desire to appear intelligent and original?
All you have to do is tap the screen and it recalibrates. Problem solved...
Though this is only a temporary fix because a second tap will return it to it's default calibration...
I can confirm this. Tested with my 5S: -2 to -4 degrees registered on several flat surfaces.
A quick test of a friend's iPhone 5 reads a perfect 0 degrees.
Hopefully it's a software update, otherwise I'll be making a Genius Bar appointment.
Other things that would not have happened
1. iOS7 would have increased functionality a lot but changed the user interface hardly at all (because it was good and getting better)
2. There would be no colored iPhones
Something that would have happened: Steve would have removed Jony Ive from control over iOS7 and anything else that has to do with user interface.
I am sick to the teeth of Apple guys accusing me of simply resisting change. I only resist change when it accomplishes nothing.
I actually had been wondering about this. I was using the 5s to hang some pictures the other day. My 5s was always a few degrees off. It's always either -2 or -3 on flat surfaces. I'm not too worried about it though. If there is an issue, I'm sure Apple will fix it soon.
Other things that would not have happened
1. iOS7 would have increased functionality a lot but changed the user interface hardly at all (because it was good and getting better)
2. There would be no colored iPhones
Something that would have happened: Steve would have removed Jony Ive from control over iOS7 and anything else that has to do with user interface.
I am sick to the teeth of Apple guys accusing me of simply resisting change. I only resist change when it accomplishes nothing.
Agreed.
I think that Sir Jony is an excellent designer of cool, clean hardware products, but the iOS7 interface takes "minimalism" to an unnecessary and confusing extreme, changing the look & feel for no apparent reason. Some of the changes in iOS7 not only accomplish nothing, they are retrograde.
Skeuomorphism is a good usability tool because it links interface functionality in an intuitive way to known objects (like a contact app that looks like an address book or a calendar that looks like a calendar) - it makes the interface more easy to use by eliminating the need to search for a function.
Usability is not the same as graphic design - usability is about making the interface disappear as far as possible - graphic design is more about pixels and paint.
I doubt you'll get much agreement about these concepts on this forum though ;-)
Now it makes sense. I built a table using no tools other than my iPhone. It has a ruler app and a leveling app so why would I need anything else? I even used two perpendicular sides as a speed square and the edge to hammer nails. When I was done I placed it in the middle of my kitchen and put my iPhone on top to do some last minute leveling. The table was soooo crooked, out of square and un-level that my iPhone slid off the table and smashed on the floor.
Since I used my iPhone as the sole tool to create my table and it smashed to bits when it slid off the table, can I sue Apple for a new phone because of faulty hardware?
@flat surface people
Floors/tables are usually flat surfaces, but it doesn't mean they are level. As the article mentions, you have to use some other accurate measuring device as a control, such as a spirt level.
I'm running iOS7 on an iPhone 4. Can't say much about the level (I'll leave that to DickPrinter's lovely post!), but the compass is inaccurate when lying on my (level or not) desk surface but corrects when I pick it up and move around. Perhaps the compass is confused about Magnetic North by the metal frames, screws etc., of the desk?
…tests conducted by Gizmodo…
And why are we believing them?
Have you no desire to appear intelligent and original?
Sar. casm.