Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
So whatever it is worth, I picked the Gold 32 GB ATT yesterday from Apple Store.

First thing I checked after coming home is gyro :)

Well it is off by -4 when flat and -2 when vertical on longer edge. Not sure what I should do though.

My phone is from week 40 (October).
 
So whatever it is worth, I picked the Gold 32 GB ATT yesterday from Apple Store.

First thing I checked after coming home is gyro :)

Well it is off by -4 when flat and -2 when vertical on longer edge. Not sure what I should do though.

My phone is from week 40 (October).

That is the newest phone I've seen reported and still a huge error
 
okay I would like to know if this works for all

In iHandy if I level it and calibrate it my gives me the correct reading for all the sides and also when it is flat on the back.

Is that correct?

if u've calibrated then yes it should. watched over the next few days for drift and report please.

note that the calibration using ihandy is for that app only if u run a gums such as asphalt that needs level info it will pull the un calibrated results.

----------

So whatever it is worth, I picked the Gold 32 GB ATT yesterday from Apple Store.

First thing I checked after coming home is gyro :)

Well it is off by -4 when flat and -2 when vertical on longer edge. Not sure what I should do though.

My phone is from week 40 (October).

damn, what factory?
we await reports from week 41 now owners.
u in the same boat as a lot if people, shall I exchange/ return while in return period?
shall I leave it and hope its software not hardware?
if hardware, and I leave beyond return period I'll get a refurb, are refurbs new?
shall I exchange/ return while in return period?
 
So whatever it is worth, I picked the Gold 32 GB ATT yesterday from Apple Store.

First thing I checked after coming home is gyro :)

Well it is off by -4 when flat and -2 when vertical on longer edge. Not sure what I should do though.

My phone is from week 40 (October).

FWIW: week 40 is last week of September not October. Still most resent phone with the problem.
 
Wait

So whatever it is worth, I picked the Gold 32 GB ATT yesterday from Apple Store.

First thing I checked after coming home is gyro :)

Well it is off by -4 when flat and -2 when vertical on longer edge. Not sure what I should do though.

My phone is from week 40 (October).

Wait to see if a software update is released and that it corrects the problem. I have a strong feeling that this will get corrected soon, the longer Apple waits the worse things will get. I heard that 7.0.3 will be out this week. If so, hope it addresses this issue. If you are not comfortable with issue, you have 14 day (I think)
 
I've been using and calibrated both apps iHandy Level and Clinometer + slope finder + bubble level (3 in 1) for about 48 hours now and haven't noticed any drift. The settings are all still 0 or 90 degrees (depending on the app). This is how is suppose to work. Now if Apple can add such a feature in the software so that it will automatically calibrate or perform a manual calibration, we should be able to use the level and play gyro games without any issues. Rumor as it, IOS 7.0.3 suppose to be released this week. I hope it gets fixed real soon.
 
I've been using and calibrated both apps iHandy Level and Clinometer + slope finder + bubble level (3 in 1) for about 48 hours now and haven't noticed any drift. The settings are all still 0 or 90 degrees (depending on the app). This is how is suppose to work. Now if Apple can add such a feature in the software so that it will automatically calibrate or perform a manual calibration, we should be able to use the level and play gyro games without any issues. Rumor as it, IOS 7.0.3 suppose to be released this week. I hope it gets fixed real soon.

I very much doubt that the software that fixes this will be released this week, it's too soon. if it is a software fix and is indeed in 7.0.3, then 7.0.3 will be released much later than this week. hope I'm wrong.
 
When you open the compass app you have to calibrate it by rolling the little ball around the circle. There is no reason Apple couldn't release an update so that when this sequence is performed it calibrates YOUR accelerometer regardless of whether it reads differently to the next person's phone.

I really believe its something that can be easily corrected by software. I think anyone returning their phones (which are otherwise perfect) are seriously jumping the gun.

It may also be the case that the initial calibration sequence that needs to be performed when opening the compass app isn't doing its job correctly thus explaining the apparent drifting some people talk about. maybe each time the app opens, the calibration sequence sets it slightly differently each time.

Oh and for what its worth, mine is off by about 1.
 
This is what I was thinking since it uses the accelerometer to roll the ball around. Maybe by rolling the ball around the accelerometer can figure out where gravity is and store it globally. It used to be the figure 8 pattern for the compass which is more accurate in my opinion but maybe they wanted to kill 2 birds with 1 stone since they incorporated the level app. It sure would be nice if it was just a simple switch they forgot to enable and 7.0.3 could fix it. Again don't hold your breath.

When you open the compass app you have to calibrate it by rolling the little ball around the circle. There is no reason Apple couldn't release an update so that when this sequence is performed it calibrates YOUR accelerometer regardless of whether it reads differently to the next person's phone.

I really believe its something that can be easily corrected by software. I think anyone returning their phones (which are otherwise perfect) are seriously jumping the gun.

It may also be the case that the initial calibration sequence that needs to be performed when opening the compass app isn't doing its job correctly thus explaining the apparent drifting some people talk about. maybe each time the app opens, the calibration sequence sets it slightly differently each time.

Oh and for what its worth, mine is off by about 1.
 
I have my replacement, it's worse -4 on back -3 on long side!

Will update with build week.

I will log the call with apple but pointless swapping again :(
 
This is what I was thinking since it uses the accelerometer to roll the ball around. Maybe by rolling the ball around the accelerometer can figure out where gravity is and store it globally. It used to be the figure 8 pattern for the compass which is more accurate in my opinion but maybe they wanted to kill 2 birds with 1 stone since they incorporated the level app. It sure would be nice if it was just a simple switch they forgot to enable and 7.0.3 could fix it. Again don't hold your breath.

How was the figure of 8 thing supposed to work? Is there any reasoning behind it except being a nicer way of saying wave the phone all over the place to hopefully get a decent average?
 
How was the figure of 8 thing supposed to work? Is there any reasoning behind it except being a nicer way of saying wave the phone all over the place to hopefully get a decent average?

Actually, the figure 8 method seems to track better when I compare it to my Silva compass. It also seems to finish the calibration quicker.
 
well, maybe since the level app was included with ios7, they will remove it in 7.0.3 and we'll have nothing to complain about. Impliment it again in a future release.
 
I checked about 10 iPhone in the Apple Store Berlin

So here is a quick report of my experience:

I checked my iPhone 5s 64GB (Week 39/October/Factory C3) against a real life level and 3x iPhone 4s on a flat surface. I checked with the new iOS 7 level app AND a free app called DUALLEVEL.

ALL iPhone 4s and the real level were 0° and my iPhone 5s was off 1° on the iOS 7 app and roughly
2° on the DUALLEVEL which seemed to have a finer readout to me.


THEN I went to the APPLE STORE IN BERLIN/Germany and checked out about 10 iPhone 5s on their presentation table against my iPhone 4s.
The table was not always leveled correctly BUT since I checked with my iPhone 4s (which was congruent with 2 other 4s) I had a sort of reference.

TO MY SURPRISE:
Each and every iPhone 5s that was displayed there was 2-4° off to my iPhone 4s.
I encountered not one phone that wasn't off. They ALL had a problem in one way or another.


So I don't know whether or not it is "fixable" with a software update!

I wait until thursday next week and if there is no update around I will send it back to get either a replacement or a refund. Depending on reports from users if their replacements got better.
I shelled out 900 EUR (1222 US Dollar) for this gadget. It should at least have the same quality as the last iPhones - or better! Especially when they advertise the accuracy of their M7 together with all the sensors.

I want to take photographs with this device and in my app Camera+ the "artificial horizon" is also off. So that means if I take this horizon tool as reference my photos will be crooked.
Not good!
 
I want to take photographs with this device and in my app Camera+ the "artificial horizon" is also off. So that means if I take this horizon tool as reference my photos will be crooked.
Not good!
Having installed Camera+ on my 5s today I can confirm the same too, it's off by roughly the same level as it is in the level app in the compass.

I also did a DFU on my phone twice and the results have stayed the same, if anything on it's back has drifted from -3 to -4 and even -5 on some occasions!
 
I've been using and calibrated both apps iHandy Level and Clinometer + slope finder + bubble level (3 in 1) for about 48 hours now and haven't noticed any drift. The settings are all still 0 or 90 degrees (depending on the app). This is how is suppose to work. Now if Apple can add such a feature in the software so that it will automatically calibrate or perform a manual calibration, we should be able to use the level and play gyro games without any issues. Rumor as it, IOS 7.0.3 suppose to be released this week. I hope it gets fixed real soon.

Likewise...the initial calibration I did in iHandy has stuck since I installed it, even with repeated close / relaunch cycles. I'm a layman in this arena, but it would seem that iHandy uses the same hardware sensors as the native app, which points to software (perhaps firmware) as the root of the problem, not neccesarily hardware.
 
----------

[/COLOR]

damn, what factory?
we await reports from week 41 now owners.
u in the same boat as a lot if people, shall I exchange/ return while in return period?
shall I leave it and hope its software not hardware?
if hardware, and I leave beyond return period I'll get a refurb, are refurbs new?
shall I exchange/ return while in return period?[/QUOTE]

+1
 
Im 100% convinced this can be fixed by calibration as we can see with ihandy and other apps that allow this. Its up to Apple to decide how to implement it. So even if we start seeing ones out of the factory in the future that are correct that doesn't necessarily mean they changed the hardware. They may start calibrating them before they leave the factories. I think the harder part for Apple is how to implement it for the units in the wild.

The only thing that bugs me a bit is the drifting but user/auto calibration would help resolve that.


Likewise...the initial calibration I did in iHandy has stuck since I installed it, even with repeated close / relaunch cycles. I'm a layman in this arena, but it would seem that iHandy uses the same hardware sensors as the native app, which points to software (perhaps firmware) as the root of the problem, not neccesarily hardware.
 
I can confirm that my 5S also has the gyro problem

Horizontal (long side on table) is correct at 0 degrees
Verticle is off by -3 degrees
Tilt is off by -2 degrees

I also find Compass/Leveler app would go out of wack. If I flip the phone a bit and turn it around, it would entirely throw off the readings. The only way to get it back to normal reading is the close the app and "re-calibrate" from the start up screen.

Also, when I use google map navigation in 1st person view, the viewpoint would be tilted in one direction. Also the movement of the status would be jaggy and jump and shakes. This also led me to believe the acceleoramter, etc. is not providing proper value either (as mentioned in the gizmodo article).
 
Last edited:
Am I the only one who has different values depending on whether on which long side I put the phone?

There's about a difference of 4 between the left and right side.
 
Am I the only one who has different values depending on whether on which long side I put the phone?

There's about a difference of 4 between the left and right side.

Are you measuring the side with the buttons so the buttons are off the edge of the table? No, then that's why the button side is off by one more degree ;)
 
Are you measuring the side with the buttons so the buttons are off the edge of the table? No, then that's why the button side is off by one more degree ;)

Actually yes I am messing with the buttons off the table.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.