Timestamping the data

[praveen2600]

I understand that now. If it is the phone's limitation then one can't do much here. 

However, the Science Journal app (I am not sure if I am allowed to compare the functions of two apps here, so apologize in advance) has this function and it doesn't show any discrepancy when used the same phone.

[Jens Noriʇzsɔɥ]

We are fully aware of other apps, so, no prob…

What exactly are you referring to? If it is timestamps: perhaps, they treat start und stop differently? Hm, on my iPhone the data in Science Journal is not at full rate, so there could be some averaging – and timestamps are artificial anyways then.

[praveen2600]

We are fully aware of other apps, so, no prob…

What exactly are you referring to? If it is timestamps: perhaps, they treat start und stop differently? Hm, on my iPhone the data in Science Journal is not at full rate, so there could be some averaging – and timestamps are artificial anyways then.

Here is what I am talking about: 

Look at the image here . I am recording the magnetic field in Phyphox and the Science Journal. 
Why the values in H2 and H9 are different from M2 and M3? 
Recording the data using system time is very helpful when I can sync data from two independent phones and do time analysis. However, getting the discrepancy in raw and metadata will leave a margin of error here. 

H15 is the recording from the Science Journal app where it starts with 0. Perhaps the science journal app records only the system time data and then gives the relative time. It is a little gray area where I will never know if there is a discrepancy in Science Journal metadata and Raw data since it doesn't give one in one file. 

Could that be a limitation of my phone?

[Jens Noriʇzsɔɥ]

First af all, phyphox gives you the time that the system provides us for each data point. As already explained, it is almost impossible that these times for the first event coincide with the moment you press the start button – that's why it's called START in L2…

Whatever Science Journal does: they give you values at a far lower rate (reproduced on two smartphones here) and – as already explained – I have no idea what these times mean at all, I would regard these as artificial, not corresponding to sensor events. Could you please elaborate on how this is better for synchronisation? I simply do not understand your point in this regard…

Speaking of synchronisation: as already explained, smartphone clocks are typically not a good time reference, off by seconds or even more. How do you manage to get two data sources into proper time sync for rates of 100 Hz?

[Rookie]

兄弟，100HZ以下的加速度计和陀螺仪的时间戳问题能解决吗？正常情况下时间为0.01s。当加速度计、陀螺仪和地磁同时工作时，地磁测量数据比前两者多。

[Jens Noriʇzsɔɥ]

Please apologise if I do not exactly answer your questions as I somehow need to rely on what google translate tells me (deepl appears to be down atm).

The answer depends a bit on what you understand as “problem”. The exact timestamps of each and every data point are now giving in the app and the export files. Some sensors are (strangely) in sync, like gyroscope and magnetometer on my iPhone (the accelerometer is not even though it should be included in the same IMU). That said, a smartphone clock is a rather bad time source. So, I am a bit unsure if there is any reasonable way to synchronise sensors of different devices.

I do not get what you mean by the magnetometer is “more” compared to accelerometer and gyroscope. If it is rate, it should be quite the contrary, see https://phyphox.org/sensordb/