EskeRahn

https://developer.android.com/reference/android/view/KeyEvent#KEYCODE_FUNCTION

In my book it always raises suspicion when a very simple thing comes in a large package I immediately think: Hey is this compiled on an infected PC and used as a mule by some malware? And I'm especially cautious/suspicious when it is an app that (for good reasons) needs high privileges. So would be nice if you could dig into the "Why". 🙂

Just curious. Does this include some library, or how does this end up at 1918KB ? Browsing through the source even the small images in various resolutions could not get it anyway near (the source adds up to 1/10th).

He he, well I would rather have @Waxberry or @Erik do that.... AND it should have a 'secret' part we as users could not see. E.g. the priority of each bug/wish. And any known bugs no users have stumbled into yet, that they hope to fix before the public even know it existed. It is common practise on security bugs to not tell about them before they are fixed.

We really should have somewhere offical... We already got this unofficial one: https://docs.google.com/spreadsheets/d/15uE12Yv5nMvIF42U33cY5FQoF6M66QIn00cC876uf7Y/edit#gid=0

Yes they are identical, that is why @Anssi Hannula starts his qwertZ-files (e.g.this) by mapping. For image, see e.g. here ...That they are identical is actually an issue when setting up a new device (WiFi and account credentials), so we need to use the fake keyboard, as the real keyboard acts as the shifted until you get a chance to tell the system that it is the qwertZ. See also this

They are electrically identical, the only difference is the print, so your app will work identical on both. So you can test if you would like to (though the print will obviously not match when tested on a qwertY device) You can see the two stock prints described at the top here

Nice. I too would like to find a good one in leather. On the glue half part, I think it should be mentioned that For those using the back camera and/or LED a lot, it is recommendable to cut a hole, and perhaps glue the whole part (like you did). For those using it less, keeping it un-cut gives a better camera protection, at the price of needing to fold it all the way open to use the camera/LED. I like you idea of using a better inner bumper case, though to me that would be overkill The best one I've found (so far) isn't real leather either. I have bought

Yes that was tricky too but once you get it right it works flawlesly. I tried to make a little help-file, to show at 100%... ADD Sigh, the forum shrinks the image ... Here the full file: https://eskerahn.dk/wp-content/uploads/2020/03/Pro1_Hall_Sensor_Top.png It is roughly 300 pixel from the top and 200 from the Right (refferenced in portrait) Note that the orientation of the magnet matters, so you might need to turn it around back to front

That you only get half, has the benefit that you can investigate the detail of the fit to the curvature towards the side edges, and how the 'lip' is grabbing the edge. For the same reason you could consider cutting the part you already made in two, to see how good it fits the end-profile.

Interesting! Just a hint on the hole for fingerprint. You should give it slanted edges, otherwise it will be hard to get to.

Very interesting, thanks! I tag @Waxberry to make sure he sees this, so It can be forwarded to the right people. 🙂 Especially interesting that they even got a workaround..

BTW $550 is really a bargain price, I bet it will not be up for long. Had someone done the same in Europe with a qwertZ model, I would be VERY tempted to buy it as a spare.

So sad. Hope to see you back.

Too bad you can not get it working. Some US carriers require some trickery. Have you seen this thread?

Oh indeed, specially newer Bluetooth versions are more lenient, but for those of us not using it, it is still waste of battery, and hence stamina. I'm not saying that all radios should be off, just those that aren't in use and isn't needed.

Maybe this code sample from here can be modified to a more general list, of course of different type and the firstOrNull should obviously be replaced with some loop private lateinit var sensorManager: SensorManager private var mSensor: Sensor? = null ... sensorManager = getSystemService(Context.SENSOR_SERVICE) as SensorManager if (sensorManager.getDefaultSensor(Sensor.TYPE_GRAVITY) != null) { val gravSensors: List<Sensor> = sensorManager.getSensorList(Sensor.TYPE_GRAVITY) // Use the version 3 gravity sensor. mSensor = gravSensors.firstOrNull { it.vendor.contains("Googl

I know nothing on the interior of Android, but could the API return an ARRAY with values for the two sensors?

The list in Quicom's "Device Info" seems longer (50 sensors), But it does not seem to appear there either. It could be the one five from the bottom with the long name "BU52061NVX ROHM_HALL_EFFECT", but it does not show the readout from the sensor. So hard to say if that is the one See this

I know what the proximity sensor is ... BUT they just MIGHT be exposing these two Hall sensors as "TYPE_PROXIMITY" (Note: TYPE), The reason to believe so is the binary type output we see in the FactoryKit test-program. And if you are looking for it as a TYPE_MAGNETIC_FIELD sensor, no wonder you can not find it.

Ah, that made more sense.... Most likely it is handled as a proximity sensor, And if we are unlucky it might not be marked for wake up. https://developer.android.com/reference/android/hardware/Sensor#TYPE_PROXIMITY https://developer.android.com/reference/android/hardware/Sensor#isWakeUpSensor() But as the other Hall sensor IS clearly able to wake up the device, it seems unlikely that they are not both.

It is easy to find, it is near the end of the volume rocker. I have posted stuff showing the whereabouts above.

The same author also have a "Transparent CPU Monitor", that can show CPU and GPU as a transparent layer. That might be helpful in debugging?

I just ran it again WITH the tick box to include the temperature, and note on the first how the temperature of ONE of the cores is sky-rocketting in the first minute, and stays up during the entire test, while the other cores are at a more moderate temperature. They all start out at around 20 (it was placed at a table at a room temperature of 21) Also note that I have been EVIL, I did NOT take it out of the flip-case, so it did NOT have the cooling from the exterior it is designed to have.

I just tried the first on a S8- (exynos, so not the same) and the Pro1. And even throttled it is higher than my S8- before throttling. (The Pro1 was disturbed by a call I did not answer during the test, the red line)