Hacker News Reader

Comment by bri3d | Hacker News Reader

https://github.com/Vladislav98759/Macbook-Lid-Angle-Sensor-C...

No, it actually does set the zero point.

Interesting but that proves the point even more --- it's hardly "calibration" when it effectively does nothing more than write constants to the EEPROM. They certainly have enough processing power in the machine to do that automatically too without needing anything more, but instead they make everyone go through a whole song-and-dance to do this trivial process; which doesn't even require Apple's involvement.

"Set the angle to 0 (closed) and press Enter. Open 10 degrees and press Enter. Repeat for every 10 degrees from 0 to 170" would be an example of actual calibration.

Pressing enter while closed might be a little difficult.

3rd party keyboard :)

Apple even makes a suitable one themselves... but the point is that a calibration procedure involves adjustment and measurement, and not merely reading some data from the sensor and writing it back. If Apple weren't deliberately trying to be hostile and sneaky, they would not have bothered with this roundabout, obfuscated process which no doubt increases their production cost too.

Minimum angle seen during a full motion would be perfectly reasonable.

The calibration measures only when the screen is open and closed. There is a grace period of a few seconds to allow you to close the screen.

But yes, if you don’t do it correctly, you will burn the service part and have to replace it again.

That seems an extreme measure to ensure the security/integrity of a screen hinge sensor input ( or protect this tiny revenue stream )

It does actually perform a security function. The lid angle sensor is used to know when the device is open or closed, and when closed, it physically disconnects the microphone. If you were to be able to recalibrate it at any time, you would leave your device vulnerable to having the microphone enabled when the lid is closed. You can argue whether that justifies the practice, but it's not as simple as just burning the EEPROM serial number in that tells it to turn the display on or off. It defends the user against an attack vector.

From that perspective making it one-time programmable is not unreasonable.

I agree with you

Though it could be simpler if it was something like a magnet on the lid that activates a magnetic switch on the bottom part (and it would be harder to have a false negative result). But Apple is going to Apple

Yes, it could be done with a Hall effect sensor or something like they used to. The cool thing about this approach is they actually use a different angle to turn the screen off as you close the lid than they do for turning it on when you open the lid, to create a better experience. Since it is a security feature, then the "open" vs "closed" state should use the same source of truth. So it's a trade-off of complexity and experience.

It feels like Apple's implementation leans more toward vendor lock-in than purely user protection

It should also be relevant for triggering the "closed let's lock the device" event, right?

Trivial with a delay, press enter and close, after 5 seconds it calibrates

You don't even need that, just ask the user to close the lid and see what the sensor settles on.

How would you know the user is ready to perform the command?

You could use the camera to estimate if the laptop is closed.

Exactly my thought, but this might not be precise enough in low light conditions? So maybe best used as a fail safe (if camera is lit don't calibrate even if the user pressed confirmed)

It wouldn’t matter… just tell the user to close and open the lid before pressing enter again, and get the maximum (or minimum) value as the lid being close all the way.

like the answer to any other similar question, with a prompt?

Prompt for what? A Key press?

You really can’t solve this on your own?

Do Macs not support USB-keyboards?

USB keyboard?

Just plug in an external keyboard.

> it's hardly "calibration" when it effectively does nothing more than write constants to the EEPROM

What do you think calibration of digital devices entails?

It does involve exactly that. Whereas in an analog device you would be adjusting a potentiometer or something similar

You're forgetting (or purposely omitting) the measurement aspect.

The measurement aspect is obvious to anyone who thinks two seconds about the point I'm making

The measurement happening on the lid is the "lid closed" point

That's not something that needs to be measured externally.

Also, one wonders just how horrible their production tolerances must be if something like this even needs that sort of calibration in the first place. No other company does this. Nonetheless, even if one accepts this bizarre excuse at face value, it doesn't take a genius to realise that the firmware can "self-calibrate" trivially: If the EEPROM in the sensor is uninitialised, then it just needs to record the max/min value the first time it's closed, and save that as the "lid closed" point.

The only plausible explanation for doing all this extra work is that this isn't merely incompetence; it's intentional malice. Given how much Apple spends on lobbying and other hostile activities, this is not surprising.

Reply on news.ycombinator.com