The touch aspect itself is really cheap. The big benefit of screens over "simpler" instruments is that regulations actually vary quite a bit across the world. Being able to use a screen that can display anything greatly reduces the sku count, simplifying supply chains and assembly lines.

It would be hard to justify not using a screen these days, IMO.

> It would be hard to justify not using a screen these days, IMO.

Unless you were concerned about the safety of someone having to look at a screen to operate vehicle controls. Otherwise you could just make it a smartphone app and ditch the screen entirely.

That would be atrocious (but not really much worse than the all-in-one infotainment screens we already have...) But my gut says automakers haven't gone there because it would make support and liability that much more complicated - "the operator was distracted at the time of the collision... but they weren't trying to turn on the defroster, they were watching Youtube!" On the plus side instead of being stuck with a never-upgraded head unit, with an orphaned 3G modem, you'd at least have connectivity that kept working.

I should have been more clear. I meant the gauge cluster. TBH, having physical controls for things like climate, wipers, and lights makes sense. I can't imagine the cost difference is meaningful.

I doubt if it's actually cheap. "The touchscreen" aka 2DIN radio/infotainment head unit slot, is required for Japanese market, as Japanese users require one of indigenous options installed for all cars. Google Maps supposedly had gotten better but aren't quite replacing it, so CP/AA doesn't work. I believe this is a unique country requirement, as many Chinese EVs as well as many European low-end cars seem to ignore it.

Not every cars are made for Japanese, far from it, but abolition of LCD based map-radio on a standardized slot is a weird hill to die on, so every cars has it anyway.

And, I think, this is where the actual "it's cheaper to..." argument begins: since every cars will have the standard issue map-radio, it _can be cheaper to move buttons into it_. Differences between the standard version and one with this context is whether the cost of the entire 2DIN screen is included in it or not - if the screen cannot be removed, to shut up whining Japanese or not, it becomes sunk cost and becomes a constant term of equation, which physical buttons have no chance of winning.

The cost of the vehicle components is a surprisingly small part of the overall cost of a vehicle. The cost of the line time and the additional tolerances to fit those cheaper components is vastly higher than the cost of stuff they're going to install anyway.

Being able to separate design decision timelines on how the UI works from manufacturing timelines is also very helpful organizationally.

I think these are the right answers. LCDs are already wired and don't need someone's hands to do it. And you can, in software, however late you want in the process, add a button for this or remove a button for that. In software development this is a given, but in real world manufacturing it's like a super power.

Try pricing out physical switches at your local electronics shop sometime. How much does say, 20 or 30 of them cost? How much a 7" touch-screen LCD cost?

Then consider how long it takes to wire up and install the LCD versus 20-30 switches (plus mounting brackets, plus functional testing etc.)

Regulation forces the manufacturers to have the LCD in the car because rear view cameras are mandatory.

Right off the bat all radio controls are free if tossed into the software.