I personally know someone who installed extra RAM on a Macbook Air at least, didn’t require modifying any firmware IIRC. It just detected it. Not sure how GPUs are these days, though.
So if you take a macbook air with soldered-on ram, which was available in 2 configurations of 16GB and 32GB, those two configs are baked into the bios, you just need the correct density, speed, and CAS latency to go from 16 to 32.
But if you stray from what’s in the firmware/bios, you’ll need to figure out how to get the ram recognized so it can be used.
In a raspberry pi, the known configs are in the firmware, so resellers were taking cheap low-ram pis, replacing the ram with double-density and selling them for a big markup from their cost, but lower than a retail RPI.
The new policy would force these resellers to mod the bios to keep doing it.
You still have to extract and modify the bios to recognize and address the new ram in those cases, you can’t just “drop in” more ram.
That’s what’s being locked down, here. Raspberry Pi devices used to not care if you did so because the firmware was open.
I personally know someone who installed extra RAM on a Macbook Air at least, didn’t require modifying any firmware IIRC. It just detected it. Not sure how GPUs are these days, though.
So if you take a macbook air with soldered-on ram, which was available in 2 configurations of 16GB and 32GB, those two configs are baked into the bios, you just need the correct density, speed, and CAS latency to go from 16 to 32.
But if you stray from what’s in the firmware/bios, you’ll need to figure out how to get the ram recognized so it can be used.
In a raspberry pi, the known configs are in the firmware, so resellers were taking cheap low-ram pis, replacing the ram with double-density and selling them for a big markup from their cost, but lower than a retail RPI.
The new policy would force these resellers to mod the bios to keep doing it.