I think they should instead just introduce a clear system for verifying that a Raspberry Pi is genuine. For example, a code printed directly on the device. If you enter that code on the website, you can see exactly which model it belongs to and how often it has been used. That way, you could determine whether a Raspberry Pi is new or used and whether it has been modified. I saw that once at a thermal paste manufacturer. They used it to solve the problem of counterfeit thermal paste.You just have to get people to actually use the system and tell anyone else who asks for support but doesn't have an original copy to GTFO.This would allow them to avoid the increased support burden without limiting people's ability to upgrade their Raspberry Pi.
They frame this as fraud-prevention but I don't really buy it.
I think of I this way. As a casual/hobbyist user, maybe I'm happy to accept a 10% risk of the device failing prematurely in exchange for saving $20 on a memory upgrade. Ok, so these unscrupulous sellers will never tell me that info up-front...but this defective-by-design system doesn't help me either. The sellers playing this shady game might not accept returns (one example of this already in the github issue linked from the main thread) and there's no sign anything's wrong until you receive and boot up. So the user is left holding the bag, with kneecapped memory capacity, and the same stability risks anyway!
If these DRAM-upgraders are really so clever, they'll find a way to bypass this (patched firmware or hacked serial in the CPU). At least, it'll work until the user is out of the 30-day period when they finally update the firmware.
Or if the CPU/mainboard can really detect this in a non-defeatable manner, why cut the usable memory? Just print something to dmesg or stderr when the user logs in. "This device's memory has been modified and may be unstable, warranty blah blah"
Apple locks down fingerprint/camera module swaps, but at least there's a plausible (afaik) security rationale for it. Seems like Raspberry just got tired of being undercut on price.
The real story here is that people are selling modded Pis as higher-memory versions. And this check actually caught them + is circumventable. The restriction is somewhat annoying to a small subset of enthusiasts but seems to catch genuine fraud.
It feels like there could be a middle ground here where the Pi would show some annoying error message at boot that the RAM doesn't match the CPU's anticipated spec.
Any sellers caught doing this would quickly be flagged by the community as unreliable; also purchase protection is pretty strong these days.
Reading through the EEPROM repo issue[1] makes me a little more sympathetic to the "feature", but I still don't like it at all.
It might help in the short term to stop dodgy resellers who mod the boards with cheap upgraded RAM to make a buck or two...
But the Pi itself is always taking one step forward, one step back in terms of firmware. I've been tinkering with some Qualcomm boards from Radxa and things are slightly better there, though you don't have anywhere near the ecosystem you get with Pi (or onboarding niceties).
I love how we think we own our hardware, but secretly we don't. Surprised though that even a Raspberry Pi now thinks this as well. They grew big _because_ of tinkering...
It would be much more user friendly if there was a way to check the serial of a particular PI and see it's original hardware config, manufacture date, and so on.
Raspberry Pi is going down the way iRobot did, succumbing to pressure from chinese players like radxa or orange pi.
Sad to see industry pioneers lose to players that copy them and undercut prices using cheap labor.
I think the bigger problem is that this was pushed as a firmware update, bricking previously working devices.
Remember when FTDI did this and suddenly medical equipment at some hospital stopped working?
They should make a form where you input the serial number and apply for a key, applying that key in config.txt disables the hardware verification checks on the bootrom.
So they applied this gate in the rpi EEPROM, which is closed source.
Alright, what would it take to make a completely open source firmware for the pi? I think there was some prior effort that got discontinued...
Why not just paint the original chips with some special ink that says VOID IF TAMPERED over the RAM chips and the board? Pretty much like other hardware manufacturers do with seals or stickers, wouldn't this solve the issue?
I like the Chromebook solution for tampering: allow people to go crazy but throw up a big scary warning on boot!
"We do it to protect you and ensure our quality"
Nah, they got big enough and now you think you can pull Apple tricks, good luck with that.
Raspberry Pis, save for RP silicon like rp2040, have always been locked down proprietary black boxes.
Rich of them to claim that "users will complain" when their platform is built for hobbiests and tinkerers. What a terrible stance.
I mean the Pi currently makes little to no sense to buy anyway. So locking down parts barely moves the needle. The value proposition is unbelievably bad at spending upwards of $423 CAD for a 16GB Raspberry Pi. Which is literally just the bare board.
The 2026 financial report is apparently this Thursday on Sept 24. So will see how this played out for them. Their stock is all over the place anyway with low float.
I think this is really dumb and stupid. Not that I don't agree with their reasoning - it's actually very reasonable - but that I have a lot of distaste from "saving the customer from themselves" type antifeatures.
Practically speaking, though, this antifeature doesn't affect most people, because I don't think most people are willing to take a heat gun to a microsoldered part just to save a few bucks on RAM. If the RAM was socketed somehow - which at this scale and performance tier is not possible - then I'd probably be way angrier.
That's quite disappointing. Whose hardware is it anyways? Yours, when you pay for it.
> We therefore remove the commercial incentive by locking devices to their original RAM size.
Get fucked. Clearly you want to make all the profit and stop people from upgrading their devices.
We can't normalize this cash-grab user-hostile late stage capitalism bullshit.
Their argument assumes users are idiots - we had swappable components for years, and while there were cheating actors, market has invented things like brands and trust, you know? You don't have to buy models from the sellers you don't trust, nobody forces you to.
[dead]
> raspberrypi locked and limited conversation to collaborators
Way to read the room, jfc. Hardbricking hardware is not the answer, see FTDI.
I get why they are doing that... but I think there should be a way for users to flash a separate firmware, maybe tied to the hardware key similar to the old h264/mpeg license, that allows the usage of a modified board.
PS: Arrest Sam Altman who caused all of this mess in the first place with his market manipulation stunt.
Raspberry Pi been scummy for a while e.g. locking codecs behind extra payments. It seems they've gone full mask off at last.
Got here via Jeff Geerling coverage[0].
There were sellers buying the smallest RAM units, changing the RAM chips, and then reselling them as the higher RAM SKUs. The sellers who do this don’t always care to use good RAM chips and may even use QA reject parts. When the unit doesn’t work correctly, the anger and RMA requests are directed back at the Raspberry Pi foundation.