Following User Outcry, AMD Reinstates Memory Encryption In Consumer CPUs
- Reference: 0184036890
- News link: https://it.slashdot.org/story/26/06/22/2039221/following-user-outcry-amd-reinstates-memory-encryption-in-consumer-cpus
- Source link:
> Following the revelation, social media was deluged by comments from AMD consumers decrying the move. They noted that AMD's quiet removal of TSME after supporting it for so long seemed underhanded. The move came solely as a result of firmware changes made in a recent update. With no physical changes required to silicon, continued support was largely, if not purely, a matter of will rather than a necessity required by changes to hardware. The critics called on AMD to reverse the move.
>
> Over the weekend, AMD said it planned to do just that in a firmware update scheduled for release next month. More often than not, the chipmaker refers to TSME as Memory Guard. "Regarding certain non-PRO Ryzen 9000-series desktop processors, a BIOS option to enable Memory Guard was previously available but was removed in a recent update," AMD said in an email. "Based on valuable community feedback, we will reinstate this option in an upcoming BIOS release in July."
>
> The company has yet to explain why it removed the protection. Critics speculate that AMD dropped it in an attempt to steer customers toward more costly CPUs. It's possible, though, that there were less nefarious reasons, such as the difficulty of continued support as chip designs changed. Another possibility is that AMD made the move for performance reasons. Encrypting and decrypting data in memory creates latency. Slowdowns are the enemy of gamers, one of the more popular customer segments using the 9000-line of Ryzen processors. Since many gamers already voluntarily disabled TSME and had little need for it in the first place, AMD may not have considered the change of much consequence.
[1] https://hardware.slashdot.org/story/26/06/15/200234/users-cry-foul-after-amd-stripped-memory-crypto-from-its-consumer-cpus
[2] https://arstechnica.com/security/2026/06/following-user-outcry-amd-reinstates-memory-encryption-in-consumer-cpus/
Never good to remove a feature (Score:2)
I think that in this age of AI data theft, perhaps it's wise to leave any kind of safeguard in place.
FBI (Score:2)
You'll understand if you think about it
Re: FBI (Score:2)
Exactly this. They want to use their magic tool they plug into usb when raiding a server to keep the memory intact then they use those piggyback power supplies to keep the machine on while they transport it back for investigation.
This only bothers their ILLEGAL activities and hardly anyone elseâ(TM)s. Glad they have to deal with it now
A trial balloon? (Score:2)
It seems unlikely that they really thought this would go unnoticed. Maybe they just wanted to see how quick and loud the pushback would be.
Corporations are forever pulling shit like this to see how much they can get away with. "Gee - if we can get away with this, maybe we can get away with something bigger in the way of monetizing something that we never charged for in the past".
Corporations are slimy antisocial motherfuckers and are ALWAYS trying to pick up a little extra coin. All those little extras add
Re: (Score:2)
> Maybe they just wanted to see how quick and loud the pushback would be.
Maybe they were counting on the pushback. The Feds step up and "suggest" that they turn the encryption off. Or things will go badly for them. They do, but it leaks out. Now the public is pissed. And it's getting near election time.
Bitlocker (Score:2)
Nightmare Eclipse showed us Bitlocker is a joke. It's not remotely real encryption and easily breakable .. on Win11/2025 server, NOT Win 10. This wasn't an exploit. It was a backdoor. Meanwhile Veracrypt needed a public backlash to get their dev signing keys reinstated so people could get their updated kernel drivers on Windows (and remember, TrueCrypt its predecessor mysteriously disappeared in 2012 with the former author telling people to use BitLocker instead!)
Now we have this. The answer should be ob
Re: (Score:3)
Is this some sort of parody of bad AC takes? If not, could have fooled me