OpenRGB can 'kill' your DDR5 rather easily.

[Kafka]

I learned about this rather unfortunately, I swapped my NVMe boot drive to my new AM5 system and everything was going rather smoothly until a reboot or two after and now my RAM is gone. I forgot to remove OpenRGB from the startup list, both sticks gone now. I'll keep it brief and to the point with evidence.

PSA:PLEASE AVOID USING THIS PROGRAM UNLESS YOU KNOW HOW TO AND CAN FIX SPD CORRUPTION AND HOPEFULLY IT ENDS AT JUST THAT.

"You must never run your system with SPD Write Disable option set to False in BIOS as a norm, even if your RAM kit requires it to control RGB, – it is better to be left without fancy lighting effects than without working memory."


At first I thought it's bad luck or something but nope.
I'd suggest to look at these:

[Meta-Thread] DDR5 Detection / SPD Corruption / Boot Issues (#4934) · Issues · Adam Honse / OpenRGB · GitLab

Description of Bug This is a meta-thread to collate all the similar DDR5 issues that are currently occurring with OpenRGB, the...

gitlab.com

https://github.com/ubihazard/ddr5-spd-recovery - "poorly coded applications made by imbeciles, e.g. OpenRGB, which fail to access devices on SMBus in a proper manner." "Avoiding the use of third-party RGB and fan control software will help minimize the risk. OpenRGB is one notable offender which has been proven to corrupt DDR5 SPD EEPROM contents many times, even if RAM didn’t have RGB leds. I.e. you might have used it to configure the lighting color of your RGB fans, without touching anything related to memory, and still end up with corrupted RAM SPD. This software is a public hazard and must be avoided." hold up his writing is this fire?

https://www.reddit.com/r/buildapc/comments/1fgbtel/has_openrgb_been_borking_my_ram_all_this_time

https://www.reddit.com/r/OpenRGB/comments/10tsvwe/openrgb_killed_my_rams_leds

I knew using this was a bad idea due to its kernel level access but shit happens.

What's more in the past iCUE and G.Skill's RGB Program also used to kill sticks as well. What a professional RGB-vomit industry out here.

 

[freeagent]

Crazy.. that is all I use lol. FWIW the Lexar RGB software is made by G.Skill haha.

 

[tpa-pr]

Interesting, I use OpenRGB for my mainboard and keyboard. Thankfully my RAM has no LEDs so it's not even detected by the software.

Good to know though, thanks for the warning!

 

[Hervon]

Fair warning. I use OpenRGB on my Gigabyte X870E Elite without issue. I may turn off RAM RGB though for better safety.

 

[freeagent]

Maybe it happens when you try to write to the stick? Like to permanently set a color or turn it off..?

Literal news to me.. been using it since 2020 when I moved to AM4..

 

[c2DDragon]

I'm not using this software but you might want to use the BIOS option : "SPD Write Disable" TRUE for safety, which is in Advanced DRAM Configuration on MSI boards and set to AUTO by default.

 

[Solaris17]

I knew using this was a bad idea due to its kernel level access but shit happens.

That’s the wildest shit I ever heard because they even have kernel access in windows lol.

All of the manufacturers do. Even icue. Looks like this software might have an issue talking on the bus but but unless you know how to write those communication protocols this sentence tells me you don’t quite know how it works either.

 

[Kafka]

I'm not using this software but you might want to use the BIOS option : "SPD Write Disable" TRUE for safety, which is in Advanced DRAM Configuration on MSI boards and set to AUTO by default.

https://gitlab.com/CalcProgrammer1/OpenRGB/-/issues/4671 Interestingly enough this may or may not be enough... who knows.

That’s the wildest shit I ever heard because they even have kernel access in windows lol.

All of the manufacturers do. Even icue. Looks like this software might have an issue talking on the bus but but unless you know how to write those communication protocols this sentence tells me you don’t quite know how it works either.

All I know for sure is that it uses the infamous WinRing0x64.sys from way back, SignalRGB for instance stopped using but now uses its own driver. And I trust what that guy wrote "poorly coded applications made by imbeciles, e.g. OpenRGB, which fail to access devices on SMBus in a proper manner." giving such (horrible) programs that much access to your system is a mistake is all I mean to say. After all iCue5 was bricking RAM as well. Whether or not it's needed for the SMBus access is different to the message I want to convey. But in the end the result is the same my RAM is bricked, wouldn't have achieved it without using a terrible program with deep system access.

 

[tpa-pr]

I'm not using this software but you might want to use the BIOS option : "SPD Write Disable" TRUE for safety, which is in Advanced DRAM Configuration on MSI boards and set to AUTO by default.

The strangeness continues, I can't find any settings about SPD write protection on my X670E AORUS Master. Maybe it only appears if I have LED RAM...?

 

[Shrek]

Windows allows kernel access to external programs?

 

[Solaris17]

Windows allows kernel access to external programs?

Where have you been for the last 30+ years?

How do you even think hardware com works?

 

[freeagent]

Well.. those posts are ancient. How relevant are they in 2025, asking out of ignorance

 

[c2DDragon]

The strangeness continues, I can't find any settings about SPD write protection on my X670E AORUS Master. Maybe it only appears if I have LED RAM...?

If you don't use one of the few last bios, maybe it was added into it. I can't tell for sure, it's not written on the website for your BIOS' board updates.

There is this news talking about them (Gigabyte) adding it into their BIOS : https://www.techpowerup.com/forums/...gcc-software-fixes-spd-related-issues.316682/

 

[Kafka]

Well.. those posts are ancient. How relevant are they in 2025, asking out of ignorance

As far as I know the program hasn't changed much since its release. It's been happening consistently year after and year and I only learned of this yesterday cause it happened to me with the latest build. The one time I didn't install Windows from scratch at that, you forget something like this and it's over, kinda crazy. It handled the drivers well tho will admit.

Could OpenRGB possibly "kill" ram? (#616) · Issues · Adam Honse / OpenRGB · GitLab

Trying to narrow down a fault, and I'm honestly at a bit of a loss as to what's going on. I had 2x8gb of HyperX Predator 3200MHz RGB...

gitlab.com

The first time it was reported and it has the main dev in it, "OpenRGB shoudn't be able to cause this on its own because all SMBus activity is performed on a single thread. Multiple apps writing at once could cause this if one app were to set the address register and the other command a write." lol lmao even.

 

[Shrek]

If you don't use one of the few last bios, maybe it was added into it. I can't tell for sure, it's not written on the website for your BIOS' board updates.

There is this news talking about them (Gigabyte) adding it into their BIOS : https://www.techpowerup.com/forums/...gcc-software-fixes-spd-related-issues.316682/

Interesting, so the SPD wasn't locked in some cases.

 

[Solaris17]

If you don't use one of the few last bios, maybe it was added into it. I can't tell for sure, it's not written on the website for your BIOS' board updates.

There is this news talking about them (Gigabyte) adding it into their BIOS : https://www.techpowerup.com/forums/...gcc-software-fixes-spd-related-issues.316682/

I am curious why SPD is generally enabled by default anyway. I think gskill keeps the RGB values on some kits on the ram eeprom so I guess that makes sense; I wonder if this is an accidental write to certain memory regions of the eeprom.

The bus is temperamental and devices are dumb at this level; that’s why having multiple sensor monitoring utilities open can fuck up readings.

 

[Ruru]

Crazy.. that is all I use lol. FWIW the Lexar RGB software is made by G.Skill haha.

I freeaget posts on a thread, it may be interesting.

And no, I'm not fucking around with ya :'D

 

[Redwoodz]

If your EXPO spd gets corrupted, you could fry your cpu. Finally an insight into fried Ryzen chips.

 

[freeagent]

I am curious why SPD is generally enabled by default anyway. I think gskill keeps the RGB values on some kits on the ram eeprom; I wonder if this is an accidental write to certain memory regions of the eeprom.

I think Steve was playing with a guy from AMD in one of his later videos, and the AMD guy said some manufacturers have left SPD open, and he was able to reprogram them. Also said SPD should not be unlocked.. As for storing to EEPROM, my Royals allow me to do that, but neither of my Lexar's do.

 

[ir_cow]

If your EXPO spd gets corrupted, you could fry your cpu. Finally an insight into fried Ryzen chips.

Never heard of this. Explain please.

I think Steve was playing with a guy from AMD in one of his later videos, and the AMD guy said some manufacturers have left SPD open, and he was able to reprogram them. Also said SPD should not be unlocked.. As for storing to EEPROM, my Royals allow me to do that, but neither of my Lexar's do.

When I flash DDR5, the only one ever fully locked is Corsair. The rest let me change all the XMP/EXPO profiles. However there is a warning on the rest saying parts of EEPROM could not be flashed and the JEDEC isn't able to be overwirtten.

 

[freeagent]

Never heard of this. Explain please.


When I flash DDR5, the only one ever fully locked is Corsair. The rest let me change all the XMP/EXPO profiles. However there is a warning on the rest saying parts of EEPROM could not be flashed and the JEDEC isn't able to be overwirtten.

I almost shit my pants when they flashed the Lexar stick.

So now I am all paranoid

Rainbows, black fans, or Armory Crate

 

[Redwoodz]

Never heard of this. Explain please.

Just logic. I was referring to the supposed outbreak of 3Dcache chips dying on certain boards. If your EXPO profile gets corrupted and you have EXPO enabled, there is no telling what could happen, especially if applying too much voltage to the cpu.

 

[freeagent]

It was just scribbling the SPD, wouldn't work because the numbers were all scrambled. I don't see how that would effect low level power settings?

 

[ir_cow]

Just logic. I was referring to the supposed outbreak of 3Dcache chips dying on certain boards. If your EXPO profile gets corrupted and you have EXPO enabled, there is no telling what could happen, especially if applying too much voltage to the cpu.

XMP/EXPO doesn't control the CPU, so corrupted SPD isn't going to kill anything but the memory. The only scenario where I can see it causing CPU damage is for motherboards that match the CPU VDDIO to DRAM voltage when XMP/EXPO is enabled. However, the max you can do is 2.1V for DRAM settings (generally), but the highest VDDIO goes to is 1.5V. The ram would die long before before the CPU.

 

[Kafka]

I almost shit my pants when they flashed the Lexar stick.

So now I am all paranoid

Rainbows, black fans, or Armory Crate

ASUS Aura: Campaigns

Personalize your gaming rig with Aura RGB lighting and synchronize effects with a vast portfolio of Aura Sync capable gear.

www.asus.com

Note: The standalone version of Aura Sync Utility will stop receiving updates soon. The latest version is 1.07.84_V2

Aura without Armory Crate? As painful as even that may be? I'd say SignalRGB but it's a bit convoluted and annoying. I'll probably use Aura Sync myself when I get new ones from the RMA.
I saw a post somewhere which I cannot find now it was showing how much gibberish it wrote to the SPD eeprom in a hex dump.
Kinda like this post https://forum.corsair.com/forums/topic/184617-corsair-icue-5-corrupting-spd-data-on-ram-modules/ but going more over it, if I had to presume my RAM probably got fried.

Maybe we should get those 3k USD en masse flashers for DDR5 then we'll always be safe... right?