Steam Deck

Steam Deck

DivideTheZero 6 mrt 2022 om 12:32
5
6
2
4
Freezing in every game, occasionally causing total system lockup
Hello, I've been having this problem ever since I got the Deck. Every game at some point, will lockup and I'll be forced to either quit the game, or occasionally force shut down the system. Possibly hardware issue?

The freezing occurs at seemingly random times, but *usually* within the first 15 minutes of launching a game. The game will freeze, the screen will go black for a sec, then when it comes back the game will be completely unresponsive. Audio still plays, and I can usually use the overlay and exit the game. Sometimes the overlay will also be unresponsive, and I have to force shut down the system. (Note: In these scenarios, I can still hear the overlay, I just can't see it)

This freeze occurs in *every single game I have tested*. H:ZD, Control, Hitman 3, No Man's Sky, Half Life 2, ME: Legendary, etc.

It appears like the GPU is crashing or resetting. Graphics settings and intensity appear to have zero impact on the issue. Happens just as often on low presets as on high presets.

So far I have tried re-imaging the deck, and opted into the beta branch. No change.
Laatst bewerkt door DivideTheZero; 6 mrt 2022 om 12:55
Origineel geplaatst door KickAssDave:
Since we're all on the same hardware, I'd like to come in here and also help others not waste any time looking in to this. Especially if you are very technically minded or work in IT and find it tempting to try and "resolve" this.

This is caused by a faulty power module. The device is a Texus Instruments ina230.
If you check your logs during boot, you'll see various errors related to i2cxx etc...
Basically this power module is faulty, fails to initialise during boot and causes issues doing what it's supposed to do... hence... random crashes.

You can try all these fixes, but they aren't really fixes. The fact it doesn't happen in every game is likely more chance than anything, but given it USUALLY crops up in more demanding games (more power... ) makes sense too.

The resulting "amdgpu" errors are just the GPU timing out because of the power delivery issues.

I spent 3 weeks trying EVERYTHING YOU CAN IMAGINE... to fix this. It is not worth your time. I have tried every Linux distro, latest mesa drivers, amdvlk drivers, latest mainline kernels (6.1rc5), zen kernels, different kernel parameters, disabling/enabling different dpm options etc. etc. etc. etc.

****TL:DR: SAVE YOURSELF THE PAIN AND JUST SEND THE DAMN THING BACK****
< >
511-525 van 742 reacties weergegeven
Origineel geplaatst door Kosmo零:
Origineel geplaatst door SilentUK:
Also agree with the point that no games crashed, and then when one does, they all start to. I completed Spiderman and Beyond Two Souls on the deck with no crashes, but once Diablo 4 started to crash, all my other games followed soon after.

I'm wondering how many people on here have replaced the Internal SSD with a new, bigger one and if that is causing any issues?
That could be because of the initial thermal damage to PCB. For example, you start to play some new game with heavy graphics that heats APU a lot. There is some problem place on PCB with bad solder under component. The heat from all your previous games wasn't enough to damage bad solder, but now it heated way more, so solder crack. Since that moment heat from any other games easily make crack in same place.

P.S. freezes happened to me even before I changed SSD.

But, the thing goes:

- I reach the point where a couple of games freezes and crashes no matter what at 20-30 minutes of Gameplay, for example, last of us and spideman yesterday.

- Then i do a factory reset, and then the same games in the same steam deck, works well for hours before they start to fail again.

So if it was something thermal related wich causes permanent damage on the hard side, a Factory reset or a re imagine won't solve anything nor temporally nor definitivelly.....that is why i think is something that changes the SO behivor in some way, like a trigger

I have even thinking about something in our steam accounts? something in the chache or who knows?

But what i can't understand is why there are some people that seems to solver this with another unit, and others that have this issue permanetly... It doesn't make sense.



Origineel geplaatst door SilentUK:
Origineel geplaatst door Alecslee:
Problem is we know for a fact there are a lot of people out there playing the most demanding games even in dock mode, full power, without capping the fps or graphics etc so at a High temps, and they aren't experiencing those freezings not even once, so if it was a thermal problem the net would be plenty of people complaining constantly and its not the case.


In fact in my case, i have more than 300 games installed in various microsd, and amongs them i have the most demanding games like elden ring, gow, guardians of the galaxy, and others, and i have only experienced this kind of freezings in about 10-12 games of those 300, and i cap all the games to run at 15-17 total watts at max, wich means 70-80° at max too, and i've played more than 100h at elden ring without a single freeze, and the same goes to guardians of the galaxy, re village or other demanding games, while for some reason, spiderman remastered is the game in wich i have more freezings by far, an the others have been: last of us (5-6 freezings) , sacboy (1-2 freezings), judgment (1-2 freezings), witcher 3 (1 freezing in 50h), Morales (1 freezing), sifu (2-3 freezings), pacer (freezes after 2-3 races) and a couple of games more, while with the other 290, i haven't experienced them, wich is so weird...maybe those games have something in common, a library maybe, and that is why once 1 is corrupted, the others follow it?

What i have clear is that the process is like:

1.- All works fine without freezings or crashes.
2.- Then a game freezes the way this thread relates.
3.- Then something happens in the SO or bios or i don't even know, and then you start experiencing them in another games that prevously were working right.
4.- Then you did a re-imagine or Factory reset and the cicle stars again from point 1.

So to me its obvious is something Software related that someway writes something where it shoulnd't have to, and then all goes crazy.

The worst part is that this is , BY FAR, the most spread bug on the deck since day one, valve have PLENTY of RMA units With those freezes, and more than half a year later they haven't told anything about, nor found a solution, and i even doubt they are trying to solve it.....

Agree with this. Diablo 4 is a good example, when that came out a lot of people are playing that for hours at a time with no freezes at all. I cant run that for more than 30 minutes without a black screen crash. Same thing with Horizon: Zero Dawn.

Also agree with the point that no games crashed, and then when one does, they all start to. I completed Spiderman and Beyond Two Souls on the deck with no crashes, but once Diablo 4 started to crash, all my other games followed soon after.

I'm wondering how many people on here have replaced the Internal SSD with a new, bigger one and if that is causing any issues?

In my case, i never opened my first deck, and the freezings happened just as when i tried another ssd with my second deck, just to test if that could be the problem, a faultys vanila ssd perhaps, but nah, the feezings appears no matter the SSD you are using, that can't be the trigger., because i haven't opened the third deck, and this new deck from my cousin, its a 64gb model without any mod too, and it has those freezings too, so 3 vanila ssd decks and one with a 1tb ssd, all with the same freezings.

I'm wonder if any of us have experienced those freezings playing only in the ssd? or we all use microsd alonsige the ssd? could be that some microsd corrupts the so? what microsd are you using?
Laatst bewerkt door Alecslee; 16 jul 2023 om 6:20
Origineel geplaatst door Alecslee:
- Then i do a factory reset, and then the same games in the same steam deck, works well for hours before they start to fail again.
That could be because your Deck's PCB is cooling while you reimaging it (that's easy for APU and it doesn't heat much).
Once some game freeze - the PCB reach the state of failure because of temp, that's why trying to run games immediatly after freeze leads to even more frequent freezes (PCB is already preheated, and then receive heat again).

Someone can confirm or deny thermal damage version by simply testing their Decks without backplate. Or even better - after changing thermal paste and without backplate. It should take more time until game freeze. BUT DO NOT run it without PCB shield plate (I did run my for test, but didn't run any games while that). Some PWM contollers are hot even without games. That shield has some thermal pads to receive heat from them. I don't want to think what will happen to them without that shield when you start some game :P

P.S. I played from SSD only when installed Windows 10. Same freeze, but Deck instantly reboots after it.
Laatst bewerkt door Kosmo零; 16 jul 2023 om 7:42
I have a replacement steam deck that should be arriving soon. I got the dispatch notification a couple of days ago. To test this further, i've not wiped my SD or SSD card from the old steam deck, so I will put them in the new replacement steam deck and see if i still get crashing in the games that were crashing before.
journalctl -t kernel|grep -i 'Killed process'

out of curiosity. Can someone post the output of this command?
Laatst bewerkt door deaddoof; 22 jul 2023 om 20:20
Origineel geplaatst door deaddoof:
journalctl -t kernel|grep -i 'Killed process'
out of curiosity. Can someone post the output of this command?
I entered this command, but it didn't produce any output. No any errors too. Terminal just switched to next line where I can enter another command.

P.S. Has anyone needed recently to provide Deck's serial number photo on back plate when asking for RMA? Tech support suddenly asked for it at a step where they usually give RMA label to download. Non of RMA instructions says they ask for a serial number photo (I already specified it in initial message).
Laatst bewerkt door Kosmo零; 24 jul 2023 om 2:21
Yes I also had to provide a serial number from the back of the steam deck
Origineel geplaatst door SilentUK:
Yes I also had to provide a serial number from the back of the steam deck
It's unclear from your answer what you prodived (photo or you retype serial number with keyboard). They ask me to MAKE PHOTO of that serial number.
Laatst bewerkt door Kosmo零; 24 jul 2023 om 3:00
I just had to type it, not provide a photo.
New steam deck arrived today. I put in my old SSD Drive and SD Card and loaded up a game and... No crashes... Yet, at least. I will note that the internals of this steam deck look different to my old steam deck, the metal cover inside was black instead of silver with no warranty screw sticker and the foil that goes around the SSD drive was much looser. Not sure if this means its a refurb or if they have changed the internals of the deck slightly. I will update this later tonight once i've tested thoroughly but so far its looking positive.
Having the same problem. Mine is a recent model, bought it on sale and received it last week. At first I thought it was a microsd issue, but then I saw all these discussions and searched for the ring gfx_0.0.0 timeout, which confirmed that I pretty much have a faulty deck.

I tried pretty much everything. I have done the APU reset, I formatted my deck and played games on the eMMC while the PSU was plugged in, I get the same error everytime.

This is pretty disappointing, especially when valve still does not acknowledge the problem and there are countries that do not support RMA. I am on my 15 day trial period and I’m not sure if I trust on getting a new one and getting the same error again. This does not seem like a reliable product unfortunately. I might just ask for a refund.
I would say it is worth doing the RMA, i haven't had a crash since getting my new one 3 days ago and I've been running it pretty extensively.
Origineel geplaatst door Lugia:
Having the same problem. Mine is a recent model, bought it on sale and received it last week. At first I thought it was a microsd issue, but then I saw all these discussions and searched for the ring gfx_0.0.0 timeout, which confirmed that I pretty much have a faulty deck.

I tried pretty much everything. I have done the APU reset, I formatted my deck and played games on the eMMC while the PSU was plugged in, I get the same error everytime.

This is pretty disappointing, especially when valve still does not acknowledge the problem and there are countries that do not support RMA. I am on my 15 day trial period and I’m not sure if I trust on getting a new one and getting the same error again. This does not seem like a reliable product unfortunately. I might just ask for a refund.
So after wasting about 3 hours of my time following steam support standard procedure (battery storage mode, disabling the advanced fan feature and re image of the deck), nothing has worked and it’s still crashing. I kinda don’t understand why I need to do all of this when the issue is clear and it has been for a long time. If this happens to be a lottery then I might get another faulty deck and I’ll lose my refund rights, I’ll be thinking about this.
Does Valve give any invoice for replacement? It's that document where Deck's price, transaction number and purchase date is written. I decided to RMA mine, and since I am in unsupported country, I need invoice for customs to receive it back here. Please, see your account purchase history if Valve gave you anything like that for replacement.
Origineel geplaatst door Kosmo零:
Does Valve give any invoice for replacement? It's that document where Deck's price, transaction number and purchase date is written. I decided to RMA mine, and since I am in unsupported country, I need invoice for customs to receive it back here. Please, see your account purchase history if Valve gave you anything like that for replacement.
I'm also in unsopperted country. Valve only gave me UPS label that I need to print and put it on the box, so it can be sent freely inside US. But every other aspect you should handle by yourself they told me.
< >
511-525 van 742 reacties weergegeven
Per pagina: 1530 50

Geplaatst op: 6 mrt 2022 om 12:32
Aantal berichten: 742