AMD RX 480 Causing Faults For PC Gamers – PCIe Slots Have Died On High-End Boards

[Misanthrope]

3 minutes ago, LAwLz said:

I put on my robe and skepticism hat when I opened this thread. Surprisingly enough some of these seem pretty legit.

 

roquen22 used an ASRock 970 Extreme4 motherboard. That's a pretty old (probably 4-5 years by now) motherboard and it was not exactly high end back then. On top of that his RX 480 was overclocked on the core (although not overvolted and stock VRAM).

 

The person on overclock.net (the one with the Rampage IV Extreme motherboard) was using 3-way crossfire.

 

 

I would not be surprised if the majority of reports are fake, but it is worrying that some of them seem legitimate. I was expecting reports like these in 6+ months. Not 2 days.

Nobody reported the 3.5 limits before it was called out I think it's a case of telling a kid not to do something being the best way to get him to do just that.

[HalGameGuru]

The audio issue I wonder at. IF it was caused by ripple it should be rectified by simply rebooting with the 480 downclocked a bit. 

 

The dead motherboard is an iffier prospect, because most people are saying the traces go direct from 24pin to PCIe slot it may have damaged the rail if the PSU couldn't handle the ripple or bad voltage swings, which would explain the other slots being dead. The other PCIe slots should work fine, if the first slot is what was damaged, unless the board has them disabled or limited on lanes. I wonder if clearing CMOS would help him get it back up and running on the other PCIe slots.

[Stefan1024]

14 minutes ago, djdwosk97 said:

AMD likely did testing on a high end motherboard, not some POS 970 board. Plus AMD may or may not have overclocked the card as high. (Assuming the report is true)

My problem is: i can-t see the point of failure.

 

PSU -> 24 pin -> copper power plane -> slot

 

The 24 pin connector can handle way more current, the power plane at least 30 A (>350 watts). The weak spot is the pcie slot (but you should see a visible damage here and the other slots must not be affected).

A cheap PSU migth be failing when to much power is drawn by the MoBo. But the MoBo itself should be fine (unless failing the PSU kocked the total system out).

[79wjd]

5 minutes ago, Stefan1024 said:

My problem is: i can-t see the point of failure.

 

PSU -> 24 pin -> copper power plane -> slot

 

The 24 pin connector can handle way more current, the power plane at least 30 A (>350 watts). The weak spot is the pcie slot (but you should see a visible damage here and the other slots must not be affected).

A cheap PSU migth be failing when to much power is drawn by the MoBo. But the MoBo itself should be fine (unless failing the PSU kocked the total system out).

I don't know enough about electrical engineering or motherboard design to say, but there was a source where someone (I believe it was someone who works for Asus/some other manufacturer) claimed sustained loads above 90w (or something similar) would likely cause damage. 

[Stefan1024]

2 minutes ago, djdwosk97 said:

I don't know enough about electrical engineering or motherboard design to say, but there was a source where someone (I believe it was someone who works for Asus/some other manufacturer) claimed sustained loads above 90w (or something similar) would likely cause damage. 

I looked up the spec. for a pcie connector by myself and it is rated at only 1.1A (we should only talk about current as this is the important metric) and going over this limit cab heat up the connector to much. But to destroy the connector the plastic has to literaly melt and you should be able the see the damage. That said, a hot PCB frm the GPU inserted to the socked makes the heat build up worse.

[sof006]

I uploaded this video to YouTube with this exact issue with my RX 480.

 

I am trying to get Scan to accept a return for the product as it is clearly faulty.

 

 

[Sauron]

A-hem can we take a second to notice that 2-3 people having an issue is not a very high number? And that wccftech is the one reporting on it?

 

Of course amd should not have exceeded specifications (if it's true, it was not 100% sure iirc) but let's take this with a grain of salt - it's not the first card to do something like that and as @LAwLz pointed out most of these can be caused by something else.

[HalGameGuru]

that 1.1 Amp reference is probably per pin. And there are multiple pins providing 12v. If memory serves someone said 5, which would jive perfectly with the rated 5.5 amp 66 watt 12v spec. but, once again that is the spec, not necessarily what it can handle overall. I don't know how the traces and leads compare to wiring but 24 gauge Cat5 wires are expected to handle 30 watts per the spec. So even if the PCIe can only handle around half that it's still slightly more than 1.1Amp per pin. more like 1.25 per pin. And that's not necessarily indicative of what it can actually handle. Without testing, and likely on a mobo by mobo basis, we cannot really say what the true ceiling is for the slot until we hit it.

[sof006]

9 minutes ago, Sauron said:

A-hem can we take a second to notice that 2-3 people having an issue is not a very high number? And that wccftech is the one reporting on it?

 

Of course amd should not have exceeded specifications (if it's true, it was not 100% sure iirc) but let's take this with a grain of salt - it's not the first card to do something like that and as @LAwLz pointed out most of these can be caused by something else.

I am having similar issues with mine described above but mine has not destroyed my PCIE slot. Mainly because mine displays the artifacts on screen after about 2-3 minutes of gameplay. Not a good start AMD. Not a good start.

[Megahurt]

You could always try to lower the power limit and try to undervolt to regain back the performance lost, but it's not a real solution and wouldn't work on all cards, cause if it did, all of them would have just shipped this way.

[Rune]

Ooooh boy. Not looking forward to seeing the returns on these.

[elago]

Did they really not test this before launch? How can this things happen nowadays, honestly.

[xAcid9]

I'm surprised nobody blaming that CX500 as usual. 

6 minutes ago, sof006 said:

I am having similar issues with mine described above but mine has not destroyed my PCIE slot. Mainly because mine displays the artifacts on screen after about 2-3 minutes of gameplay. Not a good start AMD. Not a good start.

Looks like a problematic VRAM.

[LAwLz]

33 minutes ago, HalGameGuru said:

The audio issue I wonder at. IF it was caused by ripple it should be rectified by simply rebooting with the 480 downclocked a bit. 

Not if a component in (or along the path to) the audio chipset got fried.

Rebooting doesn't repair damaged electronics.

 

 

21 minutes ago, djdwosk97 said:

I don't know enough about electrical engineering or motherboard design to say, but there was a source where someone (I believe it was someone who works for Asus/some other manufacturer) claimed sustained loads above 90w (or something similar) would likely cause damage. 

That source was from the PCPer article. Ryan asked one of the people in the motherboard industry (did not mention any names or brands) which said that motherboards are designed to handle spikes, but sustained loads of 95 watts will most likely cause damage over time.

 

 

18 minutes ago, Stefan1024 said:

But to destroy the connector the plastic has to literaly melt and you should be able the see the damage.

Why do think the plastic has to melt for it to be damaged?

Motherboards are built in layers. I don't know where the power traces for PCIe power goes, but if they are in one of the layers below the top one then it would pretty much be impossible to see any damage.

[Stefan1024]

8 minutes ago, LAwLz said:

Why do think the plastic has to melt for it to be damaged?

Motherboards are built in layers. I don't know where the power traces for PCIe power goes, but if they are in one of the layers below the top one then it would pretty much be impossible to see any damage.

I'm 99.9999% sure they have a complete power plane for 12V and GND, everything else would cause massive EMI problems. And such a power plane can easely handle >30A all day long. You are not gong to blow a power trace with a load of a 480 or any other GPU.

 

The weakest spot is the pcie slot or the PSU.

[Stefan1024]

23 minutes ago, HalGameGuru said:

that 1.1 Amp reference is probably per pin. And there are multiple pins providing 12v. If memory serves someone said 5, which would jive perfectly with the rated 5.5 amp 66 watt 12v spec. but, once again that is the spec, not necessarily what it can handle overall. I don't know how the traces and leads compare to wiring but 24 gauge Cat5 wires are expected to handle 30 watts per the spec. So even if the PCIe can only handle around half that it's still slightly more than 1.1Amp per pin. more like 1.25 per pin. And that's not necessarily indicative of what it can actually handle. Without testing, and likely on a mobo by mobo basis, we cannot really say what the true ceiling is for the slot until we hit it.

Yes it is per pin, sorry forget to say that.

And you can not directly compare wires and connectors. The contact resistance at the connector can be significant higher than the wire.

[-BirdiE-]

Yeah. Unfortunately at this point it's hard to tell how many of these reports are real, and how many are just internet trolls.

[Stefan1024]

2 minutes ago, -BirdiE- said:

Yeah. Unfortunately at this point it's hard to tell how many of these reports are real, and how many are just internet trolls.

Sadly the internet is full of trolls and AMD haters that like to write fake reports like this.

 

And there are people with already dead / damaged (before they ever used an RX480) MoBo who hope they get a free replacement.

[Sauron]

51 minutes ago, sof006 said:

I am having similar issues with mine described above but mine has not destroyed my PCIE slot. Mainly because mine displays the artifacts on screen after about 2-3 minutes of gameplay. Not a good start AMD. Not a good start.

Artifacts could be due to a number of reasons, including drivers. It's not necessarily a pcie slot problem.

[goaway]

16 minutes ago, -BirdiE- said:

Yeah. Unfortunately at this point it's hard to tell how many of these reports are real, and how many are just internet trolls.

To me it really does not matter wich claims are true or not. If AMD had just gone with an 8 pin in the first place, many if not all of the genuine complaints would not be here today, wich in turn would certainly get rid of all the fake ones. 

 

Simply put: AMD brought all this bad PR on themselves.

 

Edit: Oops double post

 

Edited July 1, 2016 by larsi239

[-BirdiE-]

1 minute ago, larsi239 said:

To me it really does not matter wich claims are true or not. If AMD had just gone with an 8 pin in the first place, many if not all of the genuine complaints would not be here today, wich in turn would certainly get rid of all the fake claims. 

 

Simply put: AMD brought all this bad PR on themselves.

 

Yeah. It's a shame, but they really did. I was really excited for them to put out a really impressive product this time around...

 

There were reports prior to release that Polaris had turned out to be "underwhelming", and I think this is just that showing through... They had to pull something like this to try and deliver on some of the promises they made.

 

I mean, it's still a decent budget card for the time being, but nowhere near as impressive as they made it out to be, especially in power efficiency.

[goaway]

2 minutes ago, -BirdiE- said:

Yeah. It's a shame, but they really did. I was really excited for them to put out a really impressive product this time around...

 

There were reports prior to release that Polaris had turned out to be "underwhelming", and I think this is just that showing through... They had to pull something like this to try and deliver on some of the promises they made.

 

I mean, it's still a decent budget card for the time being, but nowhere near as impressive as they made it out to be, especially in power efficiency.

Yeah, the whole point of polaris was always touted as being the best fps per watt/per dollar budget GPU. But looking at the power draw, if I'm not mistaken, the card is as power hungry as a 1070, but performs like a 970...

 

If nvidia gets their pricing right with the 1060, AMD is really gonna have a struggle selling 480s in my mind - wich let's be real, is really not what us consumers need right now with NV already enjoying a very comfortable lead on AMD.

[Misanthrope]

I really hope it is people trolling and taking advantage, otherwise this could be a Class action suit that would be the final nail on the AMD coffin.

[-BirdiE-]

10 minutes ago, larsi239 said:

Yeah, the whole point of polaris was always touted as being the best fps per watt/per dollar budget GPU. But looking at the power draw, if I'm not mistaken, the card is as power hungry as a 1070, but performs like a 970...

 

If nvidia gets their pricing right with the 1060, AMD is really gonna have a struggle selling 480s in my mind - wich let's be real, is really not what us consumers need right now with NV already enjoying a very comfortable lead on AMD.

 Yeah, things aren't looking great, but the 480 does have a couple things going for it:

1) It costs nearly half of a 1070

2) It performs much better in DX12 titles

 

Now it's just a question of where the 1060 fits into all of this, and how popular DX12 titles become in the near future. But overall, things aren't looking good.

[Master Disaster]

Polaris, the first step in saving AMD... 

 

Well that worked out well for them didn't it. I'm not an AMD hater, never have been however AMD have been a joke for years. Bulldozer and Piledriver were aweful, Steamroller failed, Jaguar did OK in the SoC market but failed outside of it, every GPU since the HD2x series has ran hotter than hades and we keep hearing about mythical power inside GCN that's waiting to be unlocked yet time and time again they fail. 

 

Honestly I'm expecting Zen to release late next year and perform worse than Haswell while running hotter than the sun. It just seems to be AMDs way these days. I really hope I'm wrong but I just don't see them pulling anything amazing out of the bag.