RX 480 Could fry your MOBO?

Just wanted to point out the motherboards will not burst into flames because the card uses more power from the PCIE slot. The problem here is cheaper power supplies sometimes use thinner wires which results in the wires becoming hotter, draw enough power though these wires can result in it becoming so hot it melts the plastic 24pin connector.

So as long as you don't skimp on your PSU you should be fine whether your using an expensive motherboard or a cheaper one (new or older).

In short @bimbleuk's system won't bust into flames. (Hopefully - Things do go horribly wrong sometimes)

1 Like

https://www.amazon.com/MSI-FM2-A75MA-E35-Mainboard-mATX/dp/B009LT8K26/ref=sr_1_2?ie=UTF8&qid=1467378989&sr=8-2&keywords=fm2-a75ma-e35

oops

This should law =)

1 Like

pci-e + 6 pin is rated for 150w, which is what RX480 will draw at max during full load with the stock power setting.

pci-e x 16 alone is 75w, but RX 480 has a 6 pin as well which puts it at 150w.

so why don't we stop with the Hyperbole on both sides.

Does the RX480 draw more PCI power than it should yes in some cards on some motherboards,

Do Motherboard makers feel concerned about this, from what PCPer has posted no they do not think the nonOC draw is an issue on modern motherboards.

Is there indication that on older boards you may have a problem yes, but while PCIE 3.0 devices are data backwards compatible with PCIE 2.0 that does not mean they are power compatible.

If you are planning to OC this card you will need a beefy board in terms of VRM and even then this reference card was not meant for OC.

finally this card will not fry your Mother board it will cause it to shutdown to protect itself.

if this makes anyone want to wait for AIB cards or just to pass on this that is up to you.

1) The RX 480 has passed PCIe compliance testing with PCI-SIG. This is not just our internal testing. I think that should be made very clear. Obviously there are a few GPUs exhibiting anomalous behavior, and we've been in touch with these reviewers for a few days to better understand their test configurations to see how this could be possible.

https://www.reddit.com/r/pcmasterrace/comments/4qg4p7/rx_480_fails_pcie_specification/

/END

2 Likes

Nope, it is 75W. I did bite my way through the official PDF for that. 8-pin is 150.

PCI-E is 75W, 6-pin is 75W. 150W.

2 Likes

PCI-E is rated for 75w by itself, 6 pin is rated for 75w by itself.
75w+75w = 150w

8 pin is rated for 150w by itself.

1 Like

Right. I misread that.

Still, the card draws more.

No it doesn't draw more, unless you fiddle with the power settings.
Even if it draws more, it shouldn't be a problem.

I run a custom bios on my watercooled nano, changes the core power setting from 150w to 250w, card only has a single 8 pin connector.
I haven't ran into any problems, I could be drawing 300w if I wanted and still it would be fine.

People having issues with the RX 480 is simply due to crap motherboards, RX 480 runs within specs by a good margin, typical load during games is much less than 150w.

YES YOU ARE RIGHT

but if as PCPer has had discussions with motherboard makers who have said they are not concerned about the power draw issues when not OCed then you should only be concerned about people doing their $300 and below builds using early cheap pcie 2.0 boards and low end components. Secondarily there are already reports that the 16.6.2 driver update from AMD has reduced the severity of the problem and will probably continue to as they figure out the specific issue.

Yes, and that is a good thing. As I mentioned I hoped for exactly that. An official statement would further calm some tits. I just don't understand the complete tunnel vision of some people here that are saying "standards don't matter, AMD is da best ever and its your cheap mobo's fault if your house burns down".

In the end we are all on the same side here.

I think the issue is not that the people are saying "Standards don't matter" it is more that they are saying the problem is not happening with every card or even the majority and that the standards body and Motherboard makers have all said not a big deal may negatively affect older or ultra budget motherboards in terms of shut offs but should not kill a motherboard unless you are running the card out of spec.

The point of having standards is so this does not happen.

then take it up with PCIE-SIG who has said that this is a non issue. you know the Standards group in charge of this standard.

1 Like

As someone who buys and owns really crappy Motherboards I don't have a problem with this. My MSI A68h-e33 is a piece of crap, it's not even the V2 version. It came free with the 7850k. While such boards can be problematic I am glad they are available. Usually people who have crappy MB's don't buy the latest card the day it comes out. Even with the hiccups AMD came out with a great card at a great price.
A system shutdown is not exactly frying.

How much power, 1.21 jigowatts?

Webster says the soft G (J sound) is an acceptable pronunciation. Take that grammarians!

1 Like

Sadly no as it's built to mine and will most likely never see any games unless I install something this weekend. Mining mostly works the computational and memory controller hence why I can run with a lower power target and a little lower clock speed.

For mining right now I'm seeing 100 MH/s for Ethereum but hopefully there's another 10% or so more to tweak out of it.