Saturday, July 2nd 2016

Official Statement from AMD on the PCI-Express Overcurrent Issue

AMD sent us this statement in response to growing concern among our readers that the Radeon RX 480 graphics card violates PCI-Express power specification, by overdrawing power from its single 6-pin PCIe power connector and the PCI-Express slot. Combined, the total power budged of the card should be 150W, however, it was found to draw well over that power limit.

AMD has had out-of-spec power designs in the past with the Radeon R9 295X2, for example, but that card is targeted at buyers with reasonably good PSUs. The RX 480's target audience could face troubles powering the card. Below is AMD's statement on the matter. The company stated that it's working on a driver update that could cap the power at 150W. It will be interesting to see how that power-limit affects performance.
"As you know, we continuously tune our GPUs in order to maximize their performance within their given power envelopes and the speed of the memory interface, which in this case is an unprecedented 8 Gbps for GDDR5. Recently, we identified select scenarios where the tuning of some RX 480 boards was not optimal. Fortunately, we can adjust the GPU's tuning via software in order to resolve this issue. We are already testing a driver that implements a fix, and we will provide an update to the community on our progress on Tuesday (July 5, 2016)."
Add your own comment

358 Comments on Official Statement from AMD on the PCI-Express Overcurrent Issue

#176
BiggieShady
RejZoR
Jesus, people think this will just fry the system after 3 gaming sessions.
No, just a pcie slot ... you can move 480 to a different slot and have another 3 gaming sessions on the same motherboard ;) https://community.amd.com/thread/202410
Posted on Reply
#177
GLD
I want to see AMD and/or the card manufactures release a bios update to fix the power draw problem. A fix in a driver release wont be sufficient. I want a RX 480, when they get this problem sorted. :cool:
Posted on Reply
#178
medi01
chinmi
Hahaha.... Dead on arrival... People have more and more reason to wait and buy the 1060 now...
Nah, +15% of perf for +25%+ of price, no thanks, f*ck nVidia tax.

It's a good reason to wait for 480 AIBs though.

Tatty_One
I would think it will possibly be only a matter of hours before something is done assuming that a driver update would alleviate the issue, AMD know better than most that anything negative associated with a new product launch if not dealt with promptly could have long term implications.
They said it should be out by 7th, did they change their stance?

GhostRyder
Still foolish not to just put an 8 pin as the default... If they wanted to do this to the 4gb and limit the board spec to 150 with a 6 pin then it's fine, but they should have at least with the 8gb given an 8 pin reference.

This was just a foolish design choice.
I'm pretty sure they boosted 480 at the last minute.
Raja expected Pascal to come at least several months later.

If not 1070/1080, RX 480 would look pretty impressive even at 980Mhz (what Sony PS4k is allegedly using)
Posted on Reply
#179
cdawall
where the hell are my stars
sith'ari
Mate, i haven't paid 200€ on a high-end PSU, and 400€ for also a high-end UPS, only to let an RX 480 to endanger my system!!:mad:
But , hey, with your money you can do what you want.;)
Unless you paid 20€ the motherboard you are blowing this shit way out of proportion.
Posted on Reply
#180
theoneandonlymrk
GLD
I want to see AMD and/or the card manufactures release a bios update to fix the power draw problem. A fix in a driver release wont be sufficient. I want a RX 480, when they get this problem sorted. :cool:
I've got my second one on the way from Amazon for 220 notes and I noted, its there best seller right now ,despite this immense issue ,my mobo has an onboard Molex just for extra power to the pciex for quadfire etc(Cleary asus expected odd loading via pciex) ,my boards a decent but old one and ill have 2 480s running flat out 24/7 folding ill let you all know if my pc explodes or anything.
You need to realise hardware and software are intertwined on many platforms,intel and nvidia inclusive so a driver Will be fine ty very much.
Posted on Reply
#181
RejZoR
GLD
I want to see AMD and/or the card manufactures release a bios update to fix the power draw problem. A fix in a driver release wont be sufficient. I want a RX 480, when they get this problem sorted. :cool:
And how is BIOS any different than driver? Just because it's burned into a hardware, that doesn't make it any better. And if you've seen Wattman, it's basically controlling hardware directly. So, driver or BIOS, it doesn't really matter at this point. When RX480 boots to desktop, it's not out of specs because it's not under load. But once it's in Windows, does it even matter at this point since drivers are loaded already? And seeing how Windows 10 insists on installing latest Radeon drivers no matter what, users will basically be forced to have the latest version of the driver.
Posted on Reply
#182
cdawall
where the hell are my stars
and everyone does realize we are talking about 7w over spec on the mainboard right?

Posted on Reply
#183
sith'ari
cdawall
and everyone does realize we are talking about 7w over spec on the mainboard right?
I suppose you haven't seen this video yet:

If you are bored to watch it all, go to 20:50 for a quick look.


EDIT: Or this one that says about frequent power shutdowns with older mobos:
Posted on Reply
#184
GhostRyder
medi01
Nah, +15% of perf for +25%+ of price, no thanks, f*ck nVidia tax.

It's a good reason to wait for 480 AIBs though.



They said it should be out by 7th, did they change their stance?


I'm pretty sure they boosted 480 at the last minute.
Raja expected Pascal to come at least several months later.

If not 1070/1080, RX 480 would look pretty impressive even at 980Mhz (what Sony PS4k is allegedly using)
Even if that is the case, its still foolish not to do it.

cdawall
and everyone does realize we are talking about 7w over spec on the mainboard right?


Yea, even if you buy the bottom of the barrel motherboard it still should be ok. I think some are overreacting to this though its a problem that should not exist anyways. I have loaded up a motherboard before to ridiculous levels and not managed to harm it, so I doubt most people (Unless you own like an AM1 motherboard maybe?) are not going to be able to harm it especially with an RX480.
Posted on Reply
#185
Ungari
Where is the outrage for Nvidia products that used PCIE lanes as the sole power source and routinely spiked up in excess of 200 watts?
No boos and hisses for the GTX 950 SE, or 750Ti (let's not mention 960 Strix with it's "only a 6-pin connector") since those cards have much higher power spikes than that of the RX 480?
Why did no one inquire if those cards were PCI-SIG certified?
These cards are typically mounted in lower cost basic mainboards as low tier card users aren't likely to spend on high performance mobos; why are there no danger warnings, or reports of mainboards getting blown open from these cards?
Posted on Reply
#186
Ravenas
Hmm.. So they lower the voltage through a software update.

This card is targeted towards a mainstream market that could probably care less, and are ignorant of the fact. We're talking about a 1080p $199/249 performance/per dollar beast... Not a 290x or 390x or Fury.

Mainstream consumer sees 8.9 review score, for $199/249 it's a pretty obvious choice.
Posted on Reply
#187
Dippyskoodlez
RejZoR
What "non official" driver users? There is just one driver. The official one.
To include Linux users, as well as people like me where an official, super up to date driver may not be in use:
Posted on Reply
#188
sith'ari
Ungari
Where is the outrage for Nvidia products that used PCIE lanes as the sole power source and routinely spiked up in excess of 200 watts?
No boos and hisses for the GTX 950 SE, or 750Ti (let's not mention 960 Strix) since those cards have much higher power spikes than that of the RX 480?
Why did no one inquire if those cards were PCI-SIG certified?
These cards are typically mounted in lower cost basic mainboards as low tier card users aren't likely to spend on high performance mobos; why are there no danger warnings, or reports of mainboards getting blown open from these cards?
-Apparently you have skipped a lot of posts from this thread. I would urge you to read #151. There is nothing wrong with the GTX 960's power output.
-Plus, on the video i put at #184, go at 27:10 and watch their comments for the GTX 960 power output as well.
Posted on Reply
#189
GLD
RejZoR
And how is BIOS any different than driver? Just because it's burned into a hardware, that doesn't make it any better. And if you've seen Wattman, it's basically controlling hardware directly. So, driver or BIOS, it doesn't really matter at this point. When RX480 boots to desktop, it's not out of specs because it's not under load. But once it's in Windows, does it even matter at this point since drivers are loaded already? And seeing how Windows 10 insists on installing latest Radeon drivers no matter what, users will basically be forced to have the latest version of the driver.
The way my epeen sees it, a driver can "manage" a problem and a bios update can "fix" a problem. :D
Posted on Reply
#190
cdawall
where the hell are my stars
sith'ari
I suppose you haven't seen this video yet:

If you are bored to watch it all, go to 20:50 for a quick look.
They literally just say it may be an issue. Buy something cheap slap a high watt card in and what expect everything to be perfect? This isn't an issue that will cause issues to most people, it causes an issue with someone who purchased a Dell in 2011 and expected to toss this card in and have it work perfectly...

I imagine like has been said AMD bumped card voltage to increase yield size for a known popular card and it exceeded spec because of it.


sith'ari
EDIT: Or this one that says about frequent power shutdowns with older mobos:

Do you have an older motherboard?
Posted on Reply
#191
sith'ari
cdawall
Do you have an older motherboard?
See my signature and tell me yourself!! ;)
Posted on Reply
#192
cdawall
where the hell are my stars
sith'ari
See my signature and tell me yourself!! ;)
You probably shouldn't buy this card then. Holy hell your entire issue was just avoided and just FYI even when this is fixed I wouldn't recommend you buy this card because you will still probably blow the board.
Posted on Reply
#193
Dippyskoodlez
RejZoR
And how is BIOS any different than driver? Just because it's burned into a hardware, that doesn't make it any better. And if you've seen Wattman, it's basically controlling hardware directly. So, driver or BIOS, it doesn't really matter at this point.
Bios is a hard cap, a driver requires separate installation and requires it being functional and there are many situations where a driver may not be in control of the ship, but a bios would forcefully override any issues.

Similar to the SMC on a Macbook, or the fan control on the eVGA ACX where it turns off at low temperatures. Having a card able to manage itself is key to ensuring reliability. eVGA GPUs are preferred for my configuration currently because they primarily use the pcie plugs instead of the slot for power.
Posted on Reply
#194
zAAm
cdawall
You probably shouldn't buy this card then. Holy hell your entire issue was just avoided and just FYI even when this is fixed I wouldn't recommend you buy this card because you will still probably blow the board.
You do realise people can have an opinion of something that doesn't directly affect them right? ;)

In general though (not directed at cdawall), just because people care about this issue does not automatically mean we're all nvidia fanboys and want the RX480 to fail and to burn down AMD's offices etc etc. It's a new precedent because the issue was detected with THIS card. It's not like everyone was reporting it for all the cards and now suddenly we care. Since the issue has gained some publicity however, it should ensure that manufacturers ensure future cards stay within spec, which won't be a bad thing for consumers. I for one hope the fix is easy and sales aren't affected to any great extent. Competition is always good for the consumer ;)
Posted on Reply
#195
Ungari
sith'ari
-There is nothing wrong with the GTX 960's power output.
See the 3:20 mark of this video concerning the GTX 960 Strix PCIE Power Consumption:

Posted on Reply
#196
Dippyskoodlez
zAAm
You do realise people can have an opinion of something that doesn't directly affect them right? ;)
'don't buy this' is also a bad answer: Many people will cram this into a board which it 'shouldn't be in' if that was the fix too.

Ungari
See the 3:20 mark of this video concerning the GTX 960 Strix PCIE Power Consumption:


http://www.pcper.com/reviews/Graphics-Cards/Power-Consumption-Concerns-Radeon-RX-480/Evaluating-ASUS-GTX-960-Strix

960 does not present the same issues the rx480 did.
Posted on Reply
#197
RejZoR
Dippyskoodlez
Bios is a hard cap, a driver requires separate installation and requires it being functional and there are many situations where a driver may not be in control of the ship, but a bios would forcefully override any issues.

Similar to the SMC on a Macbook, or the fan control on the eVGA ACX where it turns off at low temperatures. Having a card able to manage itself is key to ensuring reliability. eVGA GPUs are preferred for my configuration currently because they primarily use the pcie plugs instead of the slot for power.
Eh, rubbish.

- not in Windows = No 3D acceleration = no load on GPU = FINE
- Basic Display Driver = No 3D acceleration = basic load on GPU, well within specs = FINE
- Full fixed driver = 3D Acceleration = full load with 150W actual limit or redistributed load to 6pin = FINE

Please do tell me what scenario is not covered. And if you think BIOS has it's mind of its own, this is AMD we're talking about. They made the GPU's and they made the BIOS. Don't you think they know what overrides what?
Posted on Reply
#198
Dippyskoodlez
RejZoR
Please do tell me what scenario is not covered. And if you think BIOS has it's mind of its own, this is AMD we're talking about. They made the GPU's and they made the BIOS. Don't you think they know what overrides what?
Did you not see my screenshot? My GTX 970 runs "346.03".

There are also Linux users using non-proprietary blobs that provide basic acceleration. If a system fires up any compute, you could run into problems, especially considering it's likely to beat any game power draw.

I don't have any evga 'gaming' software' to run, I don't have GPU-z to tweak things, I have strictly the on card BIOS to control cooling of the ACX2.0 fans. If it were driver controlled thinking it was a blower, the 'off' feature at low load/temperatures wouldn't function.

An RX480 trying to pull >75w through my pcie slot would actually start causing stability problems if it doesn't actually cause any physical mishaps from the overcurrent, as is very well documented in eGPU setups already.

Edit: On a side note, this is why the 'bullshit' reviewer bios fiasco is also very important to me. I can't necessarily take advantage of an overclock if it's not preset.
Posted on Reply
#199
Ungari
Dippyskoodlez
'don't buy this' is also a bad answer: Many people will cram this into a board which it 'shouldn't be in' if that was the fix too.



http://www.pcper.com/reviews/Graphics-Cards/Power-Consumption-Concerns-Radeon-RX-480/Evaluating-ASUS-GTX-960-Strix

960 does not present the same issues the rx480 did.
Correct, the 960 Strix is much worse in terms of Power Consumption on the PCIE lane.
If the 960 Strix does not concern you, than the RX 480 is just fine and dandy.
Posted on Reply
#200
bpgt64
I can't even get mine to post atm. Now I am on a 420W(seasonic) psu on an mini itx board..so I didn't think it would be an issue, but I'll likely have to buy a new PSU or return the card...
Posted on Reply
Add your own comment