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

#1
the54thvoid
Good they're addressing it but they can't blame the memory speed. The GTX 1070 runs at the 'unprecedented' 8Gbps.

Then again, not a huge issue as only really affected much older mobos?
Posted on Reply
#2
ZeppMan217
So, they're gonna reduce the voltage? Lower board's max power draw? What kind of a solution can be cooked up with a driver update?
Posted on Reply
#3
chinmi
Hahaha.... Dead on arrival... People have more and more reason to wait and buy the 1060 now...
Posted on Reply
#4
arnoo1
Good but this means that the card wil probally slower
Posted on Reply
#5
WhyCry
You don't always see Wizzard making a news post.
Posted on Reply
#6
Fragment
They could just limit the power draw / undervolt the card.
There are already reports on reddit that this actually doesn't affect performance in measurable ways.
Because some cards seem to come with 1.3V vcore out of the box. Which is way over that what's needed to reach boost clock.

https://www.reddit.com/r/Amd/comments/4qupw4/super_psa_all_rx480_owners_please_attempt_to/

Maybe there is even a more elegant solution, lets wait 'til Tuesday.

WhyCry said:
You don't always see Wizzard making a news post.
Haha. Or you making any kind of post... :P
Posted on Reply
#7
$ReaPeR$
chinmi said:
Hahaha.... Dead on arrival... People have more and more reason to wait and buy the 1060 now...
yeah mate.. whatever you say..

on topic: this doesnt seem to be such a major problem, and i love how people have blown it way out of proportion, what do you think happens when you oc a card geniuses?
Posted on Reply
#8
Fragment
$ReaPeR$ said:
yeah mate.. whatever you say..

on topic: this doesnt seem to be such a major problem, and i love how people have blown it way out of proportion, what do you think happens when you oc a card geniuses?
Probably that the extra power comes from all the Wi-Fi signals that are in the air nowadays hue hue.
Posted on Reply
#9
Recus
RIP AMD

http://seekingalpha.com/article/3985508-amds-polaris-revealed-overhyped-disaster

That's why AMD community is the worst.

https://www.reddit.com/r/Amd/comments/4qfwd4/rx480_fails_pcie_specification/?sort=new
moderators here are now flairing this as 'rumor/FUD' despite plenty of testing having been done to corroborate this.
Spreading BS that GTX 960 burning motherboards too.
http://www.pcper.com/reviews/Graphics-Cards/Power-Consumption-Concerns-Radeon-RX-480/Evaluating-ASUS-GTX-960-Strix
Posted on Reply
#10
RejZoR
I was expecting AMD to be using input power monitoring and control just like NVIDIA is using on Maxwell 2 (where you can independently control how much each power input can draw power). In theory, I could limit my GTX 980 to exclusively draw power from 6pin+8pin, but I'd have to adjust the total power limit to about 225W then (oppose to current 250W with some reserve from the PCIe which is still limited to 66W).

Bottom line, clocks may dip tiny bit due to slight power restriction imposed by this fix (only on cards that have these problems), but I'm not expecting any noticeable real world difference. Especially if we consider the fact that AMD, as time progresses, optimizes drivers and gains performance opposed to NVIDIA which seems to drop it over time...

It was a bit of a cock up, but not much bigger than NVIDIA's cocked up fan profile on GTX 1080...
Posted on Reply
#11
Ferrum Master
What's the VRM IC? Anybody knows?

They can adjust power limits for each, there should be no problems. Later AIB cards will be okay out of the box.
Posted on Reply
#12
$ReaPeR$
RejZoR said:
I was expecting AMD to be using input power monitoring and control just like NVIDIA is using on Maxwell 2 (where you can independently control how much each power input can draw power). In theory, I could limit my GTX 980 to exclusively draw power from 6pin+8pin, but I'd have to adjust the total power limit to about 225W then (oppose to current 250W with some reserve from the PCIe which is still limited to 66W).

Bottom line, clocks may dip tiny bit due to slight power restriction imposed by this fix (only on cards that have these problems), but I'm not expecting any noticeable real world difference. Especially if we consider the fact that AMD, as time progresses, optimizes drivers and gains performance opposed to NVIDIA which seems to drop it over time...

It was a bit of a cock up, but not much bigger than NVIDIA's cocked up fan profile on GTX 1080...
is it really that serious of a problem though? because people seem to be panicking about this..
Posted on Reply
#13
KainXS
Its good to see they addressed it, lets see how they fix it, will they underclock and undervolt or just reduce the power limit and then limit its maximum maybe.
Posted on Reply
#14
Fiery
FinalWire / AIDA64 Developer
Ferrum Master said:
What's the VRM IC? Anybody knows?
IRF IR3567B
Posted on Reply
#15
W1zzard
WhyCry said:
You don't always see Wizzard making a news post.
bta seems asleep, I saw the email after crawling out of bed with my gf, so I thought "let's get this out to the people"
Posted on Reply
#16
BiggieShady
AMD
Recently, we identified select scenarios where the tuning of some RX 480 boards was not optimal.
Translation: Recently, we were forced to admit someone at our ranks fucked up the complete voltage table for Polaris ... poor bastard thought we were trying to hit 1.3 GHz on a 28 nm chip :laugh:

seriously, average sample (72% ASIC quality) gets to 1.3 GHz at 1.15 V ... and new boards are already undervolted in the bios version they ship with
Posted on Reply
#17
Ferrum Master
Fiery said:
IRF IR3567B
Same as always... lately, they can manage power delivery for each phase. No downclock or lower voltages, considering there is already 50% power reserve. What's the fuss really?
Posted on Reply
#18
$ReaPeR$
Fragment said:
Probably that the extra power comes from all the Wi-Fi signals that are in the air nowadays hue hue.
LOL yeah!!! hehehe
Posted on Reply
#19
arbiter
Fragment said:
There are already reports on reddit that this actually doesn't affect performance in measurable ways.
Because some cards seem to come with 1.3V vcore out of the box. Which is way over that what's needed to reach boost clock.
Likely its a safe range they can run every chip at to get to xxxx mhz some can do it at well below that voltage yes but there are a few that need it to get there.

$ReaPeR$ said:
is it really that serious of a problem though? because people seem to be panicking about this..
Um when your computer just goes poof, off and not turn on again cause the board is fried cause the power draw. Would you panic a bit?
MB power circuits more so a problem on cheaper boards can do up to the spec but not really anything past that. Its like an extension cord you pull well above the power it can handle, it gets hot and well you know the end.
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.
aka they gonna enforce clock limits so it don't boost to where it does now.
Posted on Reply
#20
Fluffmeister
$ReaPeR$ said:
is it really that serious of a problem though? because people seem to be panicking about this..
People here go into meltdown when a fan spins up when it doesn't need to, so this is borderline biblical disaster.
Posted on Reply
#21
DeathtoGnomes
chinmi said:
Hahaha.... Dead on arrival... People have more and more reason to wait and buy the 1060 now...
...there's one in every crowd...

ontopic: Sounds like a small batch of cards got the wrong burn.
Posted on Reply
#22
$ReaPeR$
arbiter said:
Likely its a safe range they can run every chip at to get to xxxx mhz some can do it at well below that voltage yes but there are a few that need it to get there.


Um when your computer just goes poof, off and not turn on again cause the board is fried cause the power draw. Would you panic a bit?
MB power circuits more so a problem on cheaper boards can do up to the spec but not really anything past that. Its like an extension cord you pull well above the power it can handle, it gets hot and well you know the end.


aka they gonna enforce clock limits so it don't boost to where it does now.
yeah i know how it works, i have seen it in practice. my point is since we havent seen many mobos die from this problem why all the panic?

Fluffmeister said:
People here go into meltdown when a fan spins up when it doesn't need to, so this borderline biblical disaster.
hahahahahahahahaha the end is nigh repent!!!!!

back on topic though, i think we need more info on this problem, it is an issue, but since the effects of it seem to be almost none existent i really dont understand the panic. most people here have good to really good hardware and most of us here are used to ocing, and when you oc something it will go beyond the specs anyway so..
Posted on Reply
#23
GC_PaNzerFIN
As usual, technical PR team tries to **** with customers and shove out load of propaganda. Thanks to beyond required effort from multiple hardware sites, this issue finally may be fixed. When you talk down these issues, you should know it is you, customer, who is getting fooled here.
Posted on Reply
#24
Ferrum Master
GC_PaNzerFIN said:
As usual, technical PR team tries to **** with customers and shove out load of propaganda. Thanks to beyond required effort from multiple hardware sites, this issue finally may be fixed. When you talk down these issues, you should know it is you, customer, who is getting fooled here.
Lately every new tech product is plagued with some sort of issues due to haste... that's the electronics biz... it has been like that always really.
Posted on Reply
#25
arbiter
$ReaPeR$ said:
yeah i know how it works, i have seen it in practice. my point is since we havent seen many mobos die from this problem why all the panic?
Card has been out for what 3-4 days at most. if MB's died already from it image what rate could be in say 3 months, or 6 months, etc. That is the real issue.
Posted on Reply
Add your own comment