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

#251
AsRock
TPU addict
ixi
Nice humour you got there mate.
Mate ?, aint that like friend ?. So you saying i was being rasist when i wasn't but i am a mate HAHA.

Anyways enough this is not what the forums for, please be more constructive and just PM me were i can just ignore you for taking out of your butt without effecting anyone else.
Posted on Reply
#252
Assimilator
jigar2speed
Thought to share it with you guys...


"Turns out pretty much every card pulls more than 75W from the slot" Well, that's a blatant lie as numerous sources with actual power measurement equipment have already determined. But I guess people would rather choose to believe an arbitrary YouTuber's tweet over multiple-pages articles of hard facts and numbers.

john_
I wonder if GTX 960 Strix was brought as a paradigm from those defending Nvidia. It's a card with an extra power connector and TDP much lower than 150W. Whatever spikes it produces the average consumption from the pcie bus will always look under the limit.
But tell someone about an overclocked GTX 950 without a 6 pin connector and ignores you
Because it's irrelevant. Let me explain this to you simply:

GTX 950 running at stock, i.e. how 100% of users will use it: adheres to the PCIe spec
GTX 950 overclocked, i.e. how a small % of users will use it: may violate the PCIe spec

RX 480 running at stock, i.e. how 100% of users will use it: violates the PCIe spec
RX 480 overclocked, i.e. how a small % of users will use it: violates the PCIe spec, probably even more

Simple numbers say that since far fewer people overclock GTX 950 than run RX 480 at stock, far fewer people will encounter issues with PCIe slot draw. Not to mention that overclocking voids your warranty anyway, so only you are responsible if your PC catches fire while you're overclocking a GTX 950. But if you're running an RX 480 at stock and it causes your PC to catch on fire, the only one to blame is the manufacturer... i.e. AMD.

Oh, and we already know that R9 295 X2 plays fast and loose with the PCIe power spec - I called AMD out on that too, BTW - but that's far less of a problem because there are so few 295s and the majority of people running them will have overspecced systems anyway.

What it boils down to is simply that if AMD hadn't been cheap f**ks and tried to shave 2 cents off the BOM by using a 6-pin connector instead of an 8-pin, they wouldn't be having this problem. That's an absolutely indefensible case of cutting corners. And personally that's why I'm so upset, because AMD has, once again, ruined what could've been a great product launch with their own incompetence. Like I said in the review thread, they never learn.

ArdWar
A bad motherboard that skimp on using proper power plane and ground plane, using skinny traces instead, is another story. It could literally burn.
That's where most of the concern comes from, because the low cost of the RX 480 means it's often likely to be paired with a cheap motherboard. Think of internet gaming cafes in Asia that are going to be buying these cards by the truckload - how high quality and well ventilated do you think those systems will be?
Posted on Reply
#253
$ReaPeR$
arbiter
High power spikes that is normal when a Dc to DC switch turns on, like a light bulb that turns on draw's a lot of power to turn on quick then drops down. Problem that could be from all this, people want to build super cheap 550$ gaming machine. Not gonna be a good thing if machine keeps shutting it self down in middle of game play. Most people probably wouldn't haven't the trouble shooting to figure out the gpu is drawing to much power from the board and causing it.

Well it is a 2 way street, the same people whined and complained about the gtx970 issue most them were not likely to ever buy one.



Problem with what you say there is spikes, if you look at all gpu's they spike to 100+ watts all the time its just nature of Dc to DC switch's. Its the over all avg draw over time that is where it gets to be the problem. drawing 225watts for matter of ms will do no damage but pulling 100watts constant for say 2-3 min can cause it as the heat is able to build up and melt something. If you go watch the video Pcper did on the issue that is one the things they cover.
thats not the same but anyway.. you cant correct with a driver the slow 512 MB of the 970..
Posted on Reply
#254
john_
Assimilator
Because it's irrelevant. Let me explain this to you simply:

GTX 950 running at stock, i.e. how 100% of users will use it: adheres to the PCIe spec
GTX 950 overclocked, i.e. how a small % of users will use it: may violate the PCIe spec
This is funny. You start about people believing an arbitrary tweet, then you do an arbitrary speculation because it suits you. Double standards?
RX 480 running at stock, i.e. how 100% of users will use it: violates the PCIe spec
RX 480 overclocked, i.e. how a small % of users will use it: violates the PCIe spec (probably even more)
In contrary to you, who will wait hell to freeze first before posting anything negative for Nvidia, I don't have a problem saying that AMD messed up here.
Simple numbers say that since far fewer people overclock GTX 950 than run RX 480 at stock, far fewer people will encounter issues with PCIe slot draw. Not to mention that overclocking voids your warranty anyway, so only you are responsible if your PC catches fire while you're overclocking a GTX 950. But if you're running an RX 480 at stock and it causes your PC to catch on fire, the only one to blame is the manufacturer... i.e. AMD.
You missed my point here. If GTX 950 goes as high as 85-90W of power usage through the pcie bus, shouldn't we have heard about people frying their motherboards? You can say that it doesn't happen and that's why we haven't heard anything. To be fair tech press likes to concentrate it's fire on AMD, so we probably will never learned what power GTX 950 with no 6pin sucks through the pcie bus under overclocking. Neither Tom's Hardware, neither PCPerspective, (neither TPU?) will come up with an article, especially if the card sucks more than it should.
Oh, and we already know that R9 295X2 plays fast and loose with the PCIe power spec - I called them out on that too, BTW - but that's far less of a problem because there are so few 295s and the majority of people running them will have overspecced systems anyway.
I bet you did. It's an AMD card.
What it boils down to is simply that if AMD hadn't been cheap f**ks and tried to shave 2 cents off the BOM by using a 6-pin connector instead of an 8-pin, they wouldn't be having this problem. That's an absolutely indefensible case of cutting corners. And personally that's why I'm so upset, because AMD has, once again, ruined what could've been a great product launch with their own incompetence. Like I said in the review thread, they never learn.
Thankfully Nvidia wasn't cheap f***ks, so they put all the features on the overpriced Founders Editions cards. Fans going bananas, power consumption at idle/multi monitor going bananas. Bananas. Bananas everywhere.

Funny how you get upset for products you will never buy from a company that you hate because it represents the competition to the company you love. In fact Nvidia fanboys are more upset than RX 480 owners themselves.
Posted on Reply
#255
jigar2speed
Assimilator
" arbitrary YouTuber
This Arbitrary youtuber is also known as Tech Reviewer and unlike you he has 654,848 subscribers to his youtube channel. You know i am just puting it forward the way they are coming up to internet, you can believe whatever you want. My 4 year old thinks my HD 7970 is an aeroplane and i don't correct her. Its not the correct time anyway. ;)
Posted on Reply
#256
silentbogo
john_
ou missed my point here. If GTX 950 goes as high as 85-90W of power usage through the pcie bus, shouldn't we have heard about people frying their motherboards? You can say that it doesn't happen and that's why we haven't heard anything. To be fair tech press likes to concentrate it's fire on AMD, so we probably will never learned what power GTX 950 with no 6pin sucks through the pcie bus under overclocking. Neither Tom's Hardware, neither PCPerspective, (neither TPU?) will come up with an article, especially if the card sucks more than it should.
Seems like you've missed the point: a GTX 950 with a 6-pin connector has a 90W rated max TDP. A GTX 950 without 6-pin PCIE power connector is limited to 75W, at least if we take specs from EVGA, ASUS, Palit and consider them true.
Additionally, here's a quote from low-power GTX950 review by @W1zzard :
During gaming, we see power consumption hover almost exactly around the 75 W mark, which is the maximum power draw from a PCI-Express slot. Since the card has no additional power connectors, this is the ideal result - close to 75 W but not significantly more.
https://www.techpowerup.com/reviews/ASUS/GTX_950/21.html

With maximum consumption of 76W and peak at 79W, which in the worst case scenario puts it at 5% over spec in short peaks, and 1.3% over spec overall (which is well within an error margin for such measurements).
Overclocking may increase the peak consumption values, but overall max won't change, because the card will throttle to stay within 75W limit.

It's not about who's defending who, it's about speculation versus facts and numbers. Even if Raja Koduri himself says that "RX480 is fine, trust me", I won't believe it because there are at least several equally reputable and less reputable people who clearly displayed the opposite by running an experiment and sharing their results with public.
Posted on Reply
#257
sith'ari
john_
...........................
Thankfully Nvidia wasn't cheap f***ks, so they put all the features on the overpriced Founders Editions cards. Fans going bananas, power consumption at idle/multi monitor going bananas. Bananas. Bananas everywhere.
Funny how you get upset for products you will never buy from a company that you hate because it represents the competition to the company you love. In fact Nvidia fanboys are more upset than RX 480 owners themselves.
You have to keep something in mind :
AMD's attempt to save money by using a 6-pin power connector instead of an 8-pin, possibly endangers my system.
What you say about NVidia,- (*although i totally disagree with you, because they made a fantastic GPU with +50% performance of a FuryX/980Ti, and you are still complaining!! )-, affects only the GPU itself, and it doesn't places in jeopardy my system.
You might enjoy taking risks about your motherboard's endurance and longevity, but personally, as i said before, i didn't pay near 600€ (*for top-notch PSU / UPS / surge protectors etc), only to let AMD's GPU to destroy my system from the inside !!:mad:
Posted on Reply
#258
rtwjunkie
PC Gaming Enthusiast
jigar2speed
This Arbitrary youtuber is also known as Tech Reviewer and unlike you he has 654,848 subscribers to his youtube channel.
Not arguing, disagreeing, or agreeing, but it needs to be pointed out that well over 90% of the people are Sheeple and will follow anything, including a guy who is entertaining on youtube. It doesn't actually say anything about his tech abilities.
Posted on Reply
#259
HD64G
sith'ari
You have to keep something in mind :
AMD's attempt to save money by using a 6-pin power connector instead of an 8-pin, possibly endangers my system.
What you say about NVidia,- (*although i totally disagree with you, because they made a fantastic GPU with +50% performance of a FuryX/980Ti, and you are still complaining!! )-, affects only the GPU itself, and it doesn't places in jeopardy my system.
You might enjoy taking risks about your motherboard's endurance and longevity, but personally, as i said before, i didn't pay near 600€ (*for top-notch PSU / UPS / surge protectors etc), only to let AMD's GPU to destroy my system from the inside !!:mad:
If you spend such amounts of money on PSU/UPS, etc and you don't habe a high quality MB, you are simply ignorant of PC and gaming tech. But since I am sure your MB is a good quality one, there is NOT A CHANCE an RX480 could damage it. Especially since AMD will fix that in 2-3 days with a new driver or anyone could fix it NOW by himself by lowering voltage a bit through Wattman.

http://semiaccurate.com/2016/07/01/investigating-thermal-throttling-undervolting-amds-rx-480/
Posted on Reply
#260
newtekie1
Semi-Retired Folder
jigar2speed
Thought to share it with you guys...


Again, the difference is that most don't pull more from the slot for long periods of time. They spike above 75w, but that is completely normal and acceptable. The problem with the RX 480 is that it pulls a lot more than 75w for a sustained period of time.

john_
I wonder if GTX 960 Strix was brought as a paradigm from those defending Nvidia. It's a card with an extra power connector and TDP much lower than 150W. Whatever spikes it produces the average consumption from the pcie bus will always look under the limit.
Actually the GTX 960 Strix was bought up because Tom's data made it look like it was doing the same thing as the RX 480. The GTX 960 Strix does spike above 75w. However, in reality it doesn't do the same thing as the RX 480 because as it has already been pointed out, the spikes don't matter. With DC to DC conversion there will always be those high spikes. What matters is the overall average. The GTX 960 Strix averages down around 30w from the PCI-E slot, and pulls everything else it needs from the external connector. The RX 480 average over 75w from the PCI-E slot.

john_
But tell someone about an overclocked GTX 950 without a 6 pin connector and ignores you, changes the subject, or wants you to believe that you can have 20% extra performance without consuming a single extra watt. Free performance.
What about an overclocked GTX 950 without a 6-pin? I directly addressed the issue of an overclocked GTX 950 without a 6-pin. I didn't change the subject, I explained to you exactly how it works several pages back. I'll explain it again. The power draw from the PCI-E slot still stays at 75w thanks to the power limiting built into the card. If the power limit is set to 75w, thanks to nVidia's GPU Boost, it is going to consume right around 75w. It doesn't matter what you overclock the card to, GPU Boost will keep the card at 75w. It is very effective at doing this. Manually raising the power limit as been a part of overclocking with nVidia for a couple generation now.
Posted on Reply
#261
john_
sith'ari
You have to keep something in mind :
AMD's attempt to save money by using a 6-pin power connector instead of an 8-pin, possibly endangers my system.
What you say about NVidia,- (*although i totally disagree with you, because they made a fantastic GPU with +50% performance of a FuryX/980Ti, and you are still complaining!! )-, affects only the GPU itself, and it doesn't places in jeopardy my system.
You might enjoy taking risks about your motherboard's endurance and longevity, but personally, as i said before, i didn't pay near 600€ (*for top-notch PSU / UPS / surge protectors etc), only to let AMD's GPU to destroy my system from the inside !!:mad:
No, it doesn't endangers your system with the 750Ti on it.

50% performance over GTX 980Ti? I think GTX 980Ti owners will want to say something here. The same can be said for Fury X owners in DirectX 12 games.

And don't worry. The UPS probably will survive.
Posted on Reply
#262
john_
silentbogo
Seems like you've missed the point: a GTX 950 with a 6-pin connector has a 90W rated max TDP. A GTX 950 without 6-pin PCIE power connector is limited to 75W, at least if we take specs from EVGA, ASUS, Palit and consider them true.
Additionally, here's a quote from low-power GTX950 review by @W1zzard :

https://www.techpowerup.com/reviews/ASUS/GTX_950/21.html

With maximum consumption of 76W and peak at 79W, which in the worst case scenario puts it at 5% over spec in short peaks, and 1.3% over spec overall (which is well within an error margin for such measurements).
Overclocking may increase the peak consumption values, but overall max won't change, because the card will throttle to stay within 75W limit.

It's not about who's defending who, it's about speculation versus facts and numbers. Even if Raja Koduri himself says that "RX480 is fine, trust me", I won't believe it because there are at least several equally reputable and less reputable people who clearly displayed the opposite by running an experiment and sharing their results with public.
I didn't missed the point. You missed all the other posts that I did about the subject and I am not going to repeat everything in detail. Just ask yourself this.
100% performance at 75W. At that review W1zzard overclocks the card and gets 20% extra performance. Not just higher clocks. 20% extra PERFORMANCE.
Now tell me. How much extra power consumption do you need for that extra 20%? 0 Watts? 5 Watts? 10 Watts? 20 Watts? And don't tell me about throttling. Throttling doesn't increase performance by 20%.

You see, there are many things that the press will not tell you. You just learned about the PCIe bus power draw because the RX480 is an AMD card. If it was an Nvidia card, you wouldn't have known about it.
Posted on Reply
#263
sith'ari
HD64G
If you spend such amounts of money on PSU/UPS, etc 1.and you don't habe a high quality MB, you are simply ignorant of PC and gaming tech. But since I am sure your MB is a good quality one, 2.there is NOT A CHANCE an RX480 could damage it. Especially since 3.AMD will fix that in 2-3 days with a new driver or anyone could fix it NOW by himself by lowering voltage a bit through Wattman.
http://semiaccurate.com/2016/07/01/investigating-thermal-throttling-undervolting-amds-rx-480/
1. Or maybe , -just maybe-, there are aren't any more Socket939 mobo's in the market and i had to buy whatever i could find regardless quality. Ever thought this possibility?
2. When something gets out of spec, then i don't need neither AMD's or your's or anybody else's reassurance. Out of specs means by default: possible danger for my system!!
3. I'm allergic to the word "will" . First they must fix it and then we are going to evaluate the results.;)
Posted on Reply
#264
cdawall
where the hell are my stars
sith'ari
1. Or maybe , -just maybe-, there are aren't any more Socket939 mobo's in the market and i had to buy whatever i could find regardless quality. Ever thought this possibility?
2. When something gets out of spec, then i don't need neither AMD's or your's or anybody else's reassurance. Out of specs means by default: possible danger for my system!!
3. I'm allergic to the word "will" . First they must fix it and then we are going to evaluate the results.;)
No one cares that you spent 600 euros on a power supply and ups for a s939 system that is essentially throw away now.
Posted on Reply
#265
sith'ari
cdawall
No one cares that you spent 600 euros on a power supply and ups for a s939 system that is essentially throw away now.
I was replying to a comment so apparently somebody cared. :rolleyes:
Posted on Reply
#266
cdawall
where the hell are my stars
sith'ari
I was replying to a comment so apparently somebody cared. :rolleyes:
Every comment has been the same. Literally no one on here understands why you are complaining, cool thing about capitalism is if you don't like a product you just don't buy it.

This particular card is a horrid idea for you because even after the software limits the PCIe draw it will still smoke your 10+ year old motherboard, which wasn't even high end 10+ years ago.
Posted on Reply
#267
newtekie1
Semi-Retired Folder
john_
I didn't missed the point. You missed all the other posts that I did about the subject and I am not going to repeat everything in detail. Just ask yourself this.
100% performance at 75W. At that review W1zzard overclocks the card and gets 20% extra performance. Not just higher clocks. 20% extra PERFORMANCE.
Now tell me. How much extra power consumption do you need for that extra 20%? 0 Watts? 5 Watts? 10 Watts? 20 Watts? And don't tell me about throttling. Throttling doesn't increase performance by 20%.
Thanks to GPU boost, basically 0w.

john_
You see, there are many things that the press will not tell you. You just learned about the PCIe bus power draw because the RX480 is an AMD card. If it was an Nvidia card, you wouldn't have known about it.
Or maybe it is just because this is the first card we've seen do this since we've had the ability to test slot power draw separately.
Posted on Reply
#268
silentbogo
john_
You see, there are many things that the press will not tell you. You just learned about the PCIe bus power draw because the RX480 is an AMD card. If it was an Nvidia card, you wouldn't have known about it.
I learned about PCI-E bus because of this:

...and this:


...and about 28 more reasons in my office - I fix this stuff occasionally, if you know what I mean.

Now, if it came to defensive insults, what makes you a specialist in this area?

P.S. Boards are not for sale! Can trade a Z77 for cheap air conditioning :toast:
Posted on Reply
#269
R-T-B
john_
I didn't missed the point. You missed all the other posts that I did about the subject and I am not going to repeat everything in detail. Just ask yourself this.
100% performance at 75W. At that review W1zzard overclocks the card and gets 20% extra performance. Not just higher clocks. 20% extra PERFORMANCE.
Now tell me. How much extra power consumption do you need for that extra 20%? 0 Watts? 5 Watts? 10 Watts? 20 Watts? And don't tell me about throttling. Throttling doesn't increase performance by 20%.

You see, there are many things that the press will not tell you. You just learned about the PCIe bus power draw because the RX480 is an AMD card. If it was an Nvidia card, you wouldn't have known about it.
You still are really not getting how nvidia boost works...
Posted on Reply
#270
Assimilator
john_
You see, there are many things that the press will not tell you. You just learned about the PCIe bus power draw because the RX480 is an AMD card. If it was an Nvidia card, you wouldn't have known about it.
Son, you just went full retard. I think we're about done here.
Posted on Reply
#271
revin
newtekie1
The problem with the RX 480 is that it pulls a lot more than 75w for a sustained period of time.
newtekie1
Or maybe it is just because this is the first card we've seen do this since we've had the ability to test slot power draw separately
This, Bingo !! That IS the issue, the current draw is continuously higher than the [revised]SIG spec allows from the thru the PCIe connector.
Posted on Reply
#272
alucasa
This whole thread is ....

LOL

We need to make a webdrama outta this. It will be pure golden.
Posted on Reply
#273
zAAm
Assimilator
Son, you just went full retard. I think we're about done here.
Agreed ;)
Posted on Reply
#274
Ungari
Assimilator
What it boils down to is simply that if AMD hadn't been cheap f**ks and tried to shave 2 cents off the BOM by using a 6-pin connector instead of an 8-pin, they wouldn't be having this problem. That's an absolutely indefensible case of cutting corners.
I thought AIBs were adding 8-pins for higher total power limits for overclocking?
I have not seen anything concrete that shows that an 8-pin would decrease the draw on the PCIE slot, as my understanding is that is regulated by the GPU itself.
Posted on Reply
#275
GhostRyder
Ok, so this thread seems to be spiraling out into a war. Should we lock and load?

In all seriousness, here is what it boils down to:

1: AMD decided to put a 6 pin instead of an 8 pin reference to look lower power instead of being smart and letting us have the clocking and less problems.
2: AMD needs to release a driver fix to stop the card from overdrawing from the PCIE and either change it to the 6 pin or limit it.
3: Even if you buy this card, your not going to kill your motherboard with it unless you have the most basic/cheap motherboard possible and even then I would be skeptical.

Fact is this should not be a problem but it is. Is it a big problem that is going to result in dead motherboards? No because motherboards especially in this day and age are pretty tough even on the cheap side. I have overloaded a motherboard's PCIE's before, it takes alot to actually do some damage to it. But the fact is AMD was beyond foolish to not only not put an 8 pin, but to let this pass through like this instead of allowing the 6 pin to take the brunt. PSU's in this day and age have an 8 pin minimum even on the most cheap entry level one you would want to buy to support your gaming rig (Speaking ~500watt). Either way though, this does not ruin the card or the value of what your getting, but it definitely makes after market variants look alot more appealing.
Posted on Reply
Add your own comment