ASUS PRIME X570 Pro review
ASUS ROG STRIX Scope keyboard review
GALAX GeForce RTX 2070 Super WTF review
ADATA Spectrix S40G 512GB NVMe M.2 SSD review
Red Dead Redemption 2: PC graphics performance benchmark review
XFX Radeon RX 5700 XT THICC III Ultra review
Promo: URcdkey Halloween sale Windows 10 Pro for 11$
Guru3D Rig of the Month - October 2019
MSI GeForce GTX 1660 SUPER GAMING X review
Gigabyte GTX 1660 SUPER GAMING OC review
AMD Gives Statement on the PCI-Express Overcurrent Problems
AMD just issued a small statement on the PCI-Express over-current Issues mentioned and reported on the web. It seems that the Radeon RX 480 reference cards are going over spec using more than 75 Watts over the PCI-Express slot and then another 75 Watt over the 6-pin power connector.
If you pull over 75 Watts from the motherboard slot, it will place more strain on it and some motherboards might note cope with that very well. To that end and to solve the issues reported they will address this issue in software. The cards do exceed 150 Watts, in fact we measured it at 166 Watts. Here is that statement:
"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)."
If this is fixed by software (driver) this likely means it will be throttling down if it hits a certain power state, so if a higher than 150 Watt power state is reached, AMD might be throttling down the cards in the future to meet the proper P state. Yesterday we have been performing some additional stress tests ourselves and cannot replicate the issues mention on the web. We do see a minor increase of temperature on the PCI-Express slot, but nothing rather concerning. Yesterday we updated the FLIR images in our reference review to show that.
__hollywood|meo
Senior Member
Posts: 2975
Joined: 2005-09-27
Senior Member
Posts: 2975
Joined: 2005-09-27
#5299602 Posted on: 07/02/2016 10:54 AM
theyre having a field day trolling at about this. the discussions ceiling is so low i cant be bothered to chime in there. more on topic, im glad AMD is doing something about this problem quickly however. making mistakes isnt wat we should base judgement on; its how mistakes are addressed & rectified that matters...coming from someone who has owned plenty of team greens gpus over the years
theyre having a field day trolling at about this. the discussions ceiling is so low i cant be bothered to chime in there. more on topic, im glad AMD is doing something about this problem quickly however. making mistakes isnt wat we should base judgement on; its how mistakes are addressed & rectified that matters...coming from someone who has owned plenty of team greens gpus over the years
Devilhunter12
Senior Member
Posts: 71
Joined: 2016-01-18
Senior Member
Posts: 71
Joined: 2016-01-18
#5299604 Posted on: 07/02/2016 10:56 AM
They will lower the clocks and their is no other logically way.
theyre having a field day trolling at about this. the discussions ceiling is so low i cant be bothered to chime in there. more on topic, im glad AMD is doing something about this problem quickly however. making mistakes isnt wat we should base judgement on; its how mistakes are addressed & rectified that matters...coming from someone who has owned plenty of team greens gpus over the years
They will lower the clocks and their is no other logically way.
pharma
Senior Member
Posts: 1093
Joined: 2003-09-10
Senior Member
Posts: 1093
Joined: 2003-09-10
#5299611 Posted on: 07/02/2016 11:08 AM
According to Ryan Smith it could affect reviews depending on what the root cause is.
"If power is somehow being wasted and it can be stopped (think leaking hose) then no. However if the issue is that power consumption is just being allowed to go too high (think a hose turned up higher) then yes, the card would need to be throttled back to avoid breaking power containment."
I guess nobody can blame AMD for last minute clocks/voltage changes to be more competitive, but they forgot (or turned blind eye) PCI-E certification. Seems like reviews will need to be redone ....
According to Ryan Smith it could affect reviews depending on what the root cause is.
"If power is somehow being wasted and it can be stopped (think leaking hose) then no. However if the issue is that power consumption is just being allowed to go too high (think a hose turned up higher) then yes, the card would need to be throttled back to avoid breaking power containment."
I guess nobody can blame AMD for last minute clocks/voltage changes to be more competitive, but they forgot (or turned blind eye) PCI-E certification. Seems like reviews will need to be redone ....
WhiteLightning
Moderator
Posts: 28046
Joined: 2007-09-19
Moderator
Posts: 28046
Joined: 2007-09-19
#5299614 Posted on: 07/02/2016 11:11 AM
If they will lower the clocks, i wonder how much of a difference it will make with benchmark scores. not that much i reckon. In any way it is sloppy work if it falls out of specification.
If they will lower the clocks, i wonder how much of a difference it will make with benchmark scores. not that much i reckon. In any way it is sloppy work if it falls out of specification.
Undying
Senior Member
Posts: 11851
Joined: 2008-08-28
Senior Member
Posts: 11851
Joined: 2008-08-28
#5299615 Posted on: 07/02/2016 11:12 AM
I dont see this effecting the non-reference cards in the future. People should not be forgeting that.
Statement is for reference cards only using one 6pin.
I dont see this effecting the non-reference cards in the future. People should not be forgeting that.
Statement is for reference cards only using one 6pin.
0blivious
Senior Member
Posts: 2576
Joined: 2006-04-25
Senior Member
Posts: 2576
Joined: 2006-04-25
#5299618 Posted on: 07/02/2016 11:20 AM
I dont see this effecting the non-reference cards in the future. People should not be forgeting that.
Statement is for reference cards only using one 6pin.
I believe custom boards will be a non-issue but it's a black eye nonetheless. It just shows some level of either incompetence at AMD or a willingness to play loose with the rules.
I dont see this effecting the non-reference cards in the future. People should not be forgeting that.
Statement is for reference cards only using one 6pin.
I believe custom boards will be a non-issue but it's a black eye nonetheless. It just shows some level of either incompetence at AMD or a willingness to play loose with the rules.
Anarion
Senior Member
Posts: 13524
Joined: 2005-08-13
Senior Member
Posts: 13524
Joined: 2005-08-13
#5299620 Posted on: 07/02/2016 11:25 AM
So there was some wonky stuff. Good thing that they admit it and are working on fixing it. I hope it doesn't lower the performance noticeably.
So there was some wonky stuff. Good thing that they admit it and are working on fixing it. I hope it doesn't lower the performance noticeably.
moeppel
Senior Member
Posts: 149
Joined: 2015-06-30
Senior Member
Posts: 149
Joined: 2015-06-30
#5299622 Posted on: 07/02/2016 11:27 AM
There actually is.
Tests have shown that undervolting a 480 results in higher performance, less power usage and subsequently less heat.
Of course each card has a different undervolting threshold before tapping out.
That said, I still consider the reference a bit of a disaster and am looking forward to AIB cards.
They will lower the clocks and their is no other logically way.
There actually is.
Tests have shown that undervolting a 480 results in higher performance, less power usage and subsequently less heat.
Of course each card has a different undervolting threshold before tapping out.
That said, I still consider the reference a bit of a disaster and am looking forward to AIB cards.
Alessio1989
Senior Member
Posts: 1400
Joined: 2015-06-11
Senior Member
Posts: 1400
Joined: 2015-06-11
#5299640 Posted on: 07/02/2016 11:54 AM
Lol too much tuning. This reminds me what did their first competitor with last cards (except thei introduced throtteling to pair the power issue).
Lol too much tuning. This reminds me what did their first competitor with last cards (except thei introduced throtteling to pair the power issue).
CalculuS
Senior Member
Posts: 2998
Joined: 2014-07-28
Senior Member
Posts: 2998
Joined: 2014-07-28
#5299651 Posted on: 07/02/2016 12:27 PM
There actually is.
Tests have shown that undervolting a 480 results in higher performance, less power usage and subsequently less heat.
Of course each card has a different undervolting threshold before tapping out.
That said, I still consider the reference a bit of a disaster and am looking forward to AIB cards.
How would that even work? The only situation I can see an undervolted card perform better if its throttling based on temparature which I don't think is the case.
There actually is.
Tests have shown that undervolting a 480 results in higher performance, less power usage and subsequently less heat.
Of course each card has a different undervolting threshold before tapping out.
That said, I still consider the reference a bit of a disaster and am looking forward to AIB cards.
How would that even work? The only situation I can see an undervolted card perform better if its throttling based on temparature which I don't think is the case.
Turanis
Senior Member
Posts: 1435
Joined: 2014-08-15
Senior Member
Posts: 1435
Joined: 2014-08-15
#5299654 Posted on: 07/02/2016 12:32 PM
Lest see this
Polaris - Power Problems or PCI Exaggeration?
Lest see this
Polaris - Power Problems or PCI Exaggeration?
Devilhunter12
Senior Member
Posts: 71
Joined: 2016-01-18
Senior Member
Posts: 71
Joined: 2016-01-18
#5299659 Posted on: 07/02/2016 12:38 PM
Lest see this
Polaris - Power Problems or PCI Exaggeration?
That guy is a troll. He is the one of those people who said Polaris can be near to GTX 980 Ti. A sensible person will never say that.
Lest see this
Polaris - Power Problems or PCI Exaggeration?
That guy is a troll. He is the one of those people who said Polaris can be near to GTX 980 Ti. A sensible person will never say that.
moeppel
Senior Member
Posts: 149
Joined: 2015-06-30
Senior Member
Posts: 149
Joined: 2015-06-30
#5299663 Posted on: 07/02/2016 12:41 PM
https://www.reddit.com/r/Amd/comments/4qoclm/german-site-explores-the-potential-for/
I cannot seem to find any true English resources on the subject. And I'm quite bad at electrical engineering
How would that even work? The only situation I can see an undervolted card perform better if its throttling based on temparature which I don't think is the case.
https://www.reddit.com/r/Amd/comments/4qoclm/german-site-explores-the-potential-for/
I cannot seem to find any true English resources on the subject. And I'm quite bad at electrical engineering
cowie
Senior Member
Posts: 13195
Joined: 2005-03-30
Senior Member
Posts: 13195
Joined: 2005-03-30
#5299664 Posted on: 07/02/2016 12:42 PM
he is not a troll but sucks on amd's ahole
not for nuttin' the guy who just posted the link Is no better
That guy is a troll. He is the one of those people who said Polaris can be near to GTX 980 Ti. A sensible person will never say that.
he is not a troll but sucks on amd's ahole
not for nuttin' the guy who just posted the link Is no better
Click here to post a comment for this news story on the message forum.
Posts: 71
Joined: 2016-01-18
Of course they will lower the clocks of 480 and card will throttle. AMD is doing a very simple logic.