RX 480 is apparently killing pcie slots

Imhotep said:
Has it occurred to you that every card has its own power circuit and the reading is a sum of all three. That is why a Rx 480 can draw more than what the specs call for. Just as any other card can and more often than not does.
Have a great weekend Gents :)
Click to expand...

the specification for a GPU it's about the TOTAL power utilized by the VIDEO CARD. not only the POWER utilized by the GPU core itself.. if not then every other GPU ever made would show numbers that will be undoubtedly way lower than what actually AMD it's trying to do now by just saying they state GPU core power which it's blatantly fake.. that's just a counter answer to this whole mess, that's all..
 
kalston said:
Had a quick glimpse but it looks like they raised power limit to 150% or am I missing something? see at 4:32 for example : "PL 150%"
Also they say 1300mhz, is that the average clock (then that's stock) or base clock? Because that would mean overclocked then, no?

That all does not change the fact that the ref rx 480 has a serious design flaw though (and should really NOT be overclocked).
Click to expand...
Valley 1300 mhz... light
Furmark stock heavy
 
rbuass said:
Valley 1300 mhz... light
Furmark stock heavy
Click to expand...

I thought it was stock as well, in their defense though, xfx is selling reference cards at 1330mhz. So... Bleep bloop, something's gotta give here

Edit: lol, I didn't realize it was you posting about your own results, sorry.

I'm just wondering what avg power did you measure on the 12v at stock
 
Ieldra said:
On the psu?




The 12v rail will have overcurrent protection of course, but the limit at which it is set must also allow for fluctuation within spec. Moreover, the damage we are talking about, that could potentially happen, is due to localized power draw; as pcper mentioned it is about power dissipation in pins and connectors.

The 12v rail on the mobo does more than just power the gtx 480
Click to expand...

Yes the PSU and I know this. I was mainly talking about protection for the CPU as it has its own main power source separate from the ATX connector (the connector that is mostly effected by the GPU power draw).
 
Ieldra said:
On the psu?




The 12v rail will have overcurrent protection of course, but the limit at which it is set must also allow for fluctuation within spec. Moreover, the damage we are talking about, that could potentially happen, is due to localized power draw; as pcper mentioned it is about power dissipation in pins and connectors.

The 12v rail on the mobo does more than just power the gtx 480
Click to expand...

On cheap mobos it's likely a slow blow poly fuse. They work fine, but the only problem with them is once they trip, they trip easier in the future. (Less tolerant to overpower)
 
  • Like
Reactions: N4CR
like this
Guys...
This forum
Ieldra said:
I thought it was stock as well, in their defense though, xfx is selling reference cards at 1330mhz. So... Bleep bloop, something's gotta give here

Edit: lol, I didn't realize it was you posting about your own results, sorry.

I'm just wondering what avg power did you measure on the 12v at stock
Click to expand...
Usual

We made a long live stream...
For example, playing GTA 5 on benchmark mode, run close the limit, no pl, no oc, no fan 100%., nothing....
This weekend we should edit best parts and update.
Is a pci ex hungry powa.... hahaha

However is a pretty good performance cost card.

:)
 
Logan321 said:
Power Consumption Concerns on the Radeon RX 480 | PC Perspective PC Perspective says their testing shows the pcie power draw exceeding spec... by 2W, and for very short intervals. I'm fairly sure that's within margin of error. I'll link more tests as I find them.
Click to expand...

RIF.

He forgot to include the 3.3Volt line which puts it OVER spec.

I must admit I mistakenly thought it was limited to 75 Watts on the 12V line, but it's 75Watts COMBINED 12 + 3.3V lines
 
Trimlock said:
Yes the PSU and I know this. I was mainly talking about protection for the CPU as it has its own main power source separate from the ATX connector (the connector that is mostly effected by the GPU power draw).
Click to expand...

Unless
Logan321 said:
Power Consumption Concerns on the Radeon RX 480 | PC Perspective PC Perspective says their testing shows the pcie power draw exceeding spec... by 2W, and for very short intervals. I'm fairly sure that's within margin of error. I'll link more tests as I find them.
Click to expand...
Being the spectacular asshole that I am I looked at the PCIe spec it specifies [email protected] max with a maximum tolerance of 9%; 71..3w

480 draws 82avg at load, stock

Also note the spec specified maximum current and voltage with a tolerance.

This means current is also over spec considering vdroop
 
DigitalGriffin said:
RIF.

He forgot to include the 3.3Volt line which puts it OVER spec.

I must admit I mistakenly thought it was limited to 75 Watts on the 12V line, but it's 75Watts COMBINED 12 + 3.3V lines
Click to expand...
I didn't know that either before this, had no idea a 3.3v line went to the slot.
 
Spidey329 said:
So I'm guessing the consensus is to just avoid the reference design? This will likely be fixed by the OEM's, as they tend to include more 12v power plugs on their board changes. That should reduce the draw from the motherboard.

PCIe standard is sustained at 75w or less. 6pin PCIe power is rated for sustained 75w, total power continuous is 150w.

From a lot of tests, the RX480 is pulling 160-170w and upwards of 200w when over clocked. That'd end up pushing both the PCIe bus and PCIe 12v past their rated standards.

Should have just slapped an 8pin on there.
Click to expand...

Yes any variant with 2x 6 pin or 1x8 pin should be safe when they are released.
 
WCCF article down. I wonder if he's watching these threads?
 
Early gen SLI/CF mobo has extra molex connector on the mobo itself to supply extra juice to the GPU. Perhaps it's time to re-introduce this.
 
Michaelius said:
Yes any variant with 2x 6 pin or 1x8 pin should be safe when they are released.
Click to expand...
So, do we have an understanding that the 8 pin or any other combination has been left for the AIB's to have something to play with in terms of overclocking heavily.
Shit, my PC just started smoking. The two XFX Rx 480 Bleck Editions are pumping smoke out the back of the case...lol
 
  • Like
Reactions: N4CR
like this
Zion Halcyon said:
Even funnier - WCCF took down their article on this.
Click to expand...

I read this article and I hate WCC, it was most likely taken down because it pretty much quoted PCPer's entire write up. Then at the bottom it tried to be different by praising AMD's superior power delivery system over Nvidia's.
 
WCCF will rip and steal every kind of news/rumor article they can find. It should be very telling they are avoiding this issue like the plague.
 
PCPer said:
The white line of the MB +12V power is going over 75 watts, but not dramatically so, while the +3.3V power is hovering just under 5 watts, for a total of ~80 watts
Click to expand...

PCPerspective was running 1080p

Toms was running 1440p
 
Trimlock said:
I read this article and I hate WCC, it was most likely taken down because it pretty much quoted PCPer's entire write up. Then at the bottom it tried to be different by praising AMD's superior power delivery system over Nvidia's.
Click to expand...

AMD's power delivery is much more 'open'
TaintedSquirrel said:
WCCF will rip and steal every kind of news/rumor article they can find. It should be very telling they are avoiding this issue like the plague.
Click to expand...

Telling of what?
 
DigitalGriffin said:
RIF.

He forgot to include the 3.3Volt line which puts it OVER spec.

I must admit I mistakenly thought it was limited to 75 Watts on the 12V line, but it's 75Watts COMBINED 12 + 3.3V lines
Click to expand...

^^ This. It did however note that the really big spec violations only happen when the power limit is raised and you were already warned about it.
 
dual zombie.jpg
think we, from TecLab, have a good solution for AMD...
No more babies cards...

20150531_145138.jpg
20150531_145055.jpg
20150531_145209.jpg
20150531_145055.jpg
20150531_145209.jpg
20150531_145055.jpg
20150531_145138.jpg
 
Last edited:
jbc029 said:
^^ This. It did however note that the really big spec violations only happen when the power limit is raised and you were already warned about it.
Click to expand...

True

On the other hand, why is there a 50% power limit increase if the board already uses 170w of its alloted 150.

And why is it going over the PCIe spec under any circumstances, there's a damned 6pin
 
Imhotep said:
So, do we have an understanding that the 8 pin or any other combination has been left for the AIB's to have something to play with in terms of overclocking heavily.
Shit, my PC just started smoking. The two XFX Rx 480 Bleck Editions are pumping smoke out the back of the case...lol
Click to expand...

No and we probably won't know ever why they did it.

IMHO most likely theory - they had power budget and clocks in mind when designing board but what came back from GF was Polaris as we know it today. And with Pascal clocking higher than everyone expected they had to squize as much performance from what they got so it's clocked as close to maximum allowed by board as possible.
 
Michaelius said:
No and we probably won't know ever why they did it.

IMHO most likely theory - they had power budget and clocks in mind when designing board but what came back from GF was Polaris as we know it today. And with Pascal clocking higher than everyone expected they had to squize as much performance from what they got so it's clocked as close to maximum allowed by board as possible.
Click to expand...
Its a conspiracy to burn all the mobos out there...Damn You AMD :D
By the way mine stopped smoking and are still running.
 
lol didn't recognize the twisted foil, is this a suicide run? Theres more volt mods done on this then the 9800pro had.
 
Michaelius said:
No and we probably won't know ever why they did it.

IMHO most likely theory - they had power budget and clocks in mind when designing board but what came back from GF was Polaris as we know it today. And with Pascal clocking higher than everyone expected they had to squize as much performance from what they got so it's clocked as close to maximum allowed by board as possible.
Click to expand...
They were trying to prove Kyle was wrong but sent the work remit to the wrong dept.
The youth training evening class thought they had done quite well. The class leader put an A+ on it and simply followed instructions, handing it back as requested.
Oops.
:p
 
You must log in or register to reply here.
Back
Top