>Polaris will consume 86w
NVIDIA BTFO
>pajeet listed only the pcie draw
>980ti performance for 200 dollars
HOW CAN THEY RECOVER?
>>55342030
>4 driver errors have been deposited into your kernal
how did everything go so wrong?
>the 480 will have 980 preformance
>>55342888
project cars is an nvidia title you fucking retard
>>55342888
hnmmmmm
>>55342030
Thats Polaris11, and that is total system power consumption.
The GTX 950 pulls about 100w, and the Polaris 11 GPU was drawing just under 50w.
>>55342866
Reference card sucks, but it's not so bad. Will stir up prices at the least, and board partner cards won't be so poo in loo. I think good cooling and power options will make this card very capable, even if overhyped. Just hope the price is right. All this hullabaloo about pic-e draw is bullshit. The 750ti regularly hit 125-140 watts.
http://www.tomshardware.com/reviews/geforce-gtx-750-ti-review,3750-20.html
@55342888
>meme cars
>>55342914
125-140 watts drawn solely from the pic-e slot that is.
>>55342888
>Project cars
Ayy
>86W
>All drawn directly from the PCIe bus
I-It's fine guys! It's just a stupid "standard"!!
>>55344811
Keep trying, shiposter.
Someone will surely think you're clever and funny one day.
>>55342030
Glad to see another GPU thread on the consumer board.
>>55342908
>performance increase over 28nm
>390 & 390X, 28nm GPUs from 2013 are faster
>>55342932
The average is only 62W though. The issue here is that traces could potentially burn up. I'm not sure how mobos are built these days, but if they were to burn up at 86W then even at 75W they would be at their limit, so the mobo would have to be a pretty shit design to not have a decent amount of headroom in terms of power capacity for some copper traces.
That being said, because the issue is thermal in nature, spikes won't have too much of an effect as long as the average isn't too high. The trace won't heat up instantaneously during a spike to 140W, but if you have 86W going through constantly then that means that the trace is definitely under more thermal stress than it was designed for (assuming it was designed for 75W and not more).
That's why the spikes on a 750Ti are pretty much irrelevant but an average over 86W is potentially more dangerous.
>>55345028
This, and the fact that it draws up to 200w in spikes when you attempt to OC makes the 750ti comparisons moot.
>>55345028
1.) The 86w figure in the OP is total system power draw
2.) There is no issue of a steady 100w being delivered through the PCI-E slot. No motherboard is going to have any traces catch fire and evaporate.
3.) Power is delivered by the PSU. The "limits" that all you autistic tech illiterate retards keep regurgitating have to do with how the PSU divides its rails and how it handles amperage across them.
4.) Power draw over the slot shown by the 960 Strix and RX 480 isn't ideal, but it isn't harmful either.
5.)Low teir mobo VRM with 4 phases and no heat sink is a far, far bigger risk, and no one here ever mentions that because you lot are too fucking retarded to know a single thing about currents.
Not long ago you average shitposters saw a Tom's review with their oscilloscope power measurements for the first time, and all of you were acting as if momentary peaks in power delivery was the same as an average. For days on end there was nonstop shitposting
>hurrr fury housefires 400w
>durrr 980ti housefires 325w
Not a single one of you has any idea what the fuck you're talking about. Even pretending for a minute to be an informed voice of authority is fucking pathetic. Go hang yourself.
>>55345142
>Total system power draw
No, everyone is reporting that this is through the PCI-E slot, the total card power draw is 160+ watts, total system draw is over 200 watts.
![Jeremy-Renner-and-Chris-Evans-laughing-at-a-joke[1].jpg](https://i.imgur.com/VRctq54.jpg "Jeremy-Renner-and-Chris-Evans-laughing-at-a-joke[1].jpg")
>the reference 480 is literally being bottlenecked by not getting enough power draw
>there are people on /g/ right now that bought a reference 480 without the intent on putting a $120 waterblock on it
https://www.youtube.com/watch?v=8Oc7zXhzlzU
lol (laughing out loud)
>>55342888
>>55342908
Graphic card company sponsored games go both ways. Project cars uses Nvidia shitworks, and Hitman uses AMD whatever the fuck they are doing now.
Until games have unbiased engines, we will never see what these cards actually do 100%,
>>55342030
idle
>>55345182
>I'm a shit eating retard
The OP image is about Polaris 11. The 86w figure is total system power draw.
Go tie a rope around your neck and end it.
>>55345219
>AMD whatever the fuck they are doing now
You mean DirectX 12?
>>55342030
86W off the PCIE slot maybe.
>>55342908
>Stop using Nvidia shilling games you fucking retard
>Here check out my AMD shilling game
AMD shills lack self awareness to a worrying degree.
>>55345028
The floor lowers the average. It's a tight wave pattern that consistently goes over 100 watts. If that wasn't destroying cheap mobos then, the 480 pulling a more stable 85 at the slot won't destroy decent ones now.
>>55342030
>AMD Internal Lab testing as of Dec 2, 2015 with an Intel Core i7 4790K
Fucking sad that AMD is so aware that their CPUs are utter garbage that they run their benches using their competitors CPUs.
>>55345142
>1.) The 86w figure in the OP is total system power draw
That's not what I'm referring to, I'm talking about the 480's average power draw from the PCI-E slot. I didn't remember the right number, it was 82W and not 86W, not a huge difference.
>2.) There is no issue of a steady 100w being delivered through the PCI-E slot. No motherboard is going to have any traces catch fire and evaporate.
I suspect so too, but if you're so certain I'd love to see a factual source on that. Being 100% fine with 75W but not 82W sounds pretty much impossible to me, but it's still out of spec technically (and at the same time you can't guarantee that the shittiest mobos out there can handle 75W anyway, though it would be unlikely)
>3.) Power is delivered by the PSU.
No shit Sherlock, what brilliant insight. I don't think anybody is worried that their PSU can't deliver an extra 7W over the 75 it has to deliver anyway. Most modern designs are single rail in any case, even multi rail designs tend to be single rail but with separate current limiters per "rail"
>>55345296
The floor lowers the average because the average is low. A conductor won't burn up if its average capacity isn't exceeded for sufficiently long periods of time. Anyway going 7W above spec is extremely unlikely to actually damage something, I agree.
Still the reference design is pretty much shit (the cooler too) so waiting for custom models and even 1060 benches is definitely the smarter thing to do.
http://www.tomshardware.com/reviews/amd-radeon-rx-480-polaris-10,4616-9.html
>We skipped long-term overclocking and overvolting tests, since the Radeon RX 480’s power consumption through the PCIe slot jumped to an average of 100W, peaking at 200W. We just didn’t want to do that to our test platform.
BLUNDER OF THE CENTURY
>>55345289
see >>55345253
Just admit it, Nvidia is the worst choice long-term if they can't take advantage of DX12. Pascal is one giant flop and nvidia will have salt poured on their wounds as more DX12 games come out.
I own a GTX 960 btw but even I can't ignore Nvidia's faggotry. I'll probably go with AMD now. But I won't get their rx 480, I'm interested how badly the rx 490 will rape the GTX 1080 especially in DX12.
>>55345639
>Just admit it
Why is it this phrase always preceeds retarded arguments?
>>55345639
>especially in DX12.
Ooooooooooooooo, DX12.
Keep clinging to Ashes of the Shitgularity as the only game your nebulous desires for DX12 amazingness will ever exist.
>>55342919
you
>>55345748
>Already seeing a big boost for AMD hardware in games Hitman, Quantum Break, Forza 6: Apex, Total War: Warhammer and Gears of War: Ulitmate Edition
>Plenty more DX12 titles in the pipeline
Feel free to keep acting like AotS is the only DX12 game out there and that no more DX12 titles will be released, but if you do, please stick to your home board of /v/.
>>55345965
The latest Dirt game too.
>>55342919
Picardia pls
>>55342030
>doesnt have a 6pin pci-e connecter
>>55345734
An attempt at intimidation.
Also off by one
