>consumes more power than the 1070
>still performs worse
At least you can use your 480 to cook some eggs on it.
POO IN LOO, PAJEET
>>55515776
It's meant to compete with the 970, so I suppose that's alright.
Would a I5 6500 bottleneck a 1070 for wqhd?
>>55516674
My phenom ii x4 did not, i doubt your new cpu can.
>>55515776
>consumes less than the 970
>destroys it
>costs less
kek
>>55516220
>>55516721
It's supposed to compete with the 1060, the 970 is retired.
>>55516746
>it's supposed to compete with something not out
also 1060 is already slower than it, has no sli, likely has all the problems other pascal cards have like pixel rates, stuttering above 60hz and displayports not working with htc vive, plus
>SOLDERED PCI EXTENSION CABLE
http://www.3dmark.com/compare/fs/9202637/fs/9188962
At least it beats a 1060
>>55516760
>has no sli
isn't that only for the 3gb cards? and who the fuck cares? who's buying a 1060 with the intention of SLIing?
>>55515776
Didn't someone try to cook an egg on the GTX 480 and failed? Maybe there is some sort of curse on products designated with a "480."
>>55515776
How much does it cost compared to the 1070?
>>55516830
there is no 3gb card
the whole "3gb has no sli, 6gb does" is a terrible rumour
>>55515776
>caring about power consumption on gaming GPUs
>Ever
I want this marketting induced meme to die.
>>55516721
it's equivalent perf/watt and performance of a reference 970 despite the: 'new architecture', new 14nm process and the fact that it's two years after the 970 launched.
an OCed 970 is vastly superior performance and perf/$ than the price gouged 480.
>>55516866
an OCed 480 destroys it.
>>55516878
the 480 has very little OC headroom and it's ROP bottlefucked into irrelevancy.
I really don't see how people can think the 480 is a good product. Nobody wants to upgrade from their 970-tier performance to 970-tier performance. AMD really needed the 480 to be 20% faster out of the box.
>>55516674
Why would one of the strongest CPUs bottleneck any card? Smhtbhqfam
>>55516866
>OCed 970 is vastly superior performance
LOL Thats a good one.
>>55515776
OP just so you know AMD believes in hardware doing all the work where nvidia believes in CPU and Software doing all the work. So AMDs cards basically will need more power as they have more hardware. It's just two different approach. Kind of like Software vs Hardware Firewall. Also AMD is betting on multi GPU in the future and Nvidia maybe betting on single.
>>55516904
well, 1060 shit tons of headroom and yet when OCed it to 1,900mhz it doesn't outperform 1,300mhz 480.
it doesn't matter if you go fast, if you can't do any calculations in your cycle your cycles are fucking useless.
http://www.3dmark.com/compare/fs/9202637/fs/9188962#
Based curry brother beats cunt eyed jew on his own ground, in his own firestrike.
480 oficially > 1060 even at green camp gsmes and green tests at lower price
>>55517054
>well, 1060 shit tons of headroom and yet when OCed it to 1,900mhz it doesn't outperform 1,300mhz 480.
[citation needed]
we haven't seen any legit benchmarks of the 1060 so far.
480 is probably the worst GPU of all time, what a disaster
>>55517079
Here is citation i go by >>55517072
This might not be 1060 but if it is, 1060 wont be worth the money.
>>55517054
1060 is litelally half locked defective 1080 chip, with half memory and half bus.
Its a halfass card.
Display you expect Jews to give you a bone for 300$?
>>55516904
>>55517054
And this is why comparing mhz/ghz between AMD and Nvidia is stupid. One can be 1.5Mhz and the other can be 2Mhz but the 1.5mhz will do more because of its design and architecture. Same thing with Tflops you cant compare them unless its with in AMD or with in Nvidia.
>>55517095
>1060 is litelally half locked defective 1080 chip,
GP106 =/= GP104
>>55517082
>every review is saying how rx480 is best 200$ you can spend on a GPU
>worst GPU
math it just doesn't add up.
>>55517092
>his might not be 1060 but if it is,
Display you know many 1.9 GHz GPUs?
16NM Pascal is the only GPUthat can possibly operate at such clocks.
Its cant be anything else.
>>55517113
>$200
GREAT meme friend
https://www.amazon.com/MSI-Radeon-RX-480-8G/dp/B01GX5Z4EM
>>55516865
>implying I can handle more than 160W of TDP on gfx
Not that guy, but my case turns into a hotbox if I put high TDP parts in there. It has very low airflow.
>>55517105
Yeah.
You take defective broken GP106, lock it down, disabke half of its cires and its a different chip.
>>55516865
I want you to move out the basement and start paying your own power bill.
>>55517113
There are plenty of reviews critical of it, especially after the power issues came to light.
I mean the 480 is actually killing motherboards and still goes over the power limit even after the patch.
480 is actually a dangerous GPU that I wouldn't even put in my system just to test.
>>55517127
well, I'm just saying it's not without a doubt, It is highly probably, yes but not certain.
i'm 99% sure it's 1060 I don't know of any other nvidia GPU unless it's 1070m or something.
>>55517130
Posting Amazon and you know you're being a dick as if we wont look at another site.
>>55517140
Muh 10 watts power consumption differebce.
Like 0.001 cent every year.
>>55517148
>It's killing motherboards
yea, nice meme.
>>55517148
at least it's mostly safe now if you use compatability mode. but since the 480's memory is entirely powered from PEG you can't actually OC it without bringing the card back to the out of spec power draw.
mistakes like this should have never made it into a retail product, AMD must have terrible QA or had to rush it out.
>>55517148
1 rupee has been deposited into your designated shitting street Pajeet.
>>55517130
This was 3 days ago a Micro Center and they will have more this week.
>>55517171
Even in compatibility mode, the 480 still goes over the power limit on both the pcie slot and the 6-pin.
480 is seriously a fucked up card that never should have even been released.
>>55517148
You should work for WCCFTECH, you just post what you see on the internet you don't fact check shit.
>>55517186
Nobody cares. It's a non-issue now. Go shit someplace else Pajeet.
>>55517216
>480 is still damaging motherboards
>not an issue
OK buddy, good luck with that.
>>55517148
>Actually
>Actually
>Actually
One reported dead
One is fake
Another is a mobo from 1999 burned with a lighter.
And a few people reported crashes.
1 thats an Nvidiot marketing meme
2 A 480 reference has issues NONE of which are going to be present in custom cards.
Custom cards wont owerdraw PCI, they will have 1350-1450 clocks and they will run cool.
>>55517153
>i'm 99% sure it's 1060 I don't know of any other nvidia GPU unless it's 1070m or something.
How did they get their hands on it?
That must be an engineering sample guarded in NVIDIA labs as a top secret.
Who the fuck can steal it?
>check store for RX480
Reference turd costs between 10-80 euros less than a GTX 980
Wow what a deal.
>>55517225
Links to Blown Mother boards please?
Other then the one I have attached, and check the condition of his board and case.
>>55517225
Provide proof other than *ONE* ASSCock thread on AMD that is confirmed bogus or STFU. No on second thoughts. Just STFU.
>>55517231
We have no idea if the non-reference 480s will even fix the issue, none of them are even out yet.
Here's my bet: all of them are based on the reference 480, therefore all of them will still suffer from the same power issues as the original 480.
>>55517225
no it doesn't damage motherboards, it draws 67w from PEG, thats 2watts over.
6pin is 90w while being over PCI SIG, it's bloody nothing for PSU.
>>55517236
>euros
No one cares about you refugees. Move to America else stfu.
>>55517254
>still goes over power limit
>doesn't damage motherboards
OK buddy
>>55517249
>We have no idea if the non-reference 480s will even fix the issue, none of them are even out yet.
Oh my fucking god kid why are you up so late?!?!?!?
>>55517256
I guess I'll just keep giving nvidia money.
>Nvidia goys on suicide watch
http://www.3dmark.com/compare/fs/9202637/fs/9188962
>>55517264
Find links to legit RX480 blown boards and I will payPal you $20 per board you find.
>>55517225
Name one report of a damaged motherboard made after the fox that isn't some dumb ass shit like some nigger overclocking three of them on a shitty psu.
There are tons of motherboards killed by 480s, just google "480 dead pcie slot" - 378,000 results
>>55517286
>implying an AMD fag actually has more than $2 in his paypal account
>>55517249
It is already established because We know that custom PCB will be full size, have 8pin, and have 2 phases from PCIe and 4 to 6 phases from 8pin.
Reference had 4phases from PCI and 2 from 6 pin for some stupid reason.
MSI PCB photos already show that it's not the case with custom.
>>55517269
Why would I care? My point is you get euro tax and your currency is different but you expect the card to be $199. Thats like saying poor countries should also have the 480 for $199 when in reality $199 in their country is like $5 USD. Like do you understand how currencies even work before you bitch about something?
>>55517317
If I had $2 it wont matter cause their are no blown boards.
>>55517303
Google "I was abducted by aliens" and you will get as many results.
>>55517326
We don't know anything until it's tested. AMD promised so much with the 480 and delivered so little, I would not be surprised if we see the same fuck up with 3rd party boards.
>>55517327
I don't expect it to be 199, I just don't expect to be able to buy a GTX 980 for almost the same money.
>>55517346
>AMD promised so much with the 480 and delivered so little
Yea you would know about being promised a lot and getting a little.
>>55517348
It's funny when you can buy a 970 for the same price that is both faster and won't kill your motherboard like a 480.
The 480 can't even compete with a 2 year old card, pretty sad.
>>55517346
They delivered exactly what they said. A £199 card that can play 1080p @ 60 fps. It was *YOU* and your shills that made up shit about it beating a 980Ti etc.
>>55517348
980 is a 2 year old card dude wtf are you on. All cars do the same shit but you don't see the new BMW being cheaper cause the old ones are still around.
>>55517371
The problem is the 980 is way faster than the 480 for almost the same price.
Makes it basically dumb to buy a 480 and that's not even considering all of the stability issues with the 480.
MFW they cant find a blown mother board but lol
Laters Nfucks.
>>55517346
>We don't know anything until it's tested.
Has anyone actually tested killing mobo with a 480?
Display you have a footage with a testthat can be repeated?
This official PCB shot is enough evidence of what i said.
>480 kills mobos i swear
>We know nothing without a test
You re just a retarded fanboy in denial.
>>55517395
There are plenty of reports of the 480 killing mobos, you can google them yourself.
There was even a guy whose motherboard would just crash over and over again because the 480 was overdrawing power.
>>55517355
>Comparing to a 2 year old card that cost twice the price back then.
This is all about making a card that competes fills an empty space that the 970 leaves. Nvidia are releasing the 1060 because the 480 scared the shit out of them losing the midrange market and making them look bad. They are no longer making the 970 and retailers are just dumping old stock.
>>55517371
It's not that the RX 480 is new, AMD products have idiot tax in Europe. Eg you're an idiot if you buy this.
>>55517354
The 970 really does have 4GB of VRAM. The last .5 is just slower compared to the rest.
>>55517395
>Display you
Fucking retardest auto-correct.
How the fuck can it turn DO into DISPLAY ffs?
>>55517409
The 1060 absolutely destroys the 480, it's not even a contest.
Basically Nvidia and AMD aren't even in the same market anymore.
AMD just makes budget crap that breaks down and kills your components.
>>55517355
>The 480 can't even compete with a 2 year old card, pretty sad
https://www.youtube.com/watch?v=PqgOfR-Oc4U
stop fucking deluding your self, you 970 cucks are the biggest retards ever. a 480 with the same core clock as a 970 performs better. it's only going to get worse for you guys once the custom 480s release.
>>55517385
>Buying old tech
>Maxwell can't even do async compute
>Can't complete in DX12
>Believes DX12 won't catch on (See Rise of the Tomb Raider a triple A title)
Keep trying
>>55517405
If you are too lazy to provide citations then I am too lazy to Google it. Put up or shut up.
>>55517405
Theres a lot of rumors and fake photos.
If it was real vendors would start replacing mobos on warranty en masse because its a warranty case and We would have oficiall evidence.
We dont.
The 480 sells like hot cakes, it sells a lot.
Mobos dont fall like flies for some reason.
>>55517426
why is the rx 480 temp missing
>>55517426
That's not true, there are plenty of benchmarks where the 970 beats the 480.
The 480 can't even beat Nvidia's 2 year old card that is selling for the same price.
>>55517421
I've seen you post this a number of times now. Copy pasting it over and over. You are too lazy to even type something original. How much are they paying you Pajeet?
>>55517385
So then why would you buy it dumb ass? Who said its here to replace a 980 why are you comparing it? All AMD said is 970 performance for $199 They did not say replace your 970 and 980's I don't understand the stupid overload going with you guys. It's your preference. You guys act like people are forcing you to buy shit. Maybe I want a rx480 because it has Display Port 1.3 for my monitor? Maybe I want it because it may work better with DX12 and the 980 does not do that? It's just a card for $199 and if you don't understand that people have different needs then you're just stupid.
>>55517450
I'm sure the lawsuits are already coming. AMD is trying to cover up the 480 power issues, but the truth is already coming out.
>>55517426
>stop fucking deluding your self, you 970 cucks are the biggest retards ever.
you linked a video that shows a properly OCed 970 shitting all over a max OCed 480.
>a 480 with the same core clock as a 970 performs better.
nobody clocks their 970 to 1350mhz. the max boost clock on a reference 970 is 1391mhz. most 970s top out at 1550mhz like the one in the video you linked.
he even had to hide the 480's temperature and hid the 1550mhz 970 half way through the video because of how hard the 480 was getting shit on. AMD truly can't make a competitive product with a two year old gimped card. Amazing.
>>55517461
There's no reason to buy a 480 when the 970 exists. Not only that but the 970 won't damage your motherboard.
>>55517452
Copy & paste fag get a trip so I can block you already.
>>55517452
>i post proof
>you literally can't
what ever helps you feel better nvidiot.
>>55517405
>There are plenty of reports of the 480 killing mobos
Why dont you google them and post it so you can shut us up?
Oh yea cause their is none. Only reports that link back to the 1 kid with soda spots all over his asrock board.
>>55517483
Sorry you can't handle the truth. Maybe you should take your head out of the sand and take a long hard look at all the issues with the 480.
>>55517452
>where the 970 beats the 480.
Theres 970 overclickeded to 1400 beats a 480 4gb at base clocks.
Yes.
There is just one autistic kid in this thread trolling with Nvidia shit. I guarantee it. Don't take the bait. Ignore him and he might get bored and leave.
>all these assblasted AMDumbos
wew
im havin a laugh
>>55517493
I encourage people to do their own research.
A good search term I already posted to search for is "480 dead pcie slot"
Tons of hits there right on the front page that will go into detail about how the 480 is actually damaging motherboards.
>>55517502
you should take your ritalin bud, you might have an aspie stroke if you don't
nvidia > Another Major Disappointment (AMD)
>>55517519
Nvidia has been making superior products for years, but some people love to cling to mediocrity.
>>55517463
>Lawsuit
>And covering
Who the fuck wont NVIDIA expose evidence then?
All you need is a manufactorer confirmed warranty case.
The rx480 is a certified PCIe device, using it on your mobo is ok and civered by warranty, a lot of people use it, where are all the warranty cases?
Where are all the dead mobos?
>Muh AMD covered it
>Muh NVIDIA cant expose it
All the Nvidiots have to clin onto is their precious 970 that they now feel defensive over being goyed into buying. AMD came along and spoiled their party and they have to pull at any little chink in the 480's armor they can find to make them feel better. Especially since the 1060 looks like it's gonna go down like a lead balloon.
http://www.3dmark.com/compare/fs/9202637/fs/9188962
>>55517536
The 480 does not pass pcie specification, that is actually a big part of the problem.
AMD could easily be sued for advertising their card as pcie, when in fact it is technically not even a pcie card because it goes over the power limit.
I found PROOF OF the RX480 burning out!!
See link AMDrones!!
https://www.youtube.com/watch?v=8AA_B1Q1VRw
>>55517554
rx 480 is PCI SIG certified card, you fucking moron.
>>55517554
>could
>maybe
>perhaps
Crystal ball gazing again?
>>55517565
kek. Nvidiots BTFO.
>>55517554
nah, if that was the case they would have been sued for the 295x2 or fury x2 since 2x 8 pin connectors isn't actually part of the pcie standard. only 6+6 and 8+6 are technically allowed.
>>55517554
Using RX480 does not void your mobo warranty.
It has a certficate.
Unfortunately it doesnt live up to that certficate, but no store would ever sells an in certified Black box with potential danger for the user.
Mobos would do that.
>>55517571
The 480 was certified before it came out, but the version they released does not pass specification.
That's the problem - advertised as pcie, not actually a pcie card.
Seriously I can't even imagine how big of a fuck up this was for AMD.
>>55517598
>The 480 was certified before it came out, but the version they released does not pass specification.
since when do you certify something after you release it? You really are a retard.
>>55517607
well that's what amd did
so they're retarded
>>55517607
>since when do you certify something after you release it? You really are a retard.
AMD is really just that retarded. They released a card that didn't even pass the specification.
>>55517618
that is how certifications work, you certify it and then you release it.
if they would of released it before certification they could of not used "PCI-Express" on their card.
>>55517598
AMD didn't get the rx 480 certified.
https://pcisig.com/developers/integrators-list
meanwhile all of nvidias chips since kepler are on the list, wtf amd.
There is one autistic kid in this thread shitposting for Nvidia. You only have to look at what they type and their writing style to see this. Ignore them and move on as it's pointless arguing with 12yo retards.
>>55517641
Interesting, the 480 was never even certified?
That explains a lot.
>>55517598
What shop would ever legaly sells an uncertified GPU?
Thats a lawsuit case. You can sue then for damage, and moral compensation.
GO FOR YOULL BE RICH
what hasnt anyone sued anyone for that?
>no lawsuit
>no warranty returns
>no evidence
>No tests
>no videos
Nvidiots as always full of shit.
>>55517658
I'm sure the lawsuits are already in the works.
>>55517658
all of them, you don't need PCIe certification to sell PCIe parts.
PCIe certification just says "we comply with their standard" nothing else, it's not mandatory and probably all audio cards don't have certification.
>>55517678
Yeah that's true.
The problem comes when AMD advertises the 480 as a pcie card, when it's not actually a pcie card.
>>55517696
Just like you are not actually human.
>>55517678
most soundcards don't pull more than 10-20w though. many GPUs pull over 150w, which is part of the reason why nvidia and amd normally go out of their way to get them up to par with the spec. AMD made a defective design and then released it anyway.
>>55517711
so? how is this relevant?
nowhere does it say "you need pcie certification if you draw over xxx Watts"
This is beauty of PC, it's open platform.
>>55517733
the point is to make sure that you don't have a product that will damage the end user's computer. AMD knew about the power draw problems, knowingly did not submit the card for review to the pcie committee, then released the defective product to the market.
AIBs will probably fix it, but the reference card should not be a defective POS for someone else to fix. it should be a bare minimum working product that partner designs are based off of.
>>55517762
Well said.
I actually almost bought a 480 to put into my secondary build, but held off because I wanted something stable. It's just too much of a risk to have a card that is not going to play nice with the rest of your components.
>>55516760
already slower? Nvidia says it's going to be faster then 980
No idea if it's another lie or not though
>>55517809
The 1060 should be quite a lot faster. The 1060 is actually faster than a 980.
>>55517819
>actually
fuck off shill. We've got no benchmarks yet. As I said: Nvidia is claming something, we cannot verify that claim at this moment.
>>55517791
First off. There's a patch that lowers power usage on the PCI-E slot to under 75w and improves performance by 3-5%. And even then, the people who broke the news about going "over spec" not only wrote it won't do permanent damage, but also wrote that it will only have problems on LGA 775 and AM2+ motherboards or older...ten year old PC parts with DDR2 memory. TPU also came out and clarified that it's a non-issue and that they were wrong about claiming anything about PCI-SIG licensing issues. It's almost like they were payed to report on the 'problem' in the first place.
Secondly, there's no way you didn't already fucking know that.
Thirdly, 5$ has been deposited into your account pajeet.
>>55517872
We have real Fire Strike scores, it confirms the 1060 is faster than a 980
>>55517872
We do have one benchmark. But it's not doing the 1060 any favors.
http://www.3dmark.com/compare/fs/9202637/fs/9188962
>>55517893
None of that is true.
The patch does not fix the power issues with the 480, it STILL goes over the pcie spec.
The problem is that it has actually burned motherboard components because there is too much power going through. The damage is very real.
Even high end motherboards were having problems, which is a real disaster when you pay so much for your motherboard, only to have your new video card damage it.
>>55517872
proof from official nvidia marketing department.
They can't legally lie about this stuff.
RX480 is done it's a fucking joke
>>55517899
>>55517919
480 is faster on that bench though
>>55517899
http://www.3dmark.com/compare/fs/9202637/fs/9188962
>>55515776
>mfw 390x
wew i win
>>55517933
As I said. It's not doing the 1060 any favors.
>>55517925
Good. It will get rid of all those shitty old motherboards. Nobody should be using a 480 on a 10 year old system anyhow.
>>55517945
Okay?
>>55517925
http://www.tomshardware.com/reviews/amd-radeon-rx-480-power-measurements,4622.html
http://www.digitaltrends.com/computing/radeon-rx-480-power-draw-fix/
5 rupees have been removed from your account pajeet.
>>55517932
>It's over AMD is finished.
Make the bars bigger. I am sure retards will believe anything Nvidia shits out.
>>55517641
what the fuck?
glad i stayed far away.
You have to compare verified benchmarks from reviews, not some random Firestrike result.
Here shows how the 1060 beats the 480 really badly:
"3DMark Firestrike Ultra scores 3014 points while the Radeon RX 480 was scoring around 2750 points. The GeForce GTX 1060 scores 11225 points in Firestrike Performance while the Radeon RX 480 was scoring 10800 points in several 3DMark (reference clocks)."
http://www.guru3d.com/news-story/nvidia-geforce-gtx-1060-6-gb-3dmark-performance-leaked.html
>>55517941
how much does your mom pay per kWh?
>>55517951
>the 480 destroys motherboards, this is good
Interesting
>>55517962
Check the pcper review
They clearly show how the 480 is STILL over the spec on both the pcie slot and the 6-pin
Pretty sad that AMD said they would fix it, and it's STILL not fixed.
>>55517925
Nice meme shill, you can go 2.4x over the pcie spec without damaging the motherboard.
>>55517985
i dont really use it that much.
gamed for a couple of months now waiting for some new games to come along
>>55518000
they clearly state that in 4k it draws 0.25A more.
also card is not PCIe certified it doesn't need to be in spec.
>>55517962
>Although the power issue is unfortunate for a company wanting to address the mass market with an affordable solution capable of high-quality VR experiences, AMD certainly didn’t hesitate to address the issue. The company saw the testing results, acknowledged the problem, publicly addressed the issue, and then immediately followed up with a promised fix. Simply awesome, AMD.
Yeah they're totally not in AMD's pocket. Trust me the 1060 is going to come out with 3 gigs of GDDR6 ram and mop the floor with the 480. You saw the slides, 1060 is the only choice for real gamers on a budget.
>>55518013
>Nice meme shill, you can go 2.4x over the pcie spec without damaging the motherboard.
Tell that to the people whose motherboards were killed by a 480
>>55518018
Glad we agree, the 480 is still out of spec and can't even pass pcie certification.
>>55518020
>>55518029
its not out of spec, because there is no spec to be in.
>>55518000
>pcper review
Yep it still isnt fixed the design of the card is fundamentally flawed.
no fucking way i'd run one in my system ever
http://www.pcper.com/reviews/Graphics-Cards/AMD-Radeon-RX-480-Power-Consumption-Concerns-Fixed-1671-Driver/Power-Testing-
>Spending all these hours defending some shitty GPU manufacturer instead of enjoying your purchase.
>>55518024
they probably were miners scum
>>55518029
As an Nvidia fan (I literally have an Nvidia sub-logo tattoo), the flat out lying your doing is making me look like a fucking tool. Cut it the fuck out. The problem has been resolved, but the 1060 is going ot be faster.
Isn't it enough for you to know that the 1060 is going to be faster and more efficient? Do you have to lie about there not being a fix for the 480?
Fuck you dude.
![21-Overview-Wattage_2-635x611[1].png](https://i.imgur.com/T0pMuEc.png "21-Overview-Wattage_2-635x611[1].png")
>>55517925
http://wccftech.com/amds-rx-480-driver-boosts-performance-reduces-power-crimson-1671/
pic related.
after patch it's at 72W from the mainboard PCIE slot.
>>55518063
>2016
>mining on GPU
people don't do that, do they?
>>55518053
They wont be enjoying it when it blows up.
i was on amd's side until they shat this rx480 out with no pci express certification.
Long term this will hurt them and damage components
>>55517130
The comments/reviews are so bad
>>55518049
it even says in that article 71W from the mainboard.
>>55518078
Only if your components suck perhaps. But 2 watts over spec isn't gonna do much to most quality boards.
>>55518049
>http://www.pcper.com/reviews/Graphics-Cards/AMD-Radeon-RX-480-Power-Consumption-Concerns-Fixed-1671-Driver/Power-Testing-
>Should these issues have been discovered and fixed before releasing and selling the Radeon RX 480 to the consumer? Absolutely. Is the power phase weighting adjustment and Compatibility Mode combination enough to make the card at or under spec for ALL workloads? Based on my testing, no. But I do believe that AMD has done it’s best to address the power consumption concerns without a hit to performance, getting the RX 480 to a much more reasonable power situation. I no longer believe that consumers should be worried about the stability of their PCs running the RX 480 with the 16.7.1 driver installed.
FUCKING PCPER IS A FUCKING AMD SHILL REEEE DON'T BUY THEIR PRODUCTS!!!! REEEEEEE REEEEEEEE
ONLY BUY NVIDIA PRODUCTS, AMD PRODUCTS WILL DESTROY YOUR LIFE AND YOUR GF WILL LEAVE YOU!!!
REEEEEE
>>55518072
>>55518092
Thats still over though isnt it.
>>55518107
>Still not certified for pcie
what are they hiding? Theres no excuse for them to release the card like this without certification.
>>55518078
Sure Chicken Little.
>>55518099
God knows how this will affect long term.
i shudder to think
>>55518130
belive me i own a 390x i know the memes they are real
490x is going to be a unreal housefire
>>55518129
Time for bed Timmy. Daddy says so.
>>55518129
no, for PCIE 75W or less is within classification
>>55518147
Have fun with your rx480 degrees
>>55518129
>Thats still over though isnt it.
75W and 5.5A is max.
>>55518156
>>55518164
its still a miniature housefire
NOBODY CARES
>>55518168
>it's not over the SIG spec
>It's still house fire
lel
>>55518164
These are the same shills who were telling everyone GTX 480 at 105c was fine and that the reports of housefires were just
>ignore fermi posts
>report fermi posts
>It's only using 600w of electricity it's fine
>>55518190
>DX11 done right
tfw I fell for the Nvidia meme once before, never again.
>>55518190
yea, but rx480 is as hot as my GTX780 and that shit runs for 3 years now.
>>55518186
>Its not even sig certified
fucking hell how can you defend this shit?
I own a amd card and i cant defend this shit
>>55518204
390x here with proper cooling any gpu can last forever
>>55518206
I own an nvidia card, but don't see the problem.
What's your point?
>>55518206
because I'm against any kind of bullshit that is not needed and all they do is add price for consumers.
>>55517968
PAJEET ON SUICIDE WATCH
>>55518228
okay you are fucking retarded.
certifications are there for a reason its to make sure the product is of a reasonable quality and clearly this isnt.
THEY CANT SELL THIS CARD ITS NOT CERTIFIED and within its first week its a literal housefire
yet ayymd will get away with it because cute red ruby underdog muh ati.
Tell you what ATi was better than this shit.
Truth is all companies just pump out crap and as usual its up to the aib's to make a decent card
>>55518258
>its up to the aib's to make a decent card
>it's up to the add in boards to make a decent card
wew m8
>>55518258
only if certification hasn't retarded as fuck limits I might of agreed with you.
like 6 pin, you can only use 2 of the 3 available +12V lines, because apparently you need 3 ground wires for 3 positive once even when your PSU clearly states that it's ground wires are way above to handle 4 positives.
>>55516904
>the 480 has very little OC headroom
Yeah I guess the 1400Mhz overclock you can get on the reference cooler means nothing, right?
Also the much faster ROPs, shaders and TMUs than the 970 obviously mean nothing.
Please, neck yourself for being so stupid.
>>55518344
>Yeah I guess the 1400Mhz overclock you can get on the reference cooler means nothing, right?
[citation needed]
Every reviewer card tops out at 1350mhz. Since reviewers get cherrypicked golden samples we know the max OC of an off the shelf card will usually be lower.
>>55518344
It's true though, the 480 is a very poor overclocker because it's basically overclocked already. That's also why it has so many power issues and is so unstable.
>>55518258
AMD just fucked up really bad. The 480 is a disaster, but AMD fans are desperately trying to rationalize it all away.
most overclockers say that the biggest issue comes from the poor cooling. Wait for non reference coolers and we'll see if they're right
>>55518258
well I guess you are for communism too.
>>55518072
Check the pcper article, it STILL goes over the power limit on the pcie slot and the 6-pin.
AMD STILL did not fix the 480, it will never get pcie certification because it doesn't pass.
>>55518397
i own intel nvidia and amd stuff none of them care about you stop defending their shit.
>>55518404
they cant fix it until a revision and the aib's havent changed shit so i doubt this will be fixed for a while.
>>55518412
>i own intel nvidia and amd
Cool, me too, wanna hang?
>>55518404
>Check the pcper article, it STILL goes over the power limit on the pcie slot and the 6-pin.
This article?
http://www.pcper.com/reviews/Graphics-Cards/AMD-Radeon-RX-480-Power-Consumption-Concerns-Fixed-1671-Driver
>The compatibility mode, despite the awful name that would leave consumers believing that the card is broken without enabling it, allows AMD to drop power draw on both +12V power sources slightly, by the same relative amount. In my testing, this actually brings the card right down to the 5.5A power draw recommendation for the PEG slot in our 1080p-based testing. I still measured slightly higher draw in Metro: Last Light at 4K, but 5.75A is at least in the right ballpark.
>>55518067
I'm glad you're an nvidia fan, but at least do some research.
Here is a direct quote from pcper article:
" With the original launch driver we saw the PEG slot pulling 6.8A or more, with the 6-pin pulling closer to 6.6A. On 16.7.1 the PEG slot draw rate drops to 6.1-6.2A. Again, that is still above the 5.5A rated maximum for the slot, but the drop is significant."
The 480 is STILL over the limit.
AMD did not fix it sadly and I doubt they ever will because the 480 power draw is fundamentally flawed.
>>55518426
>implying normal people will ever update their drivers or enable compat mode
>>55518426
>I still measured slightly higher draw in Metro: Last Light at 4K, but 5.75A is at least in the right ballpark.
Thanks for proving my case.
It's STILL over the limit.
The 480 cannot pass pcie certification even with the new patch.
>>55518412
>they cant fix it until a revision and the aib's havent changed shit so i doubt this will be fixed for a while.
Yeah it really is a fundamental problem with the 480. People are going to be very disappointed when the 3rd party boards still have this problem.
>>55518438
>The 480 cannot pass pcie certification even with the new patch.
so?
>>55518458
If you have to ask you'll never know.
>>55517395
well we do have live demo of nvidia housefire in action tho
https://www.youtube.com/watch?v=sRo-1VFMcbc
kek
>>55518471
This is most counter productive saying I have ever heard.
>>55518486
>implying there's any hope of anything productive at this point
>>55518164
>posts combined wattage
>posts one rails voltage
not realising that people here have a decent understanding of ohm law..
>>55517237
>asrock
Its amazing it didnt happen sooner
>>55518451
normal people wont care.
i think the other polaris stuff will all have problems doesnt give me hope for vega
>>55518513
voltage is max 12v it can go lower allowing to have more AMPS for the same wattage, you fucking retard.
>>55518521
because surely 480 has so many problems
lets see them all
>black screens
>dp ports fried
>stuttering about 60 hz
>drivers randomly fried cards
>constant tearing on 144hz
>cant use vive
>constantly lying about the card capabilities
oh wait no sorry all those are nvidia problems..
>>55517490
I've owned both a 970 and a 390, the 390 peforms better in games that the 970, by 10%-15%
>>55518544
I know.
can belive people drop $1200aud here for those turds
>>55518523
we are talking about 18 gauge cables here unless 480 can make them reach 140c its moot all of this shit you keep saying
12v only provides the 66 watt
3.3v provides the 9 remain watt its also the one that dynamicly balance the load between the two cables obviously you dont really know anything about electricity
back to your cave homo nvidious
>>55515776
I cri everitiem.
>>55518616
TRIGGERED
>>55515776
Holy fuck so many paid novidia shills fuck off
>>55518719
>facts are shilling
wew lad
>>55517186
The only difference between the 6 and 8 pin connectors is 2 additional grounds. The pcie power cables are vastly underrated for total power output. Much like pcie 16x used to only be rated for 25 watts. Seriously, it's a non issue unless you're using some cheap Chinese house fire psu, or a 10 year old mobo. In which case, you need a lot more than a cheap gpu upgrade to begin with. Also, so far in tests the only issues with said old ass mobos has been flat out system crashes. If the system can't support the power draw, it won't. Simple as that.
>>55519602
480 has already killed many mobos
>>55517354
>MUH 3.5GB!!!
kek everytime
have gtx 970 and never suffered anything because of the 0.5 gb being slower
>>55519611
Source: asshole
Have you found any after the driver update, by any chance
>when Nvidia shills can't attack AMD performance anymore so they resort to this retarded shit
Lel. If something consumes more power than the source can provide it SHUTS DOWN, not burns the source.
>>55517264
>implying spec limits (not physical limits, mind you) are set at the ragged edge of reliability
>>55517355
When you buy a 970 you will receive awful coil whine.
>>55515776
How much longer until AMD bites the dust?
While AMD's power design isn't flawed, nVidia simply does a lot more software side to properly implement voltage control. While AMD's ASIC power is relatively set, nVidia's is rather dynamic. In combination with the GoFlo process, the AMD's appears to be less efficient than nVidia's cards.
Another issue is that AMD strongly prefers to build an hardware infrastructure than a software one. While it provides strong advantages in some areas such as VCE, when AMD switched from VLIW4 they designed wavefront ICs, dynamic hardware scheduling, TrueAudio DSP, UVD, ACE/HWS which takes up room/power and thus contributes to a fair chunk of the power usage on GCN cards. A 110W ASIC is still incredibly efficient; loads in compatibility mode is only a bit more than 100W. However, to the average consumer, GCN's technology still isn't implemented correctly on the front-end - a bloated streaming utility, limited overclocking utilities, PRT/memory compression is poorly implemented, etc. AMD put a lot of hardware on the GCN ASIC knowing that they would have to plan ahead due to not having R&D to catch up with the software infrastructure, being way ahead of its time with UMA (for HSA), async, bare metal compute, and BRTF, and only now is nVidia catching up. It's a shame that AMD is not a financial equal with nVidia because subtly when GCN was first introduced in 2011 it could've been called a 4850 slightly out of it's time period.
>>55517501
>480
>fan @100%
>no temps
>>55519899
AMD is truly finished. They fucked up too bad this time.
>>55519848
Not true, mobos are not designed to take more power than spec. What happens beyond spec is totally unpredictable. 480 could shut down your motherboard or burn the pins making the pcie unusable. Some guy even had his sound stop working.
>>55520027
How so?
You retards do when when amd fails and then goes away, your precious nvidia will go through the roof with costs.
I am sure AMD has had an impact on the price of nvidia cards.
I am unsure why you retards lack this logically thinking. How would you enjoy your card if nvidia decided to increase every card by $100-200.
>>55520157
Nothing you can do, AMD is too incompetent to compete.
>>55517082
No way near as bad as the geforce FX series which the ATI 9800pro trashed.
Really forces you to contemplate
>>55520157
Are you telling me to buy inferior products to support competition? If they can't compete, they deserve to perish.
>>55520201
I would have loved to see /g/ in 2003. The GeforceFX was truly an embarrassing product. Particularly the DirectX 9 performance.
But I bet the nvidiots would try to say that DirectX 9 isn't a sure future bet or something.
>>55517985
I'm running two in crossfire. That costs $0.05 an hour, or, at three hours a day, $50 a year. I can afford that.
>>55517024
on dx12 when properly utilized and implemented... yea, the cpu bottlenecks the gpu.
>>55520239
lol why did they lie so much?
>>55520239
Yeah. It really makes me contemplate why they compared it to a 950 rather than a 960.
>>55516721
YoU CaN't mAkE tHiS uP
>>55517140
power bill is a non issues in any first world country that's not complete shit.
however the room you are playing in turning into a sauna after a while, that's an issue that is solved by less watts.
also, quieter cards.
>>55516866
no oc is about 5% slower, oc is about 10% faster... where is this vastly you are talking about?
>>55520239
How the hell did AMD push a 950 to use 140w? Even under full torture test the worst is around 120w and average usage is around 100w.
https://www.techpowerup.com/reviews/EVGA/GTX_950_SSC/28.html
That's not even a very efficient version. Here's the a Gigabyte FOC version which tops out around 100w.
>>55520406
Whoops! Forgot the link.
https://www.techpowerup.com/reviews/Gigabyte/GTX_950_OC/28.html
>>55520239
>amd internal lab testing
>intel cpu
>>55517977
Otherwise believable, but the fact that Fire Strike recognizes the GPU as a GTX 1060 gives it away as fake.
>>55517395
>Yes but the test is ongoing, hes trying them out over a week.
Link to start up test is here:
https://youtu.be/RiPtzcqIZ4E
>>55517473
till you can come up with just 1 motherboard, the 480 is not damaging them either, and dont get me wrong, part of the reason im holding off on a 480 is motherboard related, mine does have issue with pcie overdraw, found this out thanks to an nvidia card years back.
>>55517501
Serious question. Why would the cpu be 100% utilization on the lower clocked 970?
>>55520239
that's from the time before Nvidia unveiled up to 2.1Ghz Pascal, at that moment AMD collectively shat their pants.
>>55517501
Is this guy running on lowest settings? His game looks like shit.
>>55515776
Hello sir this /g/ an registered AMD Safe Space please discuss on /v/. Remember nvidia not support old card. Remember nvidia hate freedom and lie about 3.5 gb. Buy AMD to support gaming.
>>55519611
>>55519737
At this point with how many times he's repeated himself. I just picture appu regurgitating "thank you come again!" it's quite comical.
But seriously, pcper stated in their "our test Vs toms" video that pcie power cables are 4x over built for their specifications.
