>AMD couldn't afford to put 2 extra power pins on the RX 480
>all they had to do was put 2 more pins on and not lie
Would have been a good launch.
AMD literally bankrupt and finished?
Has the 480 been cooking some pcie slots?
>>55373033
They should have spun the 8 pin as marketing tool "look how efficient our card is at stock but if you really want to push it we even gave you a 8 pin power connector for free"
>>55373045
They still have Vega and Zen but this launch is a prime example of why their reputation is the way it is. If they simply said that the card draws up to 190w and put a 8-pin on it, it would still be a pretty good card and none of this drama would have happened.
>>55373068
"8 cores that'll shit all over Sandy Bridge"
>>55373081
>>55373081
>raises the performance bar
This is the kind of shit I'm talking about. Their marketing is fucking bullshit.
The 480 would still be a good card with an 8-pin but instead they chose to risk it and put a 6-pin on it ruining their reputation even more.
6 pin should also be able to handle 150 watt.
8 pin just has more earth pins which is extra safety but 6 pin should be able to do the same.
>>55373112
You're right but standards exist for a reason.
>>55373009
Its not about power pins. Connector replacement wont change anything. Its pcb/vrm engineer error which lead to wrong power distribution between rails.
Doesn't this meme tier 'motherboard fires' shit coe from the fact that:
1)Testers are using reference
2)One faggot tester was spotted using a '4790k' when it was a AMD mobo on his youtube <-- needs clarification
3)This shit only draws 100w when OC'd? Otherwise spikes to 80-90
4)Faggots forget the 960/750ti pulled as much through the PCI slot
5) It's ok when Nvidia does it!!!
6) Your shilling isn't stopping me from buying a Sapphire Nitro 480 when it comes out.
>>55373057
Yep
>>55373162
There is definetly something wrong with the reference model. The 960 strix was tested and the problem is not comparable to the problem on 480.
>Your shilling isn't stopping me from buying a Sapphire Nitro 480 when it comes out.
And it shouldn't since that card wont have this problem since Sapphire isn't as dumb as AMD when it comes to making a videocard.
>>55373162
>ashes of the singularity
AMD has fallen really low.
>>55373162
All the facts are here: http://www.pcper.com/reviews/Graphics-Cards/Power-Consumption-Concerns-Radeon-RX-480
The 960 has no issues, it's only the 480
>>55373162
>Your shilling isn't stopping me from buying a Sapphire Nitro 480 when it comes out.
Chip is great, ref. design is dog shit
>>55373209
>>55373162
at least all those 200 players can buy two 480 and get 1080 performance.
The worst part about it is that people think it doesn't even have 6pin power and so pulls everything from the mobo
>>55373199
>There is definitely something wrong with the reference model.
But no news of any housefires as of yet.
Many Reputable PSU testers will clarify that the 6 pin CAN deliver more than 75w (it just avoids it)
All news I find outside of this meme shithole nip board suggest that this 'issue' is nothing really to give two fucks about, just some shills on here and the odd damage control tester.
Fuck even TTL hasn't slandered the 480x as far as I've seen, and he's the biggest Nvidia stealth shill you can find.
>>55373249
> 6 pin CAN deliver more than 75w
It can deliver up to 150w, just like an 8 pin.
>>55373009
Why does AMD always fuck up the reference design?
Nvidia isnt much better, but how hard can it be to make a decent reference card?
>>55373209
It's the one game where AMD has a massive advantage over Nvidia. Unfortunately it's a garbage game that no ones plays.
>>55373009
They could, but if they would then it would be obvious for everyone that it is NOT a 150W card. As it is, we only found out after it was released.
>>55373237
>tfw wc3 has more players than their peak
>>55373081
AMD is the Donald Trump of tech.
>>55373282
>>55373291
At least is doing better than Evolve.
>>55373162
>2)One faggot tester was spotted using a '4790k' when it was a AMD mobo on his youtube <-- needs clarification
what
>>55373209
>>55373237
You Novidya fags are really assblasted a set of cheap cards slaps the shit out of your top end card?
I concede Nvidia is top of the market for high end, no quarrel with admitting that, but fuck sake lads, you know something is severely wrong with Nvidia if 2 "housefire/no driver/ poo in loo bargain basement" cards can pull ahead of a card of that caliber, it shouldn't happen at this price range. That's like have two 960s come roaring ahead of a 980ti for 1/3rd the price
>Mah 1 game
It's climbing and is currently 7 of the latest titles games I have seen where a 480 is slap bang ahead or inbetween a 970/980, exactly where it was predicted.
>Crossfire is shit
Could never be as bad as SLi, and if AMD put as much effort into Multi GPU as they did to push for better APIs, then Nvidia have something to fear
>All the good games have Nvidia
Hairworks and Gameworks do not make good games
>Enjoy Motherboard Death and no drivers
I don't buy Asrock so I'll be fine, and Drivers matured like a dream for 290/x users
>>55373284
Would anyone give a shit if it wasn't?
>>55373329
>All those "If"
Good going shills, just wait.
It will get better I promise.
I wonder why Nvidia is pushing their 1060 release. They're either scared of customers going red cuz they don't have a card to rival 480 in mid tier range or they simply are relentless and want to gun down AMD like a dog. However the fact that UK prices of 970 gtx went back up to levels of 480 (I could buy new non-ref top tier 970 like MSI or EVGA for £199 last week) mean 480 probably isn't doing as great as you think.
>>55373272
why don't AMD and Nvidia rush the aftermarket coolers? who the hell benefits from selling reference cards?
>>55373322
Heard a rumour it was Tech of Tomorrow, on his first vid, someone said it was a AMD mobo, yet he says in the specs - 4790k. Someone might be able to support or debunk perhaps?
>>55373162
>2)One faggot tester was spotted using a '4790k' when it was a AMD mobo on his youtube
dude, what?
>>55373342
>480 probably isn't doing as great as you think
OCUK reported that they ran out of RX 480s (something like 2-3k cards) and that 4GB 480 production was getting halted to favour the 8GB model.
>>55373342
>I wonder why Nvidia is pushing their 1060 release
To cannibalize the 480 sales of course, what else could it be? To inevitably gimp the shit out of the 7xx/9xx series completely and force people to upgrade again notwithstanding.
>>55373342
The 480 sold well in the UK, it might be Nvidia closing the price range to lock the market down like the transition they did with the 780ti to the 970/980, Nvidia typically group it all back up.
>Before 1060
""But the 970, it's cheap and 'as' good as 480, we swear, ignore the 3.5!!"
>1060 gets announced and leaked
"you could get a 970 for cheap, oh wait no.... 1060 is new, you have to make a gamble now anon"
>>55373337
So you're saying the 290x didn't mature and slap the 780ti around?
may be a few if's as I'm not Mystic Meg, but one thing is for certain, when the Maxwell launched, they utterly gimped the 700 series.
>>55373009
At this point i'm ready to buy 1080. Losing faith in vega.
>>55373342
>We know that about 85% of the GPU market buys GPUs between $100 and $300 US. The other 15 spend more than that or less than that. So for us, a department that wants to get our technology into the hands of as many people as possible, a GPU like the RX 480 seems pretty damn smart to me.
https://www.reddit.com/r/pcmasterrace/comments/4qfy9d/i_work_at_amd_the_time_has_come_to_ama_about/d4smj8k
as long as Nvidia doesn't have a monopoly, I guess
>>55373081
>my i5 2500 is still superior to any amd cpu
>>55373162
>Faggots forget the 960/750ti pulled as much through the PCI slot
Spikes are better than constant draw.
60 watts average vs 80
> GTX 1070s are $699+
> GTX 980s are still $599+
> RX 480s are $459+
Being a kangaroo is fucking hell
>>55373488
But dollar is worthless so 600 aussie dollar is like 134 euros
>>55373482
>literally 5w out of spec
rip biostar users
>>55373009
The funny thing is that the two extra pins on the 8 pin don't add another 12v line, all three 12v lines are on the 6 pin, but if one of those pins (the middle one) shouldn't be connected on the board itself. One of the extra pins on the 8 pin is a sense pin that would then allow the board to realize that it has an 8 pin connected and can draw 150w.
They could literally just draw more on the 6 pin. Honestly. Like 150/3 is 50, so the two lines should be able to handle at LEAST 100w instead of the 75w max for a 6 pin.
>>55373009
Because the only thing 8-pin adds os more earth pins and six pins can easily supply enough power. And while everyone has a six pin connector, not everyone has an 8 pin connector. Also drawing 8 watts to much from pcie isnt a problem except if you have 4 cards and a really shit motherboard. Litterally nvidia grasping straws trying to banter this great price performance card
>>55373515
Its 80watts on the 12v rail which is limited to 66
>>55373512
What's your point?
![0024_gossipgirl-151015700[2].jpg](https://i.imgur.com/NnDXFjp.jpg "0024_gossipgirl-151015700[2].jpg")
>>55373162
>this is the average AMDrone
>>55373249
>nothing really to give two fucks about
So apparently your internet blog reading makes you know better than AMD themselves. They just stated publicly that they're going to try to fix this hazard and hopefully it can be done in driver so they don't have to recall your card?
>>55373593
>This is the average Nvidia Shill
So you reckon custom PCBs but third party companies should fix the issue? Not that I was ever going to buy reference anyway.
>>55373683
No idea yet, I would wait for the reviews on those, right now I wouldn't buy any 480 because all of them have the issue
>>55373643
"If some of the data is to be believed, these cards are exceeding 150W total at times, which would mean there is either something causing them to run in the wrong power state, or they are just outright exeeding their power limit and need to be throttled back. As we don't do per-rail testing I don't have anything meaningful to add at this second, but it will be very interesting to see how AMD responds next week."
>these cards are exceeding 150w total at all times
Seems more like a total power issue, not a Mobo housefires, or as suggested "wrong power state"
"As we don't do per-rail testing I don't have anything meaningful to add at this second"
>"I got my moneys worth from this, and I don't want to do more work than I have to, besides now I can see how much money Nvidia will pay me for a decent review"
Like to add here - 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).
NVIDIA LOST ITS ONLY EDGE IN THE MEME WARS
>>55373788
> exceeding 150W total at times
you then type
> exceeding 150w total at all times
Which is it?
>>55373788
>more like a total power issue, not a Mobo housefires
"total power issue" is what people have been talking about all the time retard. For a card that will run for maybe 10,000 hours it's definitely a hazard
>we can adjust the GPU's tuning via software in order to resolve this issue.
This means making the card slower you retard. Not waiving a magic wand
>>55373830
They suggested the card is running flat out at all times
I meant implying with my post, and 'if' it was running 150w all the time, then it would show on the idle temps. If it was meant as in, "both 6 pin and PCI" then it's in the safe zone, so whats the issue?
>>55373841
No you bellend, people have been losing there fucking shit with crap memes about how the Mobo would explode in a ball of fire, and it certainly is not a fucking hazard as the 750ti would pump 75w+ through the PCI under a full load you absolute mongolian fuck pig.
And if the card 'is' stuck in a high power state under load and is pulling to much from the PCI, then yes they can tweak it with a driver.
>>55373930
There are already like 4 reports of mobos dying and this is only day 2, it has serious problems.
That's not even mentioning the one mobo that wouldn't even run the card without crashing because it was using too much power.
They let the hype train run away from them and now they've botched the launch with some banal driver issue, which exacerbates the known problems AMD has had in the past.
Now both NVIDIA and AMD have fucked up the launches of their most important consumer products.
I get the feeling NVIDIA pays too much attention on Tesla and AMD too much on the console market. Once again pc users are getting screwed...
>>55373930
>750ti would pump 75w+ through the PCI under a full load
1. I know you have the mentality of a 13 year old console-fanboy, but the "other side" being bad doesn't make "your side" less bad
2. Wrong. It's been said like 100 times already that "peaks" and "sustained" are completely different matters. But you just stick your fingers in your ears (or even worse doesn't even understand this basic concept)
>>55373930
It shouldn't burn your mobo no matter what "state" it's in retard. Something needs to be dialed down, probably with worse performance as a result
>>55373969
How exactly did nVidia "fuck up the launch"
>>55373958
Yeah 2 of which have been debunked as shit posters
Not saying it hasn't or will not happen, but one was saying "480 killed my board" and when he uploaded the pic, it was the 24 pin that had a burn, that's not caused by the GPU
Another has been outed as a troll. Source for the other 2?
>>55373971
http://www.guru3d.com/articles_pages/palit_geforce_gtx_750_ti_stormx_dual_review,5.html
Never said that AMD is without fault, but this is blown way out of proportion
>>55374002
Paper Launch
>>55373930
this guy put a rx 480 in an am2 mobo and it would constantly power off.
https://www.youtube.com/watch?v=rhjC_8ai7QA
he put a 980 ti (a fucking 980 ti) in the same mobo and it worked just fine.
>>55374106
That's a third party card retard. The review also doesn't even specify if it's max or avg
>>55373342
>480 probably isn't doing as great as you think.
I just checked Ebay, amazon, and newegg. Sapphire RX 480s sold out across the board. Seems like it's doing just fine.
>>55374124
How exactly is having greater demand than expected "fucking up"?
>>55374189
>sold out meme
Doesn't mean anything if you don't know how many they shipped retard
>>55374157
You know something is wrong when a 480 has more power issues than a 980 Ti
>>55374157
Use a 980ti with a 8 pin connector
Solved the puzzle, but yeah it hasn't solved the power state issue so far, wait for Tuesday for driver tweak
>>55374199
>Boss - "we hold a 85% share of the Market and our due date is soon, where are our Cards?"
"Yield problems sir"
>>55374186
Not only that it is not even a proper measurment. Looks like he is using a kill-a-watt wall wart to measure it.
>>55373009
>people with cheap mobos and psus complaining when they die under load
>>55373249
The issue has nothing to do with power draw from the 6 pin. It's drawing too much from the motherboard
>>55375201
>we want the budget audience
>fucking idiots with their cheap hardware
B R A V O A T I
T
I
>>55375335
It's actually a elaborate plan to coax cheap users in to upgrading to a zen/vega combo.
>>55374216
Microcenter only got a handful of GTX 1080's in (like 3 lol) and they sold out. Overclockers got several 1000 480's in and sold out. You do the math.
>>55374216
Oh and AMD sold more in the last few days than Nvidia sold in a month.
>Not buying the AIB Sapphire Nitro Model
>>55373081
Even that would be a good thing. The IPC of sandy bridge, with 8+ cores at an affordable price would be great.
>>55375474
Is that only the single 8 pin connector I see at the back? Putting it at the back is a pain as it extends it slightly more for the connector and cable. Not so great for some micro atx cases.
>>55375423
>uncomfirmed numbers from a single store, not even the same store
lel
>>55375435
1. Citation needed
2. the 480 is a budget card. If they sell anything less than 5X of what nVidia sells 1070/80's it's a complete disaster
>>55375549
I read a post that some store sold 1000 1070/1080's over the course of a month. The Microcenter one was 3 cards in a day. I am not sure where I saw the 1000 Nvidiia cards number either. The Overclockers one they stated they shifted about 2-3k cards in the last few days. I CBA to give citations on a shitty Chinese board I don't give a fuck about. Google it yourself.
>>55375621
>I read a post that some store
TOP
FUCKING
LEL
This is literally /pol/level research.
>I CBA to give citations
Retard, citations doesn't change that it's "uncomfirmed numbers from a single store, not even the same store"
>>55373125
They do, but everyone ignores. If you measure effectively any graphics card on the market they all draw more than the rated value on motherboards. It's pretty normal, just doesn't often come to light.
>>55375655
Of course. I could not prove it even with those links. But considering how many cards they got from Nvidia and how many they got from AMD on actual launch one can easily speculate.
They don't call Nvidia's launches a paperlaunch for nothing.
>>55373081
Still waiting for the bar to be raised...
The Bulldozer/Piledriver CPUs have some good features but the performance is not one of them.
>>55374048
>Not saying it hasn't or will not happen, but one was saying "480 killed my board" and when he uploaded the pic, it was the 24 pin that had a burn, that's not caused by the GPU
Are you computer illiterate? PCI-E slots get power from the 24 pin. If draw from PCI-E slots is too high, 24 pin will be damaged assuming it's the weakest link.
>>55373482
Spikes aren't better than draw when some of the spikes went almost 150W over spec. Look at the graphs. Sure the average is low but drawing double the rated for a split second is certainly not better than drawing 5W over spec constantly.
>>55375655
being this edgy about gaming. virgin.
>>55375686
Anyone overclocking one of those 750Ti's without a power connector would have gone over the PCIe spec but you don't hear too many stories about fried motherboards there.
>>55375729
Are YOU computer illiterate? The 24-pin can deliver FAR more power than what the 480 pulls on top of spec. No chance a 24-pin would pop from something drawing too much from PCI-E. It can handle 4x cards in there no problem, a little extra on one port won't kill it. The port might get damaged, but certainly not the 24-pin connection or cable.
>>55375739
Spikes are better because none of these components involved even know the spikes are happening. All they can "feel" is the heat generated by resistance. Which, to be fair, is higher when the card spikes and amperage goes up.
But all that means is that power consumption is not linear. Higher power draw has more transmission losses and creates higher averages. BTW the 480 spikes to 155W+ too. That's normal for shitty low end cards without a power connector. It's just that most of them manage to average it out around 40-60W.
>>55375793
I don't think you can overclock it enough without a 6 pin to draw more than 70W. The reference designs have pathetic VRM which would probably burn out if you tried to draw as much as a 480.
>>55373162
750 Ti never pulled that much. It would cook itself before doing that. I've tested a card w/o additional power connector. Overclocked to ~1450 and overvolted to 1.21v, not even then it pulled more than ~70 watts.
>>55375841
Boards that support 4xcards are high end boards which often have additional power connector on the board for PCI-E slots. It also depends on how well the connection is made, because the board only has 2 12VDC pins. If one of them isn't tight enough, it'll burn when you try to pull too much power. The RX480 takes about as much as 2 high end graphics cards from the slot, so it's like you're running a SLI setup.
Also, the problem seems to be very rare which just gives you less credibility because you're trying to claim that everything should be exploding just because some things are. I bet that none of these people who have a damaged board even have compatibility with SLI.
My board for example only has 1x16x PCI-E slot which takes all the lanes from CPU. It does not support SLI at all.
>>55375694
>But considering how many cards they [allegedly] got from Nvidia [and how many of those they allegedly sold] and how many [a completely different store with different customers allegedly] got from AMD [and how many of those they allegedly sold] on actual launch one can [pointlessly] speculate.
fixed
>480 isn't selling well
>>55375474
Are there actually Sapphire employees on here or something.
>>55376047
>a meme store with 0.1% market share
Are you really THIS desperate for good news?
A different plug would make any difference. The Polaris architecture simply that great of an improvement over Hawaii. Polaris should be doing Nano frame rates not r9 290 level.
>>55375739
both the card and mobo is designed for these spikes, what the mobo pci-e isnt designed for is costant draw of 80W, while the max is 66W. I dont understand were u faggot picked up this meme that constant overdraw from pcie is totally safe and A-OK, ig 80 is safe so is 100 rought, maybe 150W, maybe we should just remove the pins and have cards draw 200W from pci-e slot ,since its constants its safe!
When will AMD go bankrupt so other companies will start to improve ?
>>55376126
>A meme store that is up front about their sales figures sells 1000+ cards within 1-2 days vs a big box shifter like Newegg, Amazon who must be selling them by the bucketload if that 0.1% is any indicator.
Does anyone know of any games other than AoTS that supports cross brand multi-GPU setups yet? I'd like to use one alongside my 290 until I sell it on and get a VEGA card later on.
>>55373313
Definitely not. Trump is a branding genius. His "business" is literally just his brand and he made billions off of it. There's a reason why someone completely new to politics single-handily destroyed the strongest presidential lineup in republican history.
>all those 1000's of cucks that bough RX480
>all the pcie slot burning
>all those nvidia ceo's clapping hands while they are gaining 1000's of angry customers because amd destroyed there mobos
>>55376296
Overclockers is an old technology forum that opened their own store. Their customers are completely different from the majority retard. Just like 50% cards THEY sell is probably 350$+ whereas it's only 10% of total cards sold
>>55376296
Yeah some guy bought 48 RXs for mining. Its 290x/7970 once again.
>>55376358
>strongest presidential lineup in republican history.
HAHAHAHAHAHAHAHAHAHAHAHAHAHAHAAHHAAHAHAHAHHAAHAHAHAHAHAHAHAH
>trumpfag education
He's also permanently 5% behind one of the most hated democrat candidates ever and 10% behind some senile old guy. Quite an accomplishment
>>55376426
Shitposting to the max
>>>/pol/
And educate yourself
>>55376366
This.
I'm sure you can fool yourself that you're "with the good guys" by being an AMDcuck if you just get a little worse performance and compability year after year. But when they start actually destroying your computer the illusion will shatter
>>55376454
Educate myself about how you think someone republicans smeared publicly because they hate him so much (Cruz) is part of "strongest lineup in history"?
Educate myself by reading the polling numbers i read yesterday again?
>>55376220
Pajeet please download our latest "shilling with english" audio lessons. This is your final warning
>>55376552
lel
>>55373660
>0.2 TFLOPS DP
This is fake, right?
pls b fake.
>>55376948
>gaming card
>DP important
>>55377084
Doesn't matter.
0.2 TFLOPS is pathetic as fuck, even for a "gaymen card"
>>55377115
No, it's irrelevant.
Might as well say a plane is shit because it can only taxi at 30kph
>>55373009
> millions of dead mobos thanks to oced 750tis and 960 strix
why didn't AMD learn from Nvidias mistake?
>>55377236
What? I have a 960 strix, please explain further
>>55373313
you should consider suicide for the betterment of humanity.
>>55376426
>HAHAHAHAHAHAHAHAHAHAHAHAHAHAHAAHHAAHAHAHAHHAAHAHAHAHAHAHAHAH
You think that's a joke? It's not sincrely it was the most diverse and strongest lineup of a lot of candidates. Almost all of them were revered by the GOP.
> He's also permanently 5% behind one of the most hated democrat candidates ever and 10% behind some senile old guy. Quite an accomplishment
You won't be saying that when he wins in a landslide. Think about this. The biggest thing that Hillary Clinton has on him is his rhetoric and actions this political season. Something that he has never exhibited before and can easily change. Hillary Clinton CANNOT change her crookedness. More information on her scandals will definitely come out. She cannot escape confirmation bias, but Trump can. He's already won. He just has to make sure that Hillary is going to bank everything on the "crazy, racist" first.
>>55373482
Clearly not an electrical engineer. Motherboards can handle the sustained power of 5 W above specification. PCI-E power is just a standard, it's not hardware. You'd have to have a really cheap mobo from 2005 or something for this GPU to fry it.
Spikes up to 270 W on the PCI-E slot, on the other hand, those can damage the slot.
So do me a favour and end your life, shill.
>>55378443
>Clearly not an electrical engineer.
No, you clearly don't understand electronics. Spikes are without a doubt better than constant draw. Even if you don't know you should be able to use inductive reasoning. Plenty of cards spike above it, in fact most. Yet AMD is the first card with its constant draw above spec. How many problems have those cards had (in the hands of consumers) with PCI-E slots in the hands of consumers? Now how many problems are AMD users having NOW.
>>55373345
Pins wouldn't make contact you fucking retard
>>55378443
>electrical engineer
>Thinks that voltage spikes are more dangerous than overcurrent
You've clearly never experienced a housefire from the wiring in the house from 5 toasters in one wall socket.
>>55373081
Shit like this is what makes me know Zen will flop
People call Nvidia jews but AMD has never kept their word on a single thing
>>55373313
try posting this on facebook next time, where you belong, you'd probably get more likes and upvotes there
>>55373162
>3)This shit only draws 100w when OC'd? Otherwise spikes to 80-90
That's not a spike. That's average. Asus Strix 960 has average at 74W - so within limits.
>>55375523
The card is already short. You have a problem with it on the back I have a problem with it on the side.
>>55378443
The problem is that the card is drawing 80 watts from the 12v line which is rated at 5.5A or 66watts.
And thats 20% more than the specs
>>55375474
Two of these would look good in my case.
>>55378831
>>55378797
>>55378443
>>55378623
Why are you guys still talking about this? the will will be out on the 5th.
guess this is some firmware or component problem, AMD would've never gotten the PCIe certification with 10% power usege over the limit
>>55376454
>Kek educate itself meme
T. Man with no argument
>>55378861
I don't give a shit about this card, i'm calling a retard out that thinks an AMD mobo that's pinless would make contact with an intel CPU that is also pinless.
Is it that hard to read
>>55379104
wtf, yea that is retarded.
>>55377214
But you do know the Titan X had a predecessor that had 1:3 DP Tflops and was quite comfy doing a few 'non-gaymen' tasks right?
Like I know Shills like to pretend that Nvidia never cuts features off there cards, but holy fuck:
>Async Hardware
>0.5gb of VRAM
>Driver support for 1 gen old
>Hardware so it can save power on Maxwell
>Actual fucking GPU screw fittings
When a card that came out 5 year ago has more DP TFlops than your $1000 card, and you throw "it was only meant for gaymen!!" around, you know you sound like a total tool, right?
>>55378623
It was apparently a AMD Mobo as well you absolute bellwipe, why the fuck would I even suggest a AMD cpu in a Intel Mobo? This your first time on the internet when you literally need someone to explain EVERY fucking detail? Honestly end your life.
>>55379104
Don't be too harsh, maybe the poster expected you to know this would include a Intel to Intel, and AMD to AMD....
Also ToT plz leave
>>55373313
Thanks for triggering the /pol/tards. I enjoyed it.
>>55378840
White backplate!!!
This going to look sick in my Phanteks Enthoo Luxe
>>55379808
He's right tho, gaming doesn't use DP. If AMD made their shit cheaper by cutting out DP, I'd say that's a good thing.
>>55380326
So why was the Original Titan a nice 'hybrid card'? Used to hear tonnes of cunts saying how they used it for editing and video rendering etc, maybe this was just good shilling from fanboys, but cutting features rather than refining them for the SAME price is Jew tier.
>>55380370
I don't fucking know? I think it's a piece of overpriced shit. Maybe it shouldn't exist at all. How about that?
>>55374002
Saying card runs cool when it was actually playing a video
Overheating issues and can't even keep base clock
>>55377846
>It's not sincrely it was the most diverse and strongest lineup of a lot of candidates. Almost all of them were revered by the GOP.
They might of had been the biggest suckers of the GOP establishment but just because the party establishment loves a canidate doesn't mean the person will vote for the canidate
>>55378882
There are zero problems with the strix, see this:
http://www.pcper.com/reviews/Graphics-Cards/Power-Consumption-Concerns-Radeon-RX-480/Evaluating-ASUS-GTX-960-Strix
It has been tested extensively.
The 480 is the only card that has this fuckup, someone at AMD is going to get fired
>>55378443
You are very wrong, read this:
http://www.pcper.com/reviews/Graphics-Cards/Power-Consumption-Concerns-Radeon-RX-480
>>55373344
they do since reference coolers are cheap as fuck to make
Anything stopping non reference cards from being 8 pin?
>>55373081
Delaying bulldozer was the evilist thing amd has ever done.
>>55373342
>be nvidia
>make good cards but overpriced and under specced
>sell like hotcakes
its like a guy with a 12" dick he doesnt even have to try
AMD is just standing around furiously masterbating hoping everyone will notice their 6" chode
>>55382664
>Anything stopping non reference cards from being 8 pin?
nope /g/ is freaking out for no reason who the fuck buys reference cards at stock clocks at launch?
last time i did that was 7 years ago with a 285 and it was a shit
>>55373081
>>55373009
They COULD have, but then they'd be unable to pretend it's a sub-150w card, and that was their big point.
Can somebody please tell me what the problem is?
>>55382978
>>55377846
>strongest lineup of a lot of candidates
>Hamburgler
>Old man Munster (The 1st Cuban)
>Old rag who ran HP into the ground
>Paul
>Mr. Wisconsin.
>Obama knows exactly whats he doing (The 2nd Cuban)
>Dr Benadryl Carson.
>Oompa Loompa
what a ragtag band of heroic brothers
>>55383338
RX 480 will kill your motherboard
>>55373033
honestly they didn't even need the pins
just make the card draw the remaining power from the pcie plug since they can handle up to around 300 watts nicely
>>55379936
I expect the poster to have at least entry level knowledge about the hardware he's bitching about in a technology oriented board, wich isn't that much to ask. What you wrote passes as engrishh at best too.
>>55383676
But only if it's an old motherboard and even then only rarely*
>>55373033
Except for fucking complete lack of 8.1 drivers
is it time to go back to intel/nvidia combos again?
>>55383853
You could wait for the customs, since Sapphire knows how to make a card better than AMD does.
>>55373162
>3)This shit only draws 100w when OC'd? Otherwise spikes to 80-90
from the pcie connector on the board maybe
>4)Faggots forget the 960/750ti pulled as much through the PCI slot
the 960 had spikes above 75 which is ok, the 480 is almost always at or above 75 which is not ok.
drawing power from the the extra 6 pin connector would have been perfectly fine.
the 750ti at least has the excuse that it doesn't have external power. the 480 has an extra connector that it should have been using instead of going through the board which is much more dangerous.
anyways the 750ti was much better at keeping it's power below 75 watts on average, it's only really the peaks when it tends to go over.
6) Your shilling isn't stopping me from buying a Sapphire Nitro 480 when it comes out.
good, you couldn't pay me to go back to amd, but somebody has to keep them going to keep nvidia in line.
>>55383771
You mean the power supply
PCIE only handles 75W
>>55383875
yeah whatever
both are called pcie connectors for whatever reason and it gets confusing.
Ill call them aux power probably in the future.
>>55383870
Didn't someone say it has 6 pin pci-e as well?
>>55383908
No. Sapphire's Nitro has 8 pins.
>>>55375405
>>55383955
Oh well that changes everything, I might buy one of those then, or save up for a 1070
>>55383989
The 1070 is a 144p/60fps or 1080/100+fps GPU, there's no discussion between the twos.
>>55383875
>PCIE only handles 75W
Standards say it only should use 75w, it can supply more than that.
How much is it affected depends on the quality of it tough.
>>55384000
I only play at 1080p, I don't need anything higher than that for now. I can't stand multi gpu setups with terrible scaling i've experienced in the past
>>55384040
Then the 1070 is more powerful than you need. "Hopefully" (don't quote me on that) the custom 480s will bring the performances consistently at 980 level even on Dx11 (currently it has 980 performance only under Dx12). Else the 1060 is coming this month too. But the 1070 is more powerful than a 980Ti, why'd you get that for 1080p/60fps?
>>55384079
I have a i7 870 and a gtx 660 atm, it runs great for csgo and tf2 etc but when forza horizon 3 and mafia 3 and all those triple A titles come out this year they require a decent pc by todays standards so loooks like i need to upgrade everything. I'll wait and see how the 480s are before deciding since its a $350 dollar difference in Australia.
>>55384184
>http://www.logicalincrements.com/
Change the country in the upper right, dump questions in the comment section, and everything on the same line (hence, within the same tier) is compatible together.
>>55384227
Yeah I've used that website a lot, Does a /g/ user own it? I think i'll go towards excellent or outstanding with 6600k for futurability
>>55384278
Excellent would be current 1080p/60fps on most things. Outstanding would be current 1440p/60fps on most things. Obviously we don't know how it's gonna evolve with Dx12/Vulkan that promise to be less greedy, but I guess it's a safer bet if you can cough out the extra cash. But I don't read the future.
>Does a /g/ user own it?
The Falcon does trip here.
>>55384278
>>55384327
logicalincriments started as an infopic created by /g/. I'm pretty sure most of if not all of the editors were at some point posters here either as tripfags or as anons.
>>55373068
>prime example of why their reputation is the way it is
Right because 3.5GB is justified simply by 'being Nvidia'.
Fuck you anon
>itt fags who have never overclocked anything
>>55385250
3.5gb doesn't kill motherboards, it's nothing compared to this
>>55382583
>There are zero problems with the strix
>strix
That's not a reference anon
>>55385250
Why are you making this comparison? 3,5 is not that big of a deal, because the cards still perform to the benchmarks - but not buying a 970 because of 3.5 is perfectly justified.
Will rx470 be a better budget card for 1080p after this issue?
>>55373009
Actually, it worse then that.
AMD put an out-of-spec 6-Pin PCIe connector on reference 480s and don't even bother using it fully. They opt to draw from the PCIe slot.
Somebody at AMD fucked up the reference card design and it was too late for AMD to make a less minute change.
They decided to roll with it.
>>55373313
That can't the the Don, since the Don is winning!
>>55386007
I think it's more likely that the card was never intended to run at the kind of TDP that it does. The design it has is sane for slightly lower TDP.
Then again the VRM on it makes no sense to me. 6 phases is way overkill for some dinky ~110W card.
>>55386028
It is because Polaris 10 chip failed to achieve its "970-tier" performance at its intended clockspeed.
AMD was forced to pump it up but they through power consumption out the door. They couldn't afford to delay for a complete redesign and wanted to beat 1060 to $199-249 punch.
>>55386050
Certifiably false. The RX480 comes already pretty close to its optimal clockspeed
>>55386081
It's clearly not optimal when they are that close to the power envelope of their PCB. Either the PCB is wrong or the TDP is wrong. I find it hard to believe AMD is incompetent enough to design a PCB this bad.
>>55385624
Okay, say that again after your 580 burned down your house.
>>55386081
480 is not at an optimal clockspeed. It was clearly aggressively clocked in order to meet its 970-tier performance goal.
The reference 480 layout was design for something a bit less. AMD bump-up at the last minute and couldn't afford the time and $$$$ to do a redesign for the extra juice.
>>55386130
AMD hasn't released a 580 yet.
>>55386102
>>55386141
The RX480 is overvolted and has poor cooling. Some are still eating too much juice so theres something else going on along with those two.
Weird thing is that there are a lot of cards that work normally. Go figure
>>55386166
Ironic retardation is still retardation anon
>>55386242
>Weird thing is that there are a lot of cards that work normally. Go figure
They work properly, for people who bought $500-600 motherboards.
>>55386242
The problem is that power layout on 480 reference board is fucked-up.
It is drawing half of the current from PCIe slot and half from external connector. The 6-pin connector on 480 out of spec. It has three 12V power wires instead of two (the middle one is for sensing 12V current) which would make it a "pseudo" 8-pin connector. The sad part is that reference 480 doesn't use it properly.
Reference 480 were clearly rushed to the market and current board design was made for something a lot less powerful.
>>55386299
Correction, it works on an over-engineered PCIe 3.0 slot without an issue.
Not the case with subpar PCIe 3.0 slots and slots predating PCIe 3.0.
>>55386316
>over-engineered
Not really AMDs target group.
>>55386299
No no, I mean some cards are drawing as little power as they should. Its strange senpai. Maybe those people arent using fucking Furmark as their benchmark? idk
>>55386303
Hmm I see. Well I hope AIB boards are good or at least fix that issue.
>>55386378
None of them are working normally. You have to measure it at the PCIE slot to get a reading, everyone who has the equipment to measure it there has shown it is overdrawing every single time.
>>55386050
Yup basically the stock 480 is actually already overclocked, that's why it has so many power and heat issues.
>>55376948
Gayming oriented arch.
The 290x is better at compute for some reason.
oh well since it's an rx 480 thread
anyone tried the open source drivers on linux (debian or ubuntu or derivatives) ?
Are they stable compared to nouveau?
>>55373033
>>55373009
https://www.youtube.com/watch?v=2l7ErTsvhxg
its actually using around 220-250 watts
top kek
>>55373009
>Implying 2 extra grounds would of done shit
>>55387314
*would have
>>55382583
Everyone's getting fired at amd soon. Either from going bankrupt or a 480 catching fire because its a piece of shit.
>power pins
Aren't those 2 extra just earth?
A 6 pin should also be able to handle 150 watt
>>55377084
May as well go back to pixel pipelines then by that logic.
>>55386378
If your PC can't handle 24 hours of furmark and prime 95 simultaneously you should consider getting a new one
So, real talk now /g/oys.
Obviously the reference RX480 has some issues. That's not a huge surprise, you're a retard for buying reference in the first place on a card known for overclocking well. So buying a non-reference card would be a much wiser decision. That'd be great and dandy if the non-reference cards were around the RX480's much-hyped price-point (VR at $199!), but Sapphire, Asus, and MSI's aftermarket coolers are ALL 8GB only. Given non-reference cooler markup, that means these cards are going to be going anywhere from $249-$269.
So I'm an amerifat, and referencing amerifat prices here, but at that point, you're breaching into the lower end of GTX 970 prices - even ones with non-reference coolers. And the RX480 has been shown to perform nearly equal with the GTX 970 in most cases, with the exception of DX12 content.
With that in mind, is it really an amazing budget card any more if OEMs are unwilling to produce 4GB aftermarket-cooled versions of the card? I don't see the GTX970 being called a revolutionary value-level GPU. I could get an R9 290 for $180, BIOS mod it out to be a 390X, overvolt and overclock it as high as I can within a reasonable thermal envelope, and beat the performance of the 970 and RX480 for nearly $100 less. I mean, fuck, the ASUS GTX 970 recently went on sale for $199.
And is this really AMD's/the OEM's fault, or is it ours? So many people insist on shelling out the extra $40 for 8GB on a -budget card- where the framerate difference between the two tends to literally be about 2%. Have we dug our own grave?
>>55390048
>is it really an amazing budget card any more
The answer is no. A 970 is actually a much better buy at this point.
>>55390578
Nobody in their right mind would seriously consider buying a card that doesnt even have 4gb of vram in 2016, 480 is price/performance king. Just accept it and move on.
>>55390612
Unfortunately although the 970 is the top high end card on steam, it pales in comparision with the bulk of the masses still using something like a 750Ti.
Hell, I'm on a 1GB 650TI and constantly getting low video ram warnings on Forza 6
>>55390612
The 970 is the same price and same performance as a 480.
The difference is the 970 isn't plagued with all the problems the 480 has.
It's much smarter to buy a 970 right now.
>>55390652
well the problem with the 970 doesn't directly stem from low vram. It comes from the last .5gb of vram which are incredibly bottlenecked by low transfer rates. I guess the 1060 will be faster alone just because it doesn't have to deal with cripple vram.
But then again the 460/4gb might be the better buy for you.
>>55390663
It is not the same performance.
It will lose even more ground due to no dx12.
The current 480 issues will be resolved with next update and the new reference models.
Meanwhile the 970 will always be plagued with 3.5gb and will only become a liability within the next 6 months.
No matter how you turn it the 970 is the worse buy except you get it at least $70 cheaper than the 480.
>>55390812
It actually beats the 480 in some benchmarks.
The best thing though is the 970 is actually available to buy at $250, while the lowest 480 you can buy without waiting weeks for stock to come back in is $350
>>55390663
>buying a 2 year old 3.5gb dx11 card in 2016
Delusional nvidiot derected.
>>55390929
The sad thing is that a 2 year old card is better than the 480
>>55390851
What is this $350 meme it's like $200 on newegg.
>>55390941
480 beats the 970 in every benchmark in dx12 by massive margins, and the 970 slightly beats 480 in Nvidia gameworks games at dx11, yeah you're a delusional nvidiot.
>>55390952
It's sold out on newegg and everywhere
The only place you can buy it is for $350 on ebay
While the 970 is $250 brand new to buy right now
>>55391005
Move to the uk it's like £174 for the 4gb and £200 for the 8gb
>>55390048
>Make this post to discuss a lack of non-reference 4GB 480s
>Trying to promote discussion about how this hurts the price point of the 480
>/g/ immediately goes to LOL DELUSIONAL NVIDIOT vs EAT MY 3.5 AYYYYMDRONE LOL DX12
Jesus christ, you retards, can you focus what's important rather than the thing that's been discussed to death and decomposition?
>>55391021
>Jesus christ, you retards, can you focus what's important rather than the thing that's been discussed to death and decomposition?
AMDtards are not rational, pointless trying to discuss facts with them
>>55391039
To be fair, it was an immediate "lol 970 > rx480" response that sparked that shit-fling.
I personally think that they're about equal at this point, given price points and performance. If AMD had only released a 4GB version, then I would be recommending it in a heartbeat as a 970 killer, but with the 8GB version unnecessarily driving prices up its just a loss.
Seriously, why 8GB? WHY?
According to AMD, they chose the 6pin connector so that more people can use it. "GPU for the masses", even the ones with old PSUs, that do not have 8pin connectors.
>>55391351
>not using poo for thermal paste
>there are still people dumb enough to consider the 970 when 480 and 390 exists
LOL
>>55391551
>not wanting good geralt hair in witcher 3
have fun with 1 fps xdddd
RX 480 winrar senpai
>>55385451
should be fermi desu
>>55391551
>pic
top kek
>>55390983
Nope
It handily beats it at Gta v AND witches 3 without hair works
AHHHHHHHH
>>55391976
itd be more believable if the GPU was under load.
Nice try though.
>>55391019
I'd rather not move to 'new pakistan' just for a gpu.
>>55375474
Its a shame that they changed the "front" of the card with that ugly plate with holes, i liked the first nitro design
>>55377846
Go to bed, Mr. Adams.
>>55391551
>there are still people dumb enough to consider AMD.
https://en.wikipedia.org/wiki/Battered_person_syndrome
Seek help.
