OC'ed AIO RX480 rapes OC'ed 1060 >https://www.

I play games other than Ashes of Singularity

interesting. maybe this will shut up nvidiots

put the same cooler on 1060 and see it go

I'm sitting on my 1070

AMD can't even compete

rx480 beats gtx 1080
/Thread

>>56073749
You paid 200$ more for what exactly? Increase in 6fps?

Interesting that it didn't blew up as the VRM now has no airflow over its toke heatsink unlike with the original cooler

>>56073709
This.

Something I have noticed as a trend through these benchmarks is that AMD tends to have a larger gap between the average and minimum framerates while Nvidia seems to be tighter.

There were several cases where while the 480 had much higher average frame rates the minimum framerate was around or below the 1060.
I see this as an issue.

what were the temps?

do i return my 1060 now ?

Would this be a possible GPU + Cooler combo for someone in Europe?

Does anyone know the prices? Would this cost lower that 1060 cards?

>>56073543
Even if it seems like this guy won the silicon lottery, it's still hilarious to see the RX480 BTFO'ing the 1060.

I have r9 280x
should I upgrade to 470 or 480?

>>56074228
I don't think the 470 would be much of an upgrade and the minor performance bump you'd get from the 480 probably wouldn't be worth the price unless you also care about temperatures and noise levels.

soooo.... where are all the nvidiots? are they shut ?

>>56073543
Its funny that AMD needs an unfair scenario to win in benchmarks

>>56074266
>Overclocking is unfair when nvidia loses
Alright.

>>56073543
Cool story, bro! Nvidia for lyfe!

>>56073543
>rx480

call me when you actually can BUY one. this shit has been """"""released""""" two months ago and still nobody is delivering it

>>56074275
>doesnt put the same or similar cooler on the 1060
The OCed 480 without the water cooler was getting beaten most of the time by the OCed 1060. While it is interesting what the 480 is capable of when not limited by temperature, it wasn't like the 1060 was also tested with such limitations removed or that anyone would seriously consider buying a 480 for this.

>>56074282
nvidiots continuing to claim nobody has an rx480 because its sold out, its been sold out, and it still is because PEOPLE ARE ORDERING THEM SO THEY GET SHIPPED OUT THE MOMENT THEY GET THEM.

>>56074228
Unironically wait for Vega.

>custom hybrid
>just to make a midrange gpu perform better than another midrange gpu
get a life ffs

>>56074319
>sold out
and still nobody has one. have fun waiting 2+ months to get your product you preordered

Why didnt he slap this cooler on 1060? Or 1060 isnt limited by temp?

>>56074362
If I remember correctly, gamer nexus did a custom water-cooled 1060 and it didn't overcloock that well

File: 1700.jpg (216KB, 1600x900px) Image search: [Google]

216KB, 1600x900px

>>56074066
Yes, you probably should.

In a practical sense it is useless to OC both cards. Minor noticeable boost in games.

>>56073543
>amd niggers still lying to themselves

topkek

it's just pathetic at this point

>>56074401
what resolution was that test run at?

>>56074361
literally only read that its sold out. My brother ordered one, he got it, he didn't have to wait long. anhero you fucking cancerous human being.

>>56074440
Fire Strike is 1080p, extreme and ultra are just an increase in resolution.

>>56073543
OP, everyone has known for a long time that in OC conditions the RX480 wrecks the 1060. Compare HWBot data.

http://hwbot.org/hardware/videocard/radeon_rx_480/
http://hwbot.org/hardware/videocards#key=geforce_gtx_1060

>>56074380
Why would that stop you? For the purposes of comparison you don't use someone else's data.

>buy an expensive AIO card

might as well buy the next tier up if you're spending so much on a 'mid range' card

>>56073543
Get back to me when there is a one-off Frankensteined 1060 pushed to limits.

File: doom vulkan 1440p.png (19KB, 637x389px) Image search: [Google]

19KB, 637x389px

>>56073749

>>56073543
>wanting poo on your gpu
ayy

>>56074856
>nvidia hasnt received its optimizations for vulkan in doom yet

>>56074362
doesn't matter it more or less cant get past 2 ghz

>>56073844
http://gpu.userbenchmark.com/Compare/Nvidia-GTX-1070-vs-AMD-RX-480/3609vs3634


Try a huge improvement over the other one

File: XxhXDJKP2x6Z9jPMP7XgjD-650-80-1.png (153KB, 650x388px) Image search: [Google]

153KB, 650x388px

>>56074905

>excuses

>>56074953
>gpuuserbenchmark
wew lad is this a joke

>>56074978
Where's the 480?

>>56073543
https://www.youtube.com/watch?v=-ePpsT9fhpQ

>>56075088
>im poor

>>56075096
for that matter, where's the 6600k or any other current gen cpu

>>56075123
>i use userbenchmark to make buying decisions

>>56074766
Compare HWBot data.

>>56074493

File: 1471045515826-g.gif (499KB, 500x266px) Image search: [Google]

499KB, 500x266px

>>56073543
>custom cooler vs retail aib cooler
>still can't oc over 1.5ghz

>>56075146
no i use my brain literally 90% of the benchmarks ive seen show 1060>>>>>>480

>>56074228
260x user here. I'm waiting for the next generation of cards. But then, I'm pretty poor.

File: 2016-08-14_07h26_04.png (519KB, 1274x634px) Image search: [Google]

519KB, 1274x634px

>>56075165
what like this one

File: 1470344380378.png (35KB, 1053x278px) Image search: [Google]

35KB, 1053x278px

>>56075165
how about this one

File: 1470344441981.png (35KB, 1053x278px) Image search: [Google]

35KB, 1053x278px

>>56075165
what about this one

>>56075155

FUN FACT: when you overclock the coreo n an Nvidia card you are only changing the frequency the shaders run at, not any other sub systems. Overclocking on a GCN card is one frequency to rule them all which is a massive factor into why GCN scales far more linearly than anything Nvidia has.

File: 1470344503672.jpg (134KB, 614x442px) Image search: [Google]

134KB, 614x442px

>>56075165
or this one

>>56073598
>interesting. maybe this will shut up nvidiots
>maybe this will shut up nvidiots
>maybe

You must be new here.

File: 1470344566431.jpg (164KB, 614x444px) Image search: [Google]

164KB, 614x444px

>>56075165
how about this one then

>>56073543
I'm not going to buy a RX 480 for 320 EUR and an AIO for 100 EUR.

File: 1470344628481.png (121KB, 600x450px) Image search: [Google]

121KB, 600x450px

>>56075165
dis one?

File: 1470344690598.png (36KB, 682x802px) Image search: [Google]

36KB, 682x802px

>>56075165

>>56073709
>>56073953
>>56074266
>>56074362

Pascal can't oc that much more anyway. I look at the water-cooled hybrid 1070/1080. None of them can click better than a ref.

Now what the guy in the video should have done was stick that cooler on a 480 with an 8pin card. That would probably give some more oc headroom. Still tho it's a bad ass card. Clock for clock amd is way faster than Nvidia. Whish they could tap into that tho.

>>56075329
>EUR
no one cares yuropoor

>>56075365

DOMINATED

>>56073749
No it can't because it's not made to.

>>56075344
>>56075365
>let me pull out my list of the 3 benchmarks AMD wins out. You're sure to be seeing these again, and again, and again, because I have nothing else. That will surely prove that AMD doesn't lose the next 100, because I've pulled them out again!

I'm thinking of getting a 460

there is also a 470 but meh not a gamer

the 460 is available from asus and saphire - which one is better overall?

an AIO doesn't magically mean better OCs. This guy just has a golden chip, either that or he just falsified it, because he's (literally) the only person to get >1400mhz that didn't destroy their card with lots of voltage and ln2.

>>56075344
>>56075365
Why do you spend so much time and effort making excuses on an anonymous message board for a card you don't even own?

File: 1471083696980.png (15KB, 600x371px) Image search: [Google]

15KB, 600x371px

>>56075423
its a bit more than 3 family

>>56075430
sapphire

>>56075440
why are you mad when amd is better

>>56075458
Now post the rest.

>>56073543
>home made hybrid
>literally attaching a GPU AIO to a GPU
Seems about right for AMD. Claiming they did or can do something that's not actually some amazing feat.

File: xBgSksM.jpg (18KB, 340x260px) Image search: [Google]

18KB, 340x260px

ITT

File: 1471083759071.png (13KB, 600x371px) Image search: [Google]

13KB, 600x371px

>>56075510
sure thing :^)

File: untitled-14.png (36KB, 682x710px) Image search: [Google]

36KB, 682x710px

>>56075458
>I can't even compare it against the card under discussion!

File: r_600x450.png (120KB, 600x450px) Image search: [Google]

120KB, 600x450px

>>56075539
>I can't even compare it to the card under discussion!

>>56075372
I know AMD doesn't care for money, that's why they are losing it. I just need to keep pointing it out because the message doesn't seem to hit home enough.

File: r_600x450 (1).png (117KB, 600x450px) Image search: [Google]

117KB, 600x450px

>>56075220

>put expensive cooler on """"budget"""" card
AMDrones might as well go LN setup.

>people actually go through the effort of saving benchmarks just so they can win arguments on 4chan
get a fucking life you fat freak rofl

>>56075666
Satan has spoken.

>>56075543
>4k
>>56075575
>>56075600
>gaymeworks

cherry picking at its finest

>>56075666
People acually posting facts to back their arguments? LOL DUMB NERDS

>>56075939

/g/ thrives on lies and memes. Hell /v/ is actually the better board to discuss pc hardware than /g/ is.

>>56075176
The 480 is literally 4 generations ahead of that piece of trash 260x

It would be more than double the performance gain, why are you waiting when the 480 is the best priced card to come out in years?

>>56075975
>why are you waiting when the 480 is the best priced card to come out in years?

480 isn't best priced at all. you either get the 470 if you're a poorfag or you get a 1060 if you want a step up in the $250 range.

>>56075600

Do you faggots double check before posting useless crap or are you just too stupid?

>>56076151

Slow down. Buying the 1060 is the gayest thing anyone can do at this point. So ask yourself this? Are you Gay?

>aio 480

Congrats, you're only $60 away from a guy 1070 now.

>>56075220
>GTX 1060 losing in a gimpworks game
topkek

>>56075430
Sapphire has been very lackluster this gen. Currently Powercolor or MSI have the best rx 480's.

>>56076657

Really? Which one ocs the highest? I'm still undecided on which one to get.

>>56074228
Just wait, the 280x is still good.

>>56076657
I thought the ASUS card actually ended up being the best card despite being the shittiest looking one.

>>56073543
Lol who cares about some bottom of the barrel poorfag card. Seriously people would bother water cooling that? Im waiting on the water blocks for my new Titan XP cards. AMD has nothing to compete with this. I'll support the company which actually pushes the envelope.

Rx 480 might be good, but at least aib 1060 was available to buy 2 weeks ago in my country and cost me less than reference 480. So fuck AMD and their failed launch.

Still no aib rx 480 in sight too btw.

>>56076783
This I love my new NVIDIA Titan X Pascal™ cards. You would have to be poor to buy AMD :^).

>>56076762
No Asus tend to be decent but overprice compare to the other AIB's. Also asus usually just slaps on the same cooler used in their Nvidia GPUS.

File: Rise of the 480.png (300KB, 2800x1695px) Image search: [Google]

300KB, 2800x1695px

I'll just post this again. Note the Time Lie score. With a mild overclock it should match or beat the 1060 including the Jetstream model listed.

>>56077078
Seriously who gives a fuck about a 1060. It's like offcuts. AMD can't even compete with NVIDIAs entry level 1070.

File: eqAtrFH.png (269KB, 1920x1081px) Image search: [Google]

269KB, 1920x1081px

>>56077078

not sure what you're trying to show here. a max OCed 480 outperforms a reference 1060 in a completely synthetic benchmark? you realize that the 1060 can gain a shitload from a decent OC right?

there is almost never a reason to buy a 480 considering they are the same price as the 1060. the only good value card AMD has released this time around is the 470.

>>56075543
Neither is overclocked though. We are comparing overclocked cards here.

>>56075575
Didn't use the latest AMD drivers which gives RotTR a boost on AMD hardware.

>>56077197
I'm trying to say there is not much difference between MSI Gaming X cards from both teams and a mild overclock on the AMD puts it on exactly the same terms as the 1060 outside of games that favor either team. If you are talking Gimpworks vs DX12/Vulkan games that's another matter entirely and will require a big overclock on a 480 to get past an overclocked 1060 on Gimpworks titles just like in OP's video. But it can be done with some effort. I have ordered a MSI RX 480 Gaming X and look forward to seeing how far I can push it.

>>56075939
https://en.wikipedia.org/wiki/Spin_(propaganda)

I can have respect for someone like Rush Limbaugh who knows he's lying to a gullible populace, but someone who goes out and collects lies to believe?
Why is your identity connected to a video card company? "AMD CAN'T LOSE OR I AM NOTHING! I MUST COLLECT AND HOARD EVERY SCRAP THAT SAYS THEY AREN'T COMPLETE LOSERS SO I CAN CONSTRUCT A BUBBLE OUT OF IT TO LIVE MY LIFE IN!"
Why are you spending so much time and effort dismissing the hundreds of benchmarks that don't say what you'd like them to say in order to wrap yourself in the 5 benchmarks that do? Here you literally dismiss two games that you first posted benchmarks of just because I came back and posted ones that went the other way
>>56075919
So now they're >gaymeworks
because they're not in agreement with the narrative you need, and you have the gall to say I'M the one cherry-picking?

>>56077315
Because facts change over time. What your fact was a few weeks back are no longer facts since driver optimizations and other factors. Posting review samples from several weeks ago does not tally with what is achievable from RX 480's now.

holy shit these gpu threads are 100% autism

>>56074266
How is it unfair they were all overclocked except the 480 for reference

>>56074011
1060 did as well

>>56073598
>shut up nvidiots

>shut up
>nvidiots

pick one fa m

File: old man huang.jpg (25KB, 475x324px) Image search: [Google]

25KB, 475x324px

>>56077361

Take it from me goy, only release behcnmarks matter - your 780ti is fine.

>>56075155
He mentions he overclockedbthe cardvto 1517 or about but the voltage was high and likely worrisome or not stable enough.

>>56075625
he stated in the video he got a good deal and the total cost was less than an AIB 480

File: ss (2016-08-14 at 12.37.48).jpg (146KB, 1119x643px) Image search: [Google]

146KB, 1119x643px

how legit is this? my r9 290 gets ~14,000 and I've been thinking of upgrading, but I guess my card is fine?

The MSI RX 480 Gaming X is £30 cheaper than the GTX 1060 variant from where I buy. Since the performance difference is marginal and an overclock will bring it closer to the 1060 my decision is made easier because of that.

>>56074905
>nvidia hasn't received its optimizations for ashes of singularity yet
>nvidia hasn't received its optimizations for total war warhammer yet

>>56077361
You literally don't understand you're just using another axis to cherry pick along? You go along and pick the very best of every AMD driver revision, ignoring any regressions in performance, and treat it as though it comprehensively shows the performance you can expect out of the card.

Driver rev 1.01
Game A shows +10%
Game B shows +0%

Driver rev. 1.02
Game A drops 30%
Game B increases 10%

You:
"GAME A's +10% (see chart 1!). GAME B's +10% (see chart 2!). AMD INCREASED PERFORMANCE BY +10% ACROSS THE BOARD!"

"The RX 480 loses here? It doesn't matter! Don't you see it's cheaper?"
"The RX 480 + AIO water cooler costs just as much as a GTX 1070? Who cares what it costs? Don't you see it slightly beats a GTX 1060 in these three benchmarks! That proves it's the better card!"

You're all over the fucking map here.

>>56077654
There's no point to upgrade from a 290/390 to a RX 480.

>>56077654
If i were you i'd wait until VEGA and 1080Ti Non-ref models.

>>56077654

The only real upgrade path for hawaii is either a 1080 or the shiny new titan and even then you're probably looking at going up resolution as well.

>>56077666

Don't worry, Nvidia's superior drivers that /g/ tells me about should sort out all these DX12 woes right?

This is fucking pointless. Adding extra $100 to price in form of air/watercooling and fucking up your warranty in progress is laughable when you can buy a 980 Ti for $320 or 1070 for $350.

>>56077732
I just realized how lucky Americans are.

http://www.newegg.com/global/uk/Product/Product.aspx?Item=N82E16814126111&cm_re=1070_gtx-_-14-126-111-_-Product

This is insane, I have to pay $120 extra for the same card in UK.

Rx480 beats titan XP

Fuck off AMDrones

File: 1469751479787.jpg (15KB, 500x333px) Image search: [Google]

15KB, 500x333px

So first poor people buy mid-range GPUs as if they are somehow relevent

Then they fucking tell you to buy TWO of them for the stupiest reason

And now they go FURTHER into insanity adding a $100 CLOSED-loop watercooling system on their mid-range GPU

What is wrong with you?

>>56077802

>So first poor people buy mid-range GPUs as if they are somehow relevent

Its not like mid range gpus make up the majority of gpu shipments to both consumers and OEMs or anything.

File: 1459520670428.webm (3MB, 1280x720px) Image search: [Google]

3MB, 1280x720px

>>56073543
>poolaris cards need >$100 liquid coolers to match the same price range nvidia cards

>>56078020
You are forgetting that Nvidia wins mostly because of the plethora of DX11 titles. An overclocked 480 'CAN' beat a 1060 in DX11 games but also stomps the 1060 into the ground on DX12/Vulkan (ignoring shitty Talos conversion). That means going forward and overclocked 480 is on par with a 1070 for less outlay (depending where you live).

File: 1459601013598.jpg (41KB, 1280x720px) Image search: [Google]

41KB, 1280x720px

>>56077830
They are not.
It's either low-end or integrated graphics.

>>56077830

>implying the majority of OEM sales aren't shitty atom cpus + intel igpus.

only gaymer prebuilts have gaymer cards like the 480, which make up a small % of any prebuilt sale.

Don't forget when suggesting a GTX 1070 over a RX 480 with AIO cooler the $200 Goysinc tax. You will be running a 1070 on a Gsync monitor right? If not you are an idiot.

>>56078272
What are you babbling about?

>>56078272
>Don't forget when suggesting a GTX 1070 over a RX 480 with AIO cooler the $200 Goysinc tax. You will be running a 1070 on a Gsync monitor right? If not you are an idiot.

who is using gsync or freesync in 2016? they're dead memes that add input lag.

>>56078293
DSFARGEG

>>56078296
1440p Gsync is the only reason to be running a GTX 1070 bro.

>>56075975
What about a 270x? How big of a jump would the 480 be?

>>56078449

A 480/290x is roughly twice as fast.

>>56078092
Depending on processor too.

>>56078559
Well that's obvious but anyone running lesser CPU's are so poor they should probably just kill themselves anyhow.

>>56078323
Would you stop spreading disinformation?

>>56078092
>wait

lol

I have a spare corsair h60 that I replaced with air cooling for reasons. Can I just mount this shit on my 470 and oc it to the moon

>>56077315
>I have respect for people who willingly cheat the weak
Holy fucking shit this explains everything with you shills.

>>56078652
>cheaper

lol

>>56078683

If you have a way to actually secure it (as well as provide some cooling for the vrms) its most likely doable. Hell there is nothing cable ties can't solve.

>>56073543
>now the cost is higher than a 1080.
what's the point?

>>56073543
why this white guy has an indian accent?
is he an albino indian?

>>56078593
Depends what you mean by less ? i5 4690?

>>56075176
>rx460 is $150 in my country
hurr durr. buy a 260x for $60 instead.
my hd7750 died.

>>56075975
please.
after all the rebranding it's just 2 gen ahead.

>>56074228
As someone who had a 7950 and jumped to Fury X, I'm ordering you to wait and see how Vega comes out first.
If you HAVE to upgrade, find a Nitro Fury.
480 ain't shit.

>>56074718
He says in the video he spent less on a reference card +cooler than one of the custom cards or a 1060. He got the card and cooler on sale.

>>56078881

At least 7950 to fury x is a worthy upgrade. Hell depending on how you overclock your fury x you might not have an upgrade path for years.

>>56078881
How about 750ti to rx 480?

>>56078922
>Hell depending on how you overclock your fury x you might not have an upgrade path for years.

fury x has no OC headroom and it only has 4GB of vram, it's not even a good card unless you're still on 1080p.

>>56078940
that's like 750ti to gtx 970/980

>>56074228
480 or one of the Fury cards for $300.

>>56078940
I just upgraded from a 260X to an RX470. That was a huge fucking jump. I think the 750ti is slightly stronger than a 260X, but it should still be very noticeable.

>>56078941

>spicy memes

Go look up HWbot numbers - the card overclocks fine and at 4k it outperforms a titan x maxwell in TPU's relative performance charts.

>>56078975
>the card overclocks fine

a 10mhz overclock isn't fine. the card comes clocked out of the box at fmax for the process it was produced on. anything more will require tons of extra voltage and a golden chip.

>it outperforms a titan x maxwell in TPU's relative performance charts.

yeah, it performs like a 980ti and titan xm in 2013's games at 4k. in newer games at 4k and 1440p it's gonna struggle because of the vram limit.

>>56073543
No, fuck this gaymen shit.
The real reason for getting a RX480 is that the open source drivers are fantastic on 4.7.
No stuttering, no vsync issues, DRI3 is working like a boss sauce.
Nvidia has this weird shit going on with their prop drivers making gnome lag on some occations. Playing a youtube video in chrome and dragging a window over it shows that the window is trying to sync like fuck but seems to do it at every other frame or maybe trying to follow the videos sync or some shit.
Like what the fuck is nvidia doing in that hell hole of a driver?

>>56078922
Yeah. I'm happy with what I did and I don't regret it. I do wish it wasn't limited to 4GB, but that's something I knew and I wanted to see how HBM changes things.
For my monitor, which is 2560x1440@144hz, it definitely does well. I needed to move up from the 7950 since it can't do Freesync so I got the Fury X pre-ordered the very first day it was available to do so and received it on the very first day it released.
In Battlefield 4, I've lowered some very minor settings that don't really make much/any graphical difference to me while actually playing and can achieve 144fps stable, with the lowest dips at 120fps.
In games like Fallout 4, I have settings set to realistic standards considering the VRAM limit and get 70fps average, 50fps average low (though it can go down to 20fps in very specific parts of Boston, though I'm told by a friend with a 970 that it happens to him there too.)

Ultimately, I'm happy with my card, though I am interested to see if Vega has a HBM2 Fury successor. If it does and its not much of a jump, I'm waiting to see how Navi comes out. Might not bother upgrading to Vega even if it is good.

>>56078941
Actually, I get a steady 1200mhz OC with my Fury X, though it IS custom loop cooled. The original AIO wouldn't have stood a chance, though. My max temp on the GPU dropped by 22c as soon as it went custom loop.

>>56078940
Should be a somewhat decent jump.

>>56079022
Couldn't you just get a 460/470 and have the same driver compatibility?

>>56079015

>a 10mhz overclock isn't fine.

Lol.

http://hwbot.org/hardware/videocard/radeon_r9_fury_x/

>yeah, it performs like a 980ti and titan xm in 2013's games at 4k. in newer games at 4k and 1440p it's gonna struggle because of the vram limit.

Stop talking shit.

https://www.techpowerup.com/reviews/Zotac/GeForce_GTX_1080_Amp/23.html

File: guts+speccy 27-06-16 - Front.jpg (2MB, 2560x1920px) Image search: [Google]

2MB, 2560x1920px

>>56079029
Forgot pic. This is mine.
Can & will post a timestamp if necessary.

>>56078964
>>56078956
Not sure if I'm underestimating my card sometimes

But I only bought it for £75 when I'd switched from one job to another and had to wait to finish off my computer.

i5 4690k £150
Asus Pro Gamer z97 £70 (sold headset that came with it)
RMX 750 £55 (performed some RMA fuckery with a TX)

Curicial RAM 8gb 1333MhZ some old shit for free.

tfw you've had an aftermarket 290x since 2013 and it has just been getting better with every driver and is still kicking the shit out of 1080p

File: perfrel_3840_2160.png (42KB, 500x1090px) Image search: [Google]

42KB, 500x1090px

>>56079061
>Stop talking shit.

>memes about the fury x's pitiful overclocking then compares it to reference 980ti (which isn't even sold)

any non-ref completely 980ti destroys the fury x performance-wise and isn't vram limited like the fury x is. there's no good reason to buy a fury x unless you get it REALLY cheap (like $250 or less).

>>56076657
>powercolor
That indian company is a joke, anyone who buys that brand is basically shooting themselves in the foot.
>b-but it looks cool

>>56079152

>posts months old chart versus one thats 2 days old.

Well done, well done.

>>56079063
A single 360 rad is enough for those parts when the machine is heat soaked?
I honestly know nothing about water cooling.

>>56079189

the fury x results are the same, tpu doesn't rebench old cards for each review unless they change the test system (which hasn't changed since they switched to skylake on it over ~8 months ago).

>>56079177
I didn't know Taiwan is India. You learn something new everyday.

>>56079044
Yes, same driver.
Not sure how well amdgpu-pro is doing, but most of the stuff in amdgpu-pro is going into the open source stack. The only thing left, and the meaning with amdgpu-pro will be some legacy opengl profiles shit that professional users require.

>>56079208

Then how do we explain the fury x's gain on the titan? Unless maybe, just maybe it is performing better in recent games and not ones from 2013 like was suggested.

File: 75456.png (30KB, 650x400px) Image search: [Google]

30KB, 650x400px

>>56079280

it's been performing at the level of the 980ti and titan x since launch, it falls off big time at lower resolutions and gets shit on by even meagre overclocks on GM200 cards, which is why it was so unpopular.

>>56079202
They're thin rads, so having 2x360 is a good idea.
But yeah, typically a single 360 ought to be enough to handle 1 CPU + 1 GPU.
I got 2x360 because like I say, they're thinner ones and its for future ability to go MGPU without heat capacity worries.

>>56079084
I bought my RX470 for $168 last week. Feels good.

>>56079123
>Implying it's not also doing fine at 1440p for a while as well.
>Implying it wasn't the same cost as a 480 when it was going eol.

You're not saying enough good things about this card.

>>56079331
You have a folder of nothing but Ashes of the Singularity and Shadow of Mordor benchmarks, don't you?

>>56073565
so i take it you didn't bother to watch the video where he covered a varity of games?

then again why would you. you're a nvidia shill.

0.02 cents have been deposited into your paypal account rajesh. nvidia thanks you for your shills.

>>56079448
Nice. I could wait two months for some two fan 1060's or pay more than £260 for one. or get a Sapphire nitro for £240, may do it.

>>56075278
Source?

File: Boy-wearing-Cool-Short-Haircut.png (381KB, 500x500px) Image search: [Google]

381KB, 500x500px

>>56079518
Nvidiots now adding Shadow of Mordor to the list of improper games? Top fucking kek. First it was Ashes then Hitman, Batman, Doom, Forze 6, Quantum Break, GOW, Blops3, Warhammer, Division and now THIS?

>>56077732
Where?

>>56080791

Go read the NVAPI documentation for Nvidia.

I think for AMD its covered in their GCN whitepaper. https://www.amd.com/Documents/GCN_Architecture_whitepaper.pdf

File: RX-480-unboxing-4.jpg (94KB, 460x613px) Image search: [Google]

94KB, 460x613px

When the fuck is Microcenter going to get AIB 480s. Still sitting on my leafblower reference one

>>56075154
For what? I don't see a one-off frankensteined 1060 pushed to the limits.

>the rx480 is so shit than you need a AIO like the Fury X because it's a housefire

>>56073543
>the RX480 + water cooler

The price pls

>>56080858
>he doesn't know what cherry-picking is

>>56080947
Fury X doesn't need an AIO. It needs fucking custom loop cooling.
(Literally) ripping off the stock cooler and replacing it with an EKWB block + backplate for passive VRM cooling reduced my temps by 22c under load.

>>56081258

>going from 120mm rad to 240mm rad improves temps

Hell I bolted a truly monstrous heatsink to my 290x and dropped temperatures in the region of 40c on the core.

>>56081299
Actually, I've got 720mm x 26mm of rads.
But even if I only had 240, it'd still be a decent improvement. That single 120 just wasn't enough.
Should've at least offered an option where it was pre-installed with an EK/competitor block or something. Even if it cost slightly more, it'd be very much appreciated by a lot of people, I reckon.

>>56073543
>using rapes unironically instead of beats, demolishes, etc.
>in the current year

Only 15 year-old tryhards talk like that. Stop doing that, even if you are 15. And if you are, you have to be at least 18 to use this site.

As far as video content
>the 480 only significantly outperforms the 1060 in doom vulkan
>everything else is less than a 5% performance increase or still loses to the 1060

Built a custom AIO and OC'd the shit out of it to still lose or barely edge out ~2% lead except in the one area where we already all know AMD is stronger. Wow, what a fucking revelation.

>>56078727
The cost is 10% lower than 1060. He says it in the video

>>56081857
Not the same anon but I find it hard to believe he built a closed loop system with a double rad and three fans for less than $50.

>>56082006
Stop posting this stupid fake

>>56074338
get a hobby you sad fuck.

File: hqdefault.jpg (13KB, 480x360px) Image search: [Google]

13KB, 480x360px

>>56073543
>even white AMDfags sound like pahjeet

>>56082006
With that many connections, it wouldn't be dual, it'd be quad GPU.

>>56082212
Honestly, even as someone who owns a Fury X, I was certain that this guy must live in India or something. Fucking his name may as well be Raajeet.

>>56075368
no it wouldn't have, venders have already said the silicon lottery is real when it comes to oc's even approaching 1500

This guy could have used an 8 pin, but that wouldn't mean the chip in the card would have been as overclockable, but it does mean there would be more headroom if it is.

>>56077446
not to mention overall cheaper than an aftermarket solution.

File: amd-devil-13.jpg (113KB, 600x328px) Image search: [Google]

113KB, 600x328px

>>56082138
>fake

>>56082369

There is a difference betwen having 4 8pin connectors and having 9. In fact 9 would provide so much power the form factor doesn't allow for enough cooling.

>>56073543
I'd care, but I don't want to recommend a mainboard killer

>>56082427

>zero confirmed reports
>gamers nexus actually trying to kill a motherboard with a 480 and its not working

>>56082427
Literally only happens with the absolute shittiest of motherboards which are already prone to breaking anyway.
Just because PCI-E spec is 75w, doesn't mean its all they can actually handle. That's just the MINIMUM they HAVE to be able to handle.

>>56082455
>>56082508
And yet here you are, being mad as fuck.

Your favorite card is pretty lame senpaitachi

>>56073778
>/threading your own post

File: not the same.png (212KB, 2382x1131px) Image search: [Google]

212KB, 2382x1131px

>>56082556
Not the same person dude.

>>56082580
I'm aware. Should be expected. There's two billion of you nasty shits after all.

>>56082620

>makes claim
>has zero proof
>is called out on it
>memes as a response

Merely pretending?

>>56082620
I'm not an Indian. Nice of you to assume anyone and everyone who wants to call you a retard is supposedly one though. Shows how mature you really aren't.

>>56082643
I didn't expect you to get the joke, and here you are giving me a moral lecture
>hurr anon it's so rude to assume every AMDrone is a dirty Indian, shame on you
thanks man, it's been a while since I actually laughed at a 4chan post

>>56082732
Every shill is an Indian, but my post wasn't even pro-AMD. My post was specifically stating that just because its a standard, doesn't mean its all it can handle. Nice mems.

>>56076401
I'm convinced

why are you so obsessed with the 1060? are you really that threatened by its existance? jesus christ just man up and accept you're clutching at straws.

>>56073543
>>56073543

>>56082212
>what is a german accent

>>56078449
a decent upgrade

>>56078652
everyone filters you why do you even bother

>>56078964
what ever is in stock

>>56080970
its about decent cooling, a good aftermarket air cooler could do the same thing and still be cheaper than the 1060

>>56082759
shill

File: 1406692461117.png (410KB, 513x834px) Image search: [Google]

410KB, 513x834px

>>56083593
>480
>cheaper than the 1060

KEK. it's more expensive throughout europe and asia.

>>56083674
>europe and asia
why do you think i care about 3rd worlders pajeet

>>56083717
are you unironically telling me to spend the equivalent of $50 more for a 12% slower 480?

kys

>>56083755
see >>56075220
>>56075244
>>56075262
>>56075318
>>56075327
>>56075344
>>56075365
>>56075458

>>56083755
>kys
newfag detected

File: perfrel_1920_1080.png (42KB, 500x1170px) Image search: [Google]

42KB, 500x1170px

>>56083863
here, have a compiled performance summary :^)

no need to thank me for educating you.

>>56083915
>inb4 i-it doesn't c-count

>>56083915
9 games out of the 15 tested are gameworks and bf 3 is ancient, so what does this prove nvidia is better at playing shit/old games

>>56083915

>kepler dying

Gets me every time.

>>56083977
see >>56083948

stay assmad.

>>56084005
>game works games make are valid benchmarks
>b-but thats dx12 it doesn't count
once again see >>56075220
>>56075244
>>56075262
>>56075318
>>56075327
>>56075344
>>56075365

File: 1462585598173s.jpg (2KB, 125x47px) Image search: [Google]

2KB, 125x47px

>>56084030
>b-but all these popular AAA goty nominee titles d-don't count
>games running in the current industry standard api don't count
>nvidia gameworks d-don't count b-but dedicated amd shader intrinsics do!

neck yourself buddy

see
>>56083915
>>56083915
>>56083915
>>56083915
>>56083915
>>56083915

>>56080933
HWBot is for comparing OC performance, they have relevant numbers for each type of cooling system. The 1060 doesn't OC well, even on LN, and the 480 beats it in the most relevant benchmarks for gaming.

>>56082369
That card was a fucking disaster. It killed PCIE slots just from weighing so much.

File: ASUS_ARES3(1).jpg (97KB, 1200x800px) Image search: [Google]

97KB, 1200x800px

>>56084104

There is a reason why Powercolor were the only company to make an air cooled version of the chip. Pic unrelated, its the sexiest 295x2 ever made.

>>56084082
>HWBot is for comparing OC performance
Does it clock well enough to perform better than the one-off Frankenstein? Without a comparison it's only speculation as we end up comparing apples to apples to apple juice.

>>56084148
I actually really liked the 295x2 as a card, just not the air cooled version. They're still the best dual gpu card under 1k IMO. Meme'd a friend into buying one and he's already paid it off plus some from mining ETH, which is only more hilarious since it's comparable to 1070/1080 performance for DX12 titles now so there's no reason to upgrade. I think it will be the only time in history where the most cost effective choice was buying a water cooled dual gpu card from AMD.

>>56084071
>popular games, literaly all those games are shit no one cares about

>>56084239
Water cooled OC only puts a 1060 at slightly over 2GHz, Pascal is hard limited to 1.25v unless you want to load a custom BIOS and tear apart your entire card. Even then a stable 2.2GHz will be out of range for most people.

The RX480 is a better overclocker compared to the 1060 if you're going with a custom solution.

>>56074338
It reminds me of those hilarious retards here who fell for the "price/performance" meme of the utterly terrible FX lineup and then bought expensive mobos and aftermarket coolers to OC the thing all so they could still have half the IPC of an i3 but in a much more expensive build. Shit's hilarious.

File: 1430980813601.jpg (17KB, 480x360px) Image search: [Google]

17KB, 480x360px

>>56084340
>>literaly all those games are shit no one cares about
>he thinks these are all shit
>he thinks nobody plays them
>he thinks the games he linked are good and everyone plays them

amdrones for you everyone. point at him and laugh.

>>56083979
How's it dying? A 780ti basically the same as a 290X which is what it was originally competing with

>>56084369
>The RX480 is a better overclocker compared to the 1060 if you're going with a custom solution.
The question wasn't about which was the better overclocker but which was the better performer at Frankenstein levels of customization. Given how this Frankensteined 480 is just barely a bit better than a regularly OC'd 1060 it seem reasonable to assume a simiarly Frankenstiened 1060 would pull ahead.

File: 1455257163401.png (51KB, 522x683px) Image search: [Google]

51KB, 522x683px

>>56084396

File: 1403349796138.jpg (47KB, 250x250px) Image search: [Google]

47KB, 250x250px

>>56084340

Yeah let's buy brand new GPUs not to play current modern games but rather to emulate SNES ones amirite?

>b-buy amd! It's g-good! Y-you just g-get worse plethora at the vast majority of p-popular titles b-but look at these 3 b-benchmarks I research for two h-hours to find h-haha!!

Hmm I'm convinced of this fine product's quality.

>poorfags rationalizing

lmao

>>56084448
worse performance*

>>56084431
I don't think Nvidia cards actually lost performance. No source on your pic either

Plus AMD has stepped up their driver game over the last couple years, and it's no secret they struggled with driver overhead with their cards

https://www.youtube.com/watch?v=DvCsT_dk-1E

>>56084431
>no source
>no citations

http://www.bytemedev.com/the-gtx-780-ti-sli-end-of-life-driver-performance-analysis/

meh, come back when the aftermarket cards aren't shitty and actually are near MSRP and not like $50+ more

>>56084426
The Pascal chips are limited to 1.25v unless you completely tear the card apart and start replacing components. It will not OC better by just slapping on an AIO cooler. People have already rebuilt these cards, used LN, and it still performed worse than the RX480.

File: 1403281368588.jpg (97KB, 500x667px) Image search: [Google]

97KB, 500x667px

>people unironically recomending modern mid-range cards based on DX 12 expectations
>literally buying a card now so that you can get - MAYBE - better performance in unknown titles from an uncertain future when the current cards will be long obsolete

>>56084500
What about flashing a custom bios?
People got their Kepler and Maxwell cards to ridiculous speeds doing that

>>56084469
>>56084473
The argument or point isn't that Nvidia doesn't improve or update their drivers. It's about them not improving as much as AMD's GCN cards over time. Post release Nvidia trends behind AMD in most cases.

http://www.neogaf.com/forum/showthread.php?t=1220928

File: Misogi_Kumagawa.jpg (114KB, 1092x910px) Image search: [Google]

114KB, 1092x910px

>>56073543
cool !

what about the heat it generates?

is it safe for my mobo to run it at such a high clock ? Z170-A

>>56084500
>it still performed worse than the RX480.
[citation needed]
I still await some comparison between the two. If you point to the same link in the HWbot which just ends up googling it, I'm going to assume you've got nothing to support this claim because that is just trying to shift the burden of proof.

>>56084500
so you're telling me a rx 480 at 1266 stock frequency beats a 1060 overclocked to over 2.1 ghz with bios mod and custom water cooler?

bullshit much.

>>56084520
I said AMD has been upping their driver game over the last couple of years. Everyone said their drivers were shit and they finally started fixing them.
Doesn't mean Kepler is dying, the performance has just been tapped out earlier

>>56084566
Yea. I used to hate on AMD drivers too but it's been getting better. I wasn't implying Nvidia cards are shitty in anyway. I buy both cards anyways. AMD just trends better over time that's all. Doesn't mean Nvidia belts out an evil laugh every year or two and releases a driver update that kills their cards. If anything it shows Nvidia's drivers pretty much use everything the card has out of the box as far as we know.

>>56084609
Adding to this it could just be Nvidia is easier or simpler than AMD's stuff I don't know. The trends are there though that AMD can eek out more performance over time. I don't work for either of them so no clue I just buy their shit.

>>56084506
>>56084533
>>56084566
Getting over 1.25v requires BIOS+firmware+component mods for Pascal. This is well documented for the 1080 in particular.

https://xdevs.com/guide/pascal_oc/#voltsc

The confirmed world record for an OC'd GTX1060 in Time Spy is 5176.
The confirmed world record for an OC'd RX480 in Time Spy is 5343.

The RX480 OC'd outperforms the 1060 OC'd in all the Fire Strike benches as well. Which you can confirm on HWBot with the links I've already posted. This is due to the 1.25v hard limit on Pascal.

>>56084391
>Anno 2205
burnt out series sucks ass
player count: 207

>Assassin's Creed: Syndicate
More milked than call of duty at this point
player count: 315

>Batman: Arkham Knight
so bad that they had to pull it from the steam store
player count: 2167

>Fallout 4
Bastardization of the original fallout games and on a 15 year old engine
player count: 26896

>Far Cry Primal
far cry 3 reskin no. 2
player count: 365

>Just Cause 3
lame and shit just like every other just cause game
player count: 1615

>Rise of the Tomb Raider
so bad there are almost more people playing the 2013 tomb raider
player count: 1471

>The Witcher 3: Wild Hunt
Game for autists
player count: 8870

so how many people are playing these games?
all the gaymeworks shit combined: 41906
BF4: 107735

>>56084690
So is modifying the voltage limit not possible if not done at the factory by Nvidia?

>>56073543
I don't care because neither of those cards are any good.
1070 is just fine thanks.

>>56084704
and player counts for forza, gears ultimate and those games? you can't be so deluded to think these games are better and played more than any of those games in that list.

>>56084690
>The confirmed world record for an OC'd GTX1060 in Time Spy is 5176.
Source: My ass?

>>56084710
I don't see what that matters since there are no AIB's that offer such a card, but in theory yes they could.

>>56084751
Those are amd sponsored/dx 12 games im just making a point that more people are playing a 3 year old amd sponsored game than any gameworks game combined

>>56078683
There are kits for that. I had one by NZXT for my 290X. Worked wonders on it.

>>56084754
Try looking at either HWBot or 3dMark's site itself.

http://www.3dmark.com/search?_ga=1.110403916.898346781.1470167332#/?mode=advanced&url=/proxycon/ajax/search/gpu/spy/P/1096/500000?minScore=0&gpuName=NVIDIA GeForce GTX 1060

File: 1433213433459.jpg (55KB, 600x600px) Image search: [Google]

55KB, 600x600px

>>56084777
are you seriously this dense in the head?

1. obviously the most popular large scale warfare fps game will get many thousands of players. mp games > sp games in terms of player count generally.

2. if nvidia sponsored it, it would be the exact same. being amd sponsored means nothing.

3. nvidia hardware is significantly better according to literally every benchmark anywhere on bf4 dx11

4. gears ultimate is actually nvidia sponsored, not that it makes a difference. the discussion is about tpu's performance summary chart and not game sponsorships

you are fucking retarded if you're trying to claim any of those games tested by tpu are worse or played less than any of those other games linked above like forza or gears.

>>56084802
>core clock 1.7ghz

you don't even know how far it was pushed because it's not showing it. be retarded somewhere else.

>>56084391
so how are you going to argue with this >>56084704

just cut your loses and admit that im right

>>56084802
>HWBot
That's just shifting the burden of proof.

>3dMark - 5170
>Core clock: 1725
TPU got a better OC on air so I doubt that's LN.
https://www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_1060/29.html

>>56084915
3dMark doesn't report the actual OC on the card. That's the highest a 1060 has ever scored in Time Spy, according to the maker of Time Spy's own website, and on HWBot which would be the first place someone would post a world record in the OC community. Face it, higher scores aren't available because they simply don't exist, and the 1060 gets beaten by an OC'd RX480.

>>56084915
>TPU
nvidia shills see >>56084704

Wait it's $120 for the AIO $199 for the GPU
that nets us at
$319 plus build time $350 in total. Putting it in spitting distance of the GTX 1070 in terms of pricing and a whopping $100 dollars more expensive than the GTX 1060 for equal or a little better in terms of performance.

>>56084966
The idea here obviously isn't that it's practical, it's that it's possible.

>>56084966
the AIO doesn't matter, that overclock can be done on any decent cooling, like a non reference card or a decent aftermarket air cooler

>>56084957
>3dMark doesn't report the actual OC on the card.
U wut m8? Are you claiming 3dmark is reporting the standard base clock of stock at 1506 as 1725 and not overclocked base clocks?

>>56084957
>thinking the average joe will buy any gpu to install a custom bios and water cooling
>celebrating this like it's a victory

please, get a life.

>>56084984
He claims that it's cheaper than the GTX 1060 and the Sapphire Nitro+ RX 480.
Which if his claim is true then it's good for him that he magically found a free AIO and his chip is one in a million, but it doesn't mean anything when comparing the two graphics cards.

itt: amdrones clutching at straws with
>muh 50 gazillion mhz custom overclock with this $300 cooler installed on muh $250 gpu
>all games that don't favour amd (ie 99% of games) are all biased and the websites that benchmark them are all biased too!

i bet these same people unironically think adoredtv is the best hardware reviewer and benchmarker.

>>56085073
>He claims that it's cheaper than the GTX 1060
It isn't cheaper if no one can replicate his build at the same cost.

>>56085014
>He doesn't know how Nvidia handles boost clocks with Pascal.
There are actually 3 clock speeds Pascal runs at now. It's not showing the OC frequency.

Sorry, but that's the highest a 1060 has scored on Time Spy that we can confirm from both the manufacturer and the OC community, and it is lower than scores from the RX480.

>>56085056
Again, installing AIO on a mid range card isn't practical, period. But tests show the RX480 beats the 1060 in extreme OC situations, which does matter.

>>56085079
nvidiot

>>56085073
stop shilling pajeet

>>56085073
he's dense in the fucking head. this retard will probably spend 1070 amounts of money on an rx480 to only end up getting inferior performance to a 1070 even with his oc and $200 cooler.

>>56084995
gamers nexus did the same thing and was only able to get it to 1390 Mhz almost 100 Mhz lower than this guys magic chip.

>>56085092
>There are actually 3 clock speeds Pascal runs at now. It's not showing the OC frequency.
U wut m8? Are you conflating the memory clock speed with the core clock speed and the boost clock speed?

>that's the highest a 1060 has scored on Time Spy that we can confirm
Which proves what? That 3dmark's records are incomplete at best?

>>56085092
>It's not showing the OC frequency.
You're right. It's showing it's least the slowest clock speed the card, the BASE CLOCK SPEED. That cannot be a "the best LN overclocked card" if TPU got even a faster base clock on air.

>>56085130
stop shilling

>>56085131
yea and my magic 290x is at 1200 mhz, so im pretty sure that about 15% overclock on a 480 is possible

File: 1348349883371.jpg (65KB, 960x768px) Image search: [Google]

65KB, 960x768px

>>56085113

>>56085146
>What is GPU Boost 3.0
Holy shit he doesn't even know anything about the card he's trying to defend.

"The GPU in turn operates at points along the resulting curve, shifting clockspeeds based on which voltage it’s at and what the environmental conditions are."

http://www.anandtech.com/show/10325/the-nvidia-geforce-gtx-1080-and-1070-founders-edition-review/15

He can't find any higher benchmarks anywhere, at all, but still sure they exist...SOMEWHERE...
>Nuh uh! Nobody has tried to really OC the 1060 yet guys! I swear!!! I bet they did and didn't upload it to the OC community!

Yeah yeah sure. Just admit you're wrong already so you can stop frantically googling it for damage control and saying "muh burden of proof".

>>56085208
autist

File: 1426205454587.png (84KB, 480x480px) Image search: [Google]

84KB, 480x480px

>>56085230

>>56084369
Whats funny is I said a while ago that the 480 was shaping up to be a 30-35% overclocker, sure you (kinda) need an 8 pin design and some beefy cooling but it's there.

1120*1.35=1512 and it seems, going by this video, even reference cards can, not always, hit that mark.

>>56085164
It only shows core clock on 3D Mark. That isn't the actual operating frequency of the card. Even TPU states the same thing about their OC results.

" Please note that although the 1060 Founders Edition card has been overclocked to 1731 MHz for the base and 1933 MHz for the boost, it’s actually operating at around 2.1 GHz in our tests due to the way Nvidia’s GPU Boost 3.0 technology works."

Which is from the page you linked.

>>56085210
Abandoned the "three clock speeds" argument? Good. Now please show when the clock speed is ever lower than base clock speed except when idle. Your own citation demonstrates it is never lower than base clock speed. We can also note Anandtech's 1060 review (since you linked the 1080 & 1070) that:
>In fact the average clockspeed is a little higher than I would have expected. GTX 1060 never averages less than 1822MHz
Far higher than the best timing Base clock of 1725mhz.

>>56085272
>30-35%
Sorry but you were wrong.

>>56074401

>>56085210
not him but you are severely stupid. if you turn up core voltage the card won't go from base clock 1.5ghz to over 2.1ghz. that's not how gpu boost works. a base clock of 1.7 ghz doesn't even get close to 2.1 ghz boost.

source: nvidia owner

stop talking about things you don't know anything about.

>>56085210
>Holy shit he doesn't even know anything
were you expecting anything else? nvidiots are mongs

>>56085305
Pardon? 1120mhz base time 35% is 1512mhz. OPs video shows a card that can hit 1470-1510, between the percent i've been saying.

Someones unverified 1700mhz run has nothing to do with it.

File: 1440918631088.jpg (32KB, 400x300px) Image search: [Google]

32KB, 400x300px

>>56073543
>why does a good card shit on a shit card

>>56085316
>pascal
>getting to 2 ghz
that's gold sample tier now fuck off

>>56085282
>>56085316
Which is what I was saying, that the card is running far above the base clock, but not reporting it in Time Spy. The card in the #1 spot on their records is running a heavy OC, but it isn't showing what the average frequency was because it's only reporting the base. If you notice, all the results below the top spot have core clocks above 2 GHz. The card in the #1 spot is running faster than ones that are below it for base clock. It's possible that the firmware and BIOS have been modified to accomplish this.

>>56085432
the main point here is that we don't know how far it was pushed. there may be 1060's which can achieve higher.

>>56085282
Pascal cards under load change between Base, Boost, and Max frequency dynamically. Effectively making them run at three different frequencies, Time Spy only records base, it does not poll to find the highest clock achieved.

>>56085408
every pascal card can achieve 2 ghz retard.

>mainstream GPU
>having to install highend cooling for it to even compete
kek

>>56085480
>every pascal card can achieve 2 ghz retard.
nvidia shills everyone

File: wmzmmq.jpg (26KB, 512x446px) Image search: [Google]

26KB, 512x446px

>>56085464
Not with that lovely 1.25v limiter on Pascal.

>>56085490
>>56085501

congrats. you just showed how retarded you are.

2ghz =/= 2.1 ghz. every pascal card can achieve 2ghz via boost with most around 2.05/2.075. 2.1 is another story.

File: It_dc9dda_5887448.jpg (67KB, 640x640px) Image search: [Google]

67KB, 640x640px

>>56074401
>GPU Clock 1.7Ghz
>1.7Ghz

Nice try, but that shit is fake as fuck.

>>56077666
Nvidia cards wreck AMD cards in total warhammer

>>56074228
Do you have anything more than a 1080p 60hz display?

>>56085464

4881 @ 2177 Core
vs
5176 @ ????

2.4 boost/max was where most people were having problems on Pascal, since that's the second highest score, I think it's safe to say it's at least in that range.

So if Pascal scaled perfectly in frequency to match the higher score?
It would put the card at about 2550 boost, which is approaching close to the max frequencies achieved on the 1080 before component mods. It's not scaling perfectly though because it's Pascal.


>>56085557
https://www.youtube.com/watch?v=HYg123osc04

>>56085613
no they dont see >>56075365

>>56085632
i'm pretty sure a magical 1060 will pop up just like that magical 480 did with that score. 99% of people won't be able to hit over 1.4/5 ghz on a 480 and 2.5 on 1060.

this whole discussion is stupid anyway. the 1060 is superior at regular achievable factory clocks where it matters (ie dx11 games) and you have to be some serious next level shill to buy a more expensive 480 to stick a $120 cooler on it.

>>56085645
What the fuck happened? a week ago sites were reporting that nvidia was faster even in DX12

>>56085645
>current year
>using windows 10

>>56085705
Sure sure, but for the record, the RX480 technically is a more powerful chip with OC, even considering that Time Spy doesn't have full Async in it's DX12 implementation.

>>56085729
those sites are shills

>>56085731
>current year
>being a freetard

>>56085729
it was probably tested with a realistic cpu other than >>56075365 which is using a heavily oc 6700k which masks all the 480's driver problems. it's common knowledge that the 480 doesn't gain as much when paired with lesser cpu.

>>56085729
They might have been testing an AIB vs an 1120 core clocked RX480.

File: driveroverheadisameme2.png (206KB, 1924x1083px) Image search: [Google]

206KB, 1924x1083px

>>56085768
>480's driver problems
opinion disregarded

File: overheadisameme.jpg (135KB, 570x1057px) Image search: [Google]

135KB, 570x1057px

>>56085768
>480's driver problems
get a trip so i can filter you

>>56085752
>even considering that Time Spy doesn't have full Async in it's DX12 implementation.

no. it's implementation of async is perfectly fine for being a neutral benchmark. all these assmad amd fanboys are complaining about async "not being fully implemented" because it doesn't use specific gcn shader intrinsic extensions which would make it biased towards amd gcn hardware. all games which give amd big gains over dx11/ogl like doom or hitman are specifically optimized for amd hardware via their gcn shader extensions. it's all on the gpuopen website if you don't believe me.

>>56085841
>being this assmad

File: 1469311822999.png (498KB, 1920x1080px) Image search: [Google]

498KB, 1920x1080px

>>56085799
>posting a situation of a gpu bottleneck and not a cpu bottleneck

how can anyone be this dense?

>>56085865
>look mommy i posted it again

File: c3r_1920vh.png (16KB, 582x368px) Image search: [Google]

16KB, 582x368px

>>56085824
here, have a chart with an actual source other than some slav chart some random butthurt amdrone made :^)

>>56085888
>pclab.pl
>.pl
lol thats still a slav chart dumbass

File: 1465402320579.png (167KB, 709x636px) Image search: [Google]

167KB, 709x636px

>>56085865
>>56085888
>>>/v/
its time to leave now

>>56085841
DX12 and Vulkan are designed to offer optimizations at a hardware level, which means it's just about optimizing for each set of hardware. I was simply mentioning that more performance is still possible from AMD. Is it really neutral if they don't attempt optimizing for both types of cards?

File: 1430397168041.gif (1MB, 400x368px) Image search: [Google]

1MB, 400x368px

>>56085914
>slav chart
>in english
>slav language = english now

KEK. leave and never come back.

File: stock-photo-a-hand-picking-cherry-from-tree-82047283.jpg (377KB, 1500x1096px) Image search: [Google]

377KB, 1500x1096px

>>56085888
>>56085865