>GTX 1070 performance

>>62146195
Whats the problem?

>>62146195
aren't you tried?
are gimpwroks effects enabled or not?

Incoming 'just wait' remarks

>>62146233
people also saying ti runs well, this graph doesn't look like it so I assume gimpworks are enabled which makes it amazing for vega to keep up at all

>>62146243
>just wait
>just undervolt
>just wait for drivers
>just wait

>>62146195
>Nvidia sponsored Gameworks title
ftfy

>>62146195
>resolution above FHD
Irrelevant. If I wanted that I'd buy a console.

>>62146260
>just cherry pick benchmarks from games that support your favourite nvidia card

>>62146272
how does this make any sense to you?

File: 1496252026564s.jpg (5KB, 250x177px) Image search: [Google]

5KB, 250x177px

Gimpwork + Beta game.....

>>62146286
>meanwhile amdrones post dirt 4

something janky is going on
detail and textures are not even that high yet cards are only around 60
>inb4 rendering water you cannot even see
>inb4 1 mil tris on the guns
1070 should be ploughing through that

either gimpworks or shit engine

>>62146393
people say it runs well though, with some drops, even on amd
so it's another case of "honest" and "real case" benchmarking with cranking everything up to unrealistic standards that have zero visual impact
also fucking TXAA, god..

>>62146195
>average FPS's only as if it's says anything in concrete

there you go anon, a more comprehensive review:
http://www.gamersnexus.net/game-bench/3037-destiny-2-gpu-benchmark-frametimes-graphics-optimization

>>62146410
SM*

>>62146289
It means I'll just get the cheapest one on the list, with the lowest power draw. Because that's all I need. And it should give a stable 60fps for all relevant games.

>>62146410
>people say it runs well though, with some drops, even on amd
AMDfags would say that.

>>62146195
>friendly reminder that nvidia(tm) geforce(tm) gtx(tm) cards are custom pcbs for current gen games with half the tflop perf of amd video cards
gaymers >> /v/
power users welcome on this board

>>62146442
will see this evening how my 390 handles it
but it's rarely the case if game runs bad and /v/ not full of "end of pc" and the like threads

>>62146441
if you enjoy 512x512 shadows, sure

File: newking11 - Copy.jpg (149KB, 1016x849px) Image search: [Google]

149KB, 1016x849px

>>62146195
Poor 980Ti getting gimped hard. It was on par with GTX1080 year ago... They dont even try to hide it anymore.

>>62146462
>having shadows enabled

Why the gap between the 1080 and the ti is that huge?

>>62146481
It makes no sense to me. Its like they have some kind of a Ti switch.
>1080Ti detected
>enable +30% performance mode

>>62146472
New games are optimized for new GPUs, as drivers and game engines mature they begin to utilize features only seen on new GPUs.

File: [2017.08.29_22.16.55].jpg (2MB, 3840x2160px) Image search: [Google]

2MB, 3840x2160px

yeah, nah
there is something janky as fuck going on
something is eating the shit out of computes

>>62146502
good chance of that

>>62146195
Raja should resign. Ditch HBM2 for a while.

>gimpworks beta title

File: perfrel_2560_1440.png (35KB, 500x890px) Image search: [Google]

35KB, 500x890px

>>62146195
Wow, nice cherry pick. Too bad about the composite results from the end of that same test though.

>>62146580
I can see what from these screenshots, global illumination done by nvidia rays, same shit was going on in fo4 for no reason.

>>62146580
>post yfw activision is just using destiny 2 on pc to mine ethereum

>>62146195
delete this

>>62146472
I think the difference is from the fact that most people won't be running their 980Ti benchmarks OC'd for some ungodly reason.

>>62146195
>Blizzard literally states that SMAA is broken
>on their site
>somehow it is relevant benchmark

>>62146645
>980ti is worse than FuryX
>And 1070
lol what?

>>62146747
980Ti's always been ~5-10% behind the 1070 with stock speeds.

>>62146747
drivers got better, and that's coming from nvidiabutterup

>>62146703
Yeah, its an absolute monster if a card when overclocked. Mine would hit a solid 1470mhz, many people got much higher than that even. So many benchmarks never account for that 30-50 percent frequency uplift, outside of launch day reviews

>>62146868
Mine only stays stable in really picky games like OW/Hitman at 1400, otherwise I'd be able to push it to around there in shit like 3dmark without issue :^(

>>62146195
https://www.hardwareluxx.de/index.php/artikel/hardware/grafikkarten/44084-amd-radeon-rx-vega-56-und-vega-64-im-undervolting-test.html

a wild undervolt appears

File: Comb23082017023226.jpg (255KB, 1280x1440px) Image search: [Google]

255KB, 1280x1440px

>>62146195
>BETA
>game nobody plays
Keep telling yourself that.

>>62146539
wait are you saying that maxwell 2.0 is different than maxwell 3.0? thats cute

>smaa on 1440p

>>62146472
>It was on par with GTX1080 year ago...
No it wasn't, it was always slower than the 1080 outside of a few anomalies.

That said, it was as fast as the 1070 and now it's not so the gimping is definitely real.

>>62146195
How on earth all of those cards perform so low? The game doesn't look that special.

Also, fuck 980ti gimping.

>>62146580
I literally can't tell the difference between these screens.

>>62146979
390 was as fast as a 970 at launch, now it's 10% ahead. times change.

>>62147118
3 of them 50 fps faster than the other, are you blind?

>>62147190
u fukn wot m8, there's no FPS counter in those screen

>>62146868
The average of is 1392 on hwbot. Also most cards already boosted up to 1200 when tdp and temps allowed so it's actually not a huge improvement in reality.

>>62147189
I'm aware, the 970 was always shit though.

>>62147189
Fucking hell. I just bought a used 970 over 390. Thought I was getting a good deal but should fucking never trust Nvidia fuck.

>>62147381
The 390 is pretty shit too, I bought one on release and never got good frames in new games at the time.

t. assmad amdrone who will be buying Nvidia next generation

>>62146747
NVIDIA SourMilk® technology

>>62146393
Have never played/seen the game but your making me interrested.
Can someone that owns the game use PerfStudio to see how much tesselation is happening on models and post screenshots? This was also a huge problem with Crysis 2 with DX11 enabled. Would be funny if it's also rendering water when there isn't any like with Crysis 2.

There's always so much look at these benchmarks in these threads without looking into why and discussing the technology responsible for it. Let's make this threads /g/ instead of /v/.

>>62146195
Why is scaling so bad?
1080 is double the 1060, but it's only 40% faster.

>>62146472
No it wasn't. Since launch it's been 1060>=980 and 980Ti>=1070 with the 900 series having higher power draw.

I hope the 56 turns out to be good. I've got a 980Ti so I don't really need to upgrade, but if the shader and drivers get fixed and the undervolt meme has some truth to it, maybe it'll be worth it in a year when prices drop.

>>62146580
Current state of pcgayming, it a snake oil

>>62148105
>when prices drop
That doesn't look likely

>>62146580
Real time graphics is all about tricks that emulate lighting. Its like everything is a bad approximation of the real thing, so when you want more realism the approximations have to become more accurate and difficulty goes up rapidly.

>>62148210
In a year when ethereum tanks and Volta is out, AMD won't have a lot of a choice if they want to keep selling.

>>62147838
open beta in 2-4 hours
download it yourself
but it's tessellation again, god rays turned to 11 usually do that

>>62148271
Does it have a Linux port? I don't have a Windows machine ready for testing.

File: newking1 - Copy.jpg (108KB, 1568x724px) Image search: [Google]

108KB, 1568x724px

>980ti
..............

File: newking5 - Copy.jpg (139KB, 1006x753px) Image search: [Google]

139KB, 1006x753px

What the heck man...

File: newking8 - Copy.jpg (132KB, 998x743px) Image search: [Google]

132KB, 998x743px

F

>>62147815
>>62146778
>>62146747
1070 was alway faster than 980ti idiot...

>>62146539
no company optimizes for cards, the optimize for card features.

If your card takes advantage of something it will run better on that card, what we see here is nvidia sunsetting driver support for the 900 series and only offering up game breaking fixes rather than any support after the fact.

File: 1503561450718.jpg (437KB, 2560x1440px) Image search: [Google]

437KB, 2560x1440px

>>62146962
>people play dirt 4

>>62148439
It was faster than stock 980Ti. OC 980Ti was on par with GTX1080...

>>62146979
at launch a 980ti was within a 2% deviation above or below a 1080 with, there was no real performance uptick from old nvidia cards to new ones, and there was no better performance per $ uptick at eithers launch because of the founders edition and then 3rd parties launching cards at or above founders.

>>62148493
>OC 980Ti was on par with GTX1080...
Autism: the post

>>62147909
likley vram

>>62148295
I would've liked to see Vulkan performance
the devs made their engine multi core from the start

>>62148493
Not it's not, we can see in the posted graphs it's only faster than the 1080 in one game

>>62148488
I don't think you understand, those dirt 4 benchmarks were done with no AA or CMAA. Which means the drivers for Vega haven't been optimized fully yet. AMD knows this yet they hastened to release pro drivers first because of miners and professionals.

But that also brings up an important point: is AA necessary at a 1440p resolution on computer monitors? Most people won't be using their 60" 4K flatscreen TVs for gaymen.

>>62146580
>>62147118
>see top left
>first thing I spot is the aliasing
>"surely that'll look a lot better on high settings"
>go to bottom right
>aliasing is still there
I mean the shadows and lighting are a bit better, as well as the textures, but highest settings are probably incredibly disproportionally difficult to run compared to low.

File: 1501458960889.jpg (966KB, 2661x1638px) Image search: [Google]

966KB, 2661x1638px

Both are easily overclock able.

64 = 1080
56 = 1070

Just because they are 18 months late use double the power and have no stock and require tonnes of tweaking and have no aftermarket coolers and are insanely over priced doesn't mean amd fucked up...

No seriously ryzen is great


Wait we're talking about Vega ain't we?

>>62146472
>>62146472
>Poor 980Ti getting gimped hard. It was on par with GTX1080 year ago...

t. bitter 980ti owner

>>62148978
Learned my lesson the hard way.

t.ex 480GTX owner.

>>62148742
They need to release tweakable bios,much progress has been done on loonix where you can boot bios on top of the existing one but you cant do that on windows.

File: __________________.jpg (96KB, 850x566px) Image search: [Google]

96KB, 850x566px

>>62146195
What makes the 1080Ti so different that is absolutely rapes everything else

>>62146441
>26.8fps

File: Untitled.png (32KB, 655x921px) Image search: [Google]

32KB, 655x921px

So what about fixing voltages and adjusting the clocks by yourself?
Did anyone but that one german site do it?
I've only seen individuals on hw forums do it everyone else just moves the voltage a little and calls it a day

>>62149120
Drivers,nvidia does some voodoo on top of dirty tactics

>>62146472
980ti has matched a 1070 from day one

>>62149120
no_gimp_365.41

>>62149143
>>62149149
is there broofs of this what is stopping Zhangxui from using chink magic to use Ti drivers on a regular 1080

File: image.jpg (19KB, 326x416px) Image search: [Google]

19KB, 326x416px

>>62149136
>gains 10% performance by using less voltage
>any coolers are confirmed to be gimping the card at any level
>people fighting over how to get 100% of their GPU working
Why is amd such a god damn mess

>>62149165
Experience i'd wager.
Nvidia was doing it a long time ago, after that physx driver spoofing on AMD they really upped their driver locking game and it's not easy to do much.

File: 980Ti benchmarks from Anandtech.jpg (776KB, 3360x1944px) Image search: [Google]

776KB, 3360x1944px

>>62148533
>at launch a 980ti was within a 2% deviation above or below a 1080 with
you're wrong
http://www.anandtech.com/show/10325/the-nvidia-geforce-gtx-1080-and-1070-founders-edition-review/19

>>62149206
No idea.
They did the same overvolting firestorm on polaris.
This is so incompetent it looks like someone is doing it on purpose

>>62149165
example new Final Fantasy XV

Supports Nvidia
>Nvidia Game Works
>Nvidia Turf Effects
>Nvidia Hair Works
>Nvidia Flow
>Nvidia HFTS
>Nvidia VXAO
>Nvidia Ansel

AMD tech
>none

>>62146580

https://www.youtube.com/watch?v=d5ZsaavKNR8

>>62146286
>games that support your favourite nvidia card
>games that support nvidia

Are you bragging about not having support for AMD? I'm confused.

>>62149232
If there is X16 AA doesnt mean you have to use it on 4K.
I mean people do but w/e

>>62149217
What's the point of a liquid cooled model if 100% of the cards throttle at stock anyways

>>62148363
This is one of the most cherry picked pictures I've seen today.

Comparing a fucking Lightning model vs stock cards is just beyond retarded even for /g/.

File: 1486406923017.gif (3MB, 286x258px) Image search: [Google]

3MB, 286x258px

>all these salty 980ti owners ITT

C'mon guy, if you dont have money to upgrade every gen for those mad gains just roll with AMD.

File: ehhhh.jpg (131KB, 492x462px) Image search: [Google]

131KB, 492x462px

>>62149245
Hey clown, long time no see. Where have you been, jail?

>>62149255
Don't ask me.
They have someone retarded enough to okay a watercooler while ignoring that he ramped up the voltage so high it consumes more power than a 1080ti while the card runs using 1070 levels of performance if you bother to move the voltage slider in the driver.

>>62149318
>long time no see
48 hours is like 2 weeks in neet time for you.

>>62146272
You're a special snowflake, aren't you?

>>62149289
>Comparing a fucking Lightning model vs stock cards is just beyond retarded even for /g/.
Whats the difference when GTX1000 series cards overclock themselves to the max with turbo boost 3.0? Also you can see there stick 980Ti included into the chart.

Even if AMD magic out primitive shaders it won't matter since the cards are all selling for $100 more than they should

>>62149410
more like 200-300 dollars or euros above the msrp

File: 1469892718501.jpg (68KB, 633x758px) Image search: [Google]

68KB, 633x758px

>>62149428
When will minerbulls stop cucking us bros?

wait, is SMAA better than MSAA?

File: 1448346725022.jpg (32KB, 305x315px) Image search: [Google]

32KB, 305x315px

>>62149330
Awwww, I missed you too bby.

File: 1493696194352.png (2MB, 1200x891px) Image search: [Google]

2MB, 1200x891px

>>62149443
AHAHAHAHAHAHAHHAHAHA

File: 1470032812996.jpg (83KB, 622x757px) Image search: [Google]

83KB, 622x757px

>>62149461
A-at least i will buy it cheaper once you're done haha

>>62149453
I was trying to goad you into telling me how long its been since I posted since I was curious but you don't seem to be working.

>>62149461
>that cable management
Ew. Truly worst of the worst

File: RX Vega 56 vs GTX 1070 32 game aeverage.jpg (185KB, 1600x900px) Image search: [Google]

185KB, 1600x900px

>>62148488
At least post the newest one.

>>62149069
Is the 1750mhz GPU clock limit still there? God this GPU gen is such shite for tweaking
>>62149120
Brute Force vs relying on clever programming and architectural optimization.

Vega 64 will probably never reach it's full potential just like fury x and 290-390 because amd can't write driver's for shit.

They made something special with mantle and vulkan than us it for all of 2 games bf4 and doom then we never see anything great like that again.

Look at all those people claiming dx12 would save amd perofrmance in the mean time it runs even worse.

>>62149496
I hope you're not implying 5% performance is worth getting a card that consumes so much power, although winter is coming so it could be worth it.

>>62149448
Objectively?
Yes in all cases since it doesn't take any performance by comparison.
MSAA will work better in a few cases but it costs crazy levels of performance by comparison.

>>62149410
Nvidia does the same with blurring textures on the fly and cutting geometry for more FPS.
They are not hiding it.

File: 1496876457729.jpg (186KB, 900x1200px) Image search: [Google]

186KB, 900x1200px

>>62149224
All those proprietary crap, And Gaymer think this is good....

>>62149531
It's not worth getting any card now.
And it's not like you can't move a slider in the driver suite and make it consume the same amount of power as a 1070.

>>62149448
>he doesn't use CMAA
look at him and laugh

File: image_17.jpg (44KB, 499x500px) Image search: [Google]

44KB, 499x500px

>>62149482
I've been stalking you in almost every thread your trip shows up in. I haven't seen you for ~4 days. N-not like I was worried about you or anything you 馬鹿.

>>62149547
oh yes and G-Sync but no Freesync.
Nvidia is working with the studios directly thats why they have better shit on average.

>>62149136
Gamers Nexus did it during a live Youtube video. They did not spend much time on it but got it down to 1070mV which matches with that German sites results.

>>62149448
Is SMAA the best cost/performance type of AA?

>>62149554
>CMAA
The only one anyone will laugh at is you.
You can't equate a method that works in every program out there to a method that requires the developer to implement it.

>>62149580
Yeah a dude on here linked me that video.
They streamed it when they got the card, but it was really rushed and not conclusive.
They did prove the concept though, i wish more people tested that shit

>>62149607
You literally dont know shit CMAA as anything can be forced through drives and works better than it fucking should for minimal performance impact.

>>62149607
Are you retarded?

>>62149633
How do i force it then.
I inject smma with the injector by mrhaandi or reshade, how do i inject CMAA?

File: 1439761159915.png (75KB, 342x245px) Image search: [Google]

75KB, 342x245px

>>62149496
>mfw mainly playing PUBG and considering to buy vega 56

>>62149601
Yes. I've been ignoring AA types delivered by games for years while injecting smaa with RadeonPro instead. It visually looks like 4x MSAA while costing about 3 or 4 fps.

File: 3D-Settings-1024x546.png (110KB, 573x217px) Image search: [Google]

110KB, 573x217px

>>62149658

>>62146195
drivers are massively immature. frametimes are all over the damn place for vega, these chips are in the same place the 7950/7970 were at launch. Pretty good, but with a huge amount of room for improvement.

>>62149700
Wait
It's intel integrated gpu drivers?
lmao
I have a real GPU sorry.

>>62149554
Even intel admits CMAA is worse than SMAA. IN terms of quality:

FXAA < CMAA < SMAA < MSAA

In terms of cost:

MSAA > SMAA > CMAA > FXAA

>>62149756
>you cant use free performance bump while looking good because ????

>>62149778
I really dont like how FXAA looks on lower resolutions so i dont really use AA in the first place.

>>62149780
Sorry i like playing above 20 fps with a real gpu.

>>62149820
sure bud

>>62148720
I only notice 2 real differences:
1) The wall immediately on the left has a lower res texture on the top 2 pictures than on the bottom.
2) While the resolution of the texture on the floor tiles is lower, there is also significantly less aliasing as a result in the Low settings pic.

>>62149496
>buying 1070 Founders Edition more than a year after release
Can you even find them now? All of the pre-overclocked 1070s are cheaper than a Vega 56.

>>62149778
>FXAA ever
FXAA is about the worst AA method I've ever seen. It just blurs everything. There's a way to achieve FXAA without a performance penalty at all. Covering your screen in a thin vasaline layer.

>>62149904
He used both a FE and a aftermarket 1070 in his testing to be fair.

http://www.youtube.com/watch?v=zfLuJajkwcY

Not sure why he used the FE in the overall chart and not the Gaming X though. It would make the cards tie pretty much otherwise.

>>62149939
If the products released at roughtly the same time I would get using the founder's edition but it's been more than a year now and pretty much every 1070 on the market is overclocked with a dual fan cooler.

>>62149687
really? I'm usually avoiding injectors. How reliable is it?

>>62150071
Haven't had any problems at all with it. Works like a charm as it should.

>>62146195
>he fell for amdeme

>>62146261
>nvidia sponsoring meaning shit all at game title that cost 300mil to develop
Are you serious? These gaming companies could literally buy nvidia if they wanted, they dont give a shit about nvidia or amd

>>62150328
lol

>check jewtube comments on gaymersnexus video on Destiny 2 beta (pre-alpha goyworks edition)
>it's full of stupid gaymer kiddies jerking their dicks off over how doomed AMD is
>check shitty thread on /g/
>it's full of stupid gaymer kiddies jerking their dicks off over how doomed AMD is

Where were you when /g/ is basically the same as Youtube comment section retards?

>>62150595

>Where were you when /g/ is basically the same as Youtube comment section retards?

11th November 2011.

>>62150717

well thats oddly specific. as some guy from b can you tell me what happened then?

Vega 56 is worth it. I got my 1070 for $380 days before the mining boom

>>62150819

"[...] is an open world action role-playing video game developed by Bethesda Game Studios and published by Bethesda Softworks. Wikipedia
Initial release date: 11 November 2011"

>>62150956
thanks but jesus its been that long since it was released? why is this thing even relevant anymore

>>62150595
Yeah we basically became /v/ along the way jerking of to benchmarks without understanding what it means instead of technoli/g/y that dives into the how and why.
I'm wondering what effect enabling the primitive shaders will do to effect the goyworks performance. Vega should according to the whitepaper be able to discard 17 triangles per clock compared to the 4 now. This should increase tesselation performance by a lot if there is more early culling of triangles thanks to the primative shader. However I don't know the specifics of all the goyworks technologies from Nvidia but if I where to guess I think it relies a lot on tesselation since Nvidia gpu's have been way better at tessellation than AMD's gpu's. Would the primative shaders close the gap when the're enabled is what I'm wondering the most.

>>62148488
more than literally anything on the microshit store

File: 1341038096354.jpg (112KB, 449x401px) Image search: [Google]

112KB, 449x401px

>He bought a 1070

>>62146539
>blogger knowledge

>>62151076
You can offload everything goyworks-related onto the "primitive shader".
Should be nice.
WELL WHEN IT'D ACTUALLY WORK RIGHT?

>>62146195
It's literally an Nvidia sponsored Beta

>>62151251
Yeah that's what I'm suspecting.
The launch of VEGA was an absolute disaster and I won't defend it. I think the reason AMD released it before being finished is because of the shareholders which is understandable but pretty messed up as well.
As to the actual technology I think it's pretty interesting. I almost played any games at all so I'm not interested in buying it and I think the price is too high because of supply, demand and HBM cost and also Nvidia has better cards for gaming at the moment but I really like keeping up with VEGA because of what it's trying to do.

The way the caching has been overhauled is a really nice addition to the architecture. Since it can access the cache before and after a calculation should relieve a lot of the frontend bottlenecks that GCN was suffering from. I like GCN but also understand it's weak points and it seems RTG did as well. Right now I think there is a lot of performance to gain from the architectural changes. Like for calculating geometry the way NGG fast path works would make a lot of improvements overall. Also it should be very nice for vr since it only has to calculate the geometry once instead of twice. The way the primitive shaders work makes early culling possible and should accelerate tessellation a lot and changes like the HBCC are awesome for workstation applications as well.

I like VEGA as a test bed for Navi and testing the waters for the upcoming Raven Ridge apu's. I think VEGA also should be pretty efficient if all the features are enabled. The reason I think it's consuming so much power is that AMD pushed VEGA to hard giving it to much voltage and clocking it way too high and not because the arch itself is inherently inefficient.

I wonder who was getting married to their 9yo cousin.

https://twitter.com/gchip/status/902578993920073728

>>62146261

>Just use AMD sponsored games like we do!!!!!!!

>>62151579
Well, both Instinct and Pro launches would be good.
It's just the gayming launch is bad.
I don't even know why did they ever launch it.

>>62151689

>not wanting chocolate loli

>>62151304
People stubbornly ignore that.
As for why,you tell me.

>>62151733
These usually run well on everything.
>>62151771
Pedo pls.

>>62148656
>Most people won't be using their 60" 4K flatscreen TVs for gaymen
Because they will be using their 4k computer monitors.

>>62149469
Good luck, bitcoin just hit $4500.

>>62151737
I'd say shareholders. Holding of on the launch would have been a good thing in this case.
However next to the pro drivers I've noticed there has been a lot of work done on the amdgpu linux driver as well. AMD unfortunately doesn't have as much coders for the drivers so there almost hasn't been any work done on the gaming drivers unfortunately for this launch.

I hope it will be easier for AMD in the future since the arch is more software defined now and if the drivers work it should be up to the engines to optimize for the gpu instead of the drivers.

>>62149553
The 1080ti is the only card that isn't overpriced, and it is only worth the $700 if you are gaming in 4k or want max fps VR.

>>62150983
Because it is cheaper to milk the game for every last penny by relreleasing it with minimal tweaks then it is to make a new one.

>>62151910
Who cares there's no RX Vega AiB cards.
Fuck, there's no cards at all.

>>62152069
I ordered a 56 yesterday. They told me on their webpage that is was in stock. But now they tell me they won't be getting more stock until Friday. Retailers are lying scumbags and AMD are just incompetent.

>>62152069
I've already said the launch was atrocious and that I won't defend it.

>>62152192
AMD has nothing to do with that, the supply chain is very long. It's fabs -> packaging -> AiBs -> distributors -> resellers.
Now find who's guilty.

>>62152218
Distributors

>>62152218

Jews. Its always the Jews.

>>62151733
why does it seem like the AMD sponsored games are usually always better gamer over all?

>>62152387
because AMD doesn't sponsor trashy console ports.

>>62150328
>Are you serious? These gaming companies could literally buy nvidia if they wanted, they dont give a shit about nvidia or amd

Wrongo.

>>62150328
No.

>>62152387
Because AMD uses more open standards which everyone benefits from while Goyworks games use proprietary techniques that puts extra burden on competitors (and a bit on Nvidia).

Fun fact. Nvidia cards are faster with AMD's TressFX than with their own Hairworks.

Goyworks needs to die.

>>62146195
>Fury X STILL slower than 980 Ti

F-Finewine right guys?

File: 7s5QTxg.png (600KB, 800x533px) Image search: [Google]

600KB, 800x533px

I will always buy Nvidia™ because I only play games The Way It's Meant to be Played™. Nvidia also pioneers innovative new technologies like PhysX™, Gameworks™ and the highest quality driver to ever grace Windows.

When I boot up with a brand new Nvidia™ Geforce™, I can experience the game just like it's meant to be played. Nvidia™ also delivers a far more smooth experience.

Nvidia Geforce™ is also very power efficient. A graphic card is the most power hungry device in your house. Refrigerators, AC units, water heaters, dishwashers, lights, etc all use significantly less power than a graphic card. Which is why Nvidia™ puts gamers first by ensuring that their gaming experience is of the highest quality while looking out for gamers by giving them the most value in their electrical bill

At this point in time, there's really no reasons to consider an AMD graphic card at all. I tried one one time, it caused so much heat that it exploded. It also consumed so much power that it gave on an EMP and destroyed the rest of my computer.

Nvidia™ also pioneered how useless GPGPU is with CUDA™. Years ago, everyone thought GPGPU, CUDA™, and OpenCL were the future. Now, Nvidia™ has removed those useless features from their GPUs and increased efficiency. Now you can save thousands a year in electricity thanks to Nvidia™ ensuring that useless features like GPGPU are "optimized" for gamers.

It's quite clear that OP's an AMD shill trying to convince you to settle on something less than The Way It's Meant to be Played™. Nvidia™ is the only real way to play games. We have seen recently that they offer incredible libraries for software developers like Nvidia Gameworks. He is probably too poor to afford the Nvidia Geforce Experience and can not afford to play any games The Way It's Meant To be Played™.

Don't be a poor gamer with bad drivers and a huge power bill. Play games with the Geforce™ Experience™: The Way It's Meant To Be Played™

File: chrome_2017-08-29_15-59-14.png (14KB, 802x259px) Image search: [Google]

14KB, 802x259px

>>62146195
Curious if that benchmark was ran using crimson 17.8.2

>>62153517
Thanks leather jacket men.

>>62153621
I don't know but I'm guessing that it won't matter to much since the game may use a lot of tessellation from what I've read here.
The real performance improvements should come from enabling the primitive shaders since they should support a culling rate of 17 triangles instead of the 4 now. That would give real performance improvements in heavily tessellated games.

>>62149461
Get a real job, mother fucker.

>>62149211
wow what a shock not looking at ocvoc

File: perfrel_2560_1440.png (39KB, 500x1050px) Image search: [Google]

39KB, 500x1050px

>>62146472
>It was on par with GTX1080 year ago
No, no it was not.

>>62149574
Are you a cute trap that stalks some random tripfag or something
I'm confused, what am I witnessing here

>>62151579
I'm confused
The tech is there, but AMD pushing their card (They always have hardware wise) is negated because programming is THAT much more efficient?
Is the 1080ti running less bulk but being that much more effective because of it? Damn

>>62154438
I guess a comparison may be a person with 6 arms folding envelopes. He (VEGA) is only using 4 of his arms and tries to compensate by trying way to fast with 4 arms (high clocks and power usage) while he could have been faster using 6 arms even at a slower pace (I think it will achieve more performance even if clock speed and power is lower).
This might be a weird comparison but it's the best I could come up with. AMD should focus now on making drivers that make VEGA use the 6 arms in the comparison I made.
I hope it still is understandable.

>>62151112
>wait vega

>>62146472
>Poor 980Ti getting gimped hard. It was on par with GTX1080 year ago... They dont even try to hide it anymore.

So? You should be upgrading your video card every 18 months / 2 years anyway

I don't understand why people buy Nvidia cards. All the best monitors right now only support Freesync

>>62152387
Mankind divided says hello

>>62154817
>best monitors
This is the opposite

>>62154438
AMD cards have "traditionally" had trouble in the sense that AMD could not feed all the hardware, thus essentially leaving the card underutilized to some degree. Their drivers are also shittier, especially in DX11 and below, where they used to need better single-threaded CPU performance than NVIDIA.

This was somewhat alleviated in Vulkan/DX12, which is why AMD cards saw a large boost while NVIDIA still does not. NVIDIA never suffered from underutilization and shit DX12 drivers like AMD, so there wasn't any place to squeeze much extra GPU-bound performance from.

>>62155355
So nvidia doesn't have shit dx12 drivers.
The already managed to get dx12 levels of efficiency with dx11.
An example is Nvidias driver in dx11; it's heavily multithreaded.

Dx12 just takes everything AMD sucked at and removed it.
Driver overhead and drivers in general is something AMD has failed at in the past. But, dx12 changes all that.

>>62155479
NVIDIA's DX11 driver is very good, yes, but that doesn't mean it still isn't heavier on CPU than DX12, even for NVIDIA. Sadly there are few technically-sound DX12 games/ports, but you can have a look at RotTR (especially DX12 mGPU) and AotS, even NVIDIA performs better in DX12, especially in natively-handled mGPU instead of SLI.

>>62149461
If I had one wish, I'd wish for all Cryptocurrency to immediately crash and never become viable.