Rx Vega vs 1080ti blind test

Get in here
youtube.com/watch?v=2CE-wSU1KMw
Article
hardocp.com/article/2017/07/26/blind_test_rx_vega_freesync_vs_gtx_1080_ti_gsync
Looks like the Vega will deliver

Other urls found in this thread:

youtube.com/watch?v=TuOQ-E2czNE
youtube.com/watch?v=-Ze5FLQtV94
youtube.com/watch?v=jRsDOftpjs0&index=14&list=PLVM1qRtVEeEoFkoPmBiunxP-JEpYqv4Ps
youtube.com/watch?v=THeBms4QEEM&t=1569s
forbes.com/sites/jasonevangelho/2014/05/12/amds-project-freesync-gets-momentum-as-adaptive-sync-gets-added-to-displayport-spec/#5589a0873c58
twitter.com/SFWRedditGifs

I'm fairly sure my R9 290X could almost perform as well on Doom.

That's a terrible game to gauge performance.

>no graphs
>"feel"
This is audiophie tier compariosn. I was right about the whole "guise look at this gpoo+memesync is the same as 1080+gsync" marketing stunt.

yeah because only blind person will buy AMD lmao

R9 290x master race here.

Why is AMD so bad at advertising?

NO CHARTS, AMD DEAD ON ARRIVAL

Wow the comments got brigaded by WCCFtech trolls. Fantastic.

Do we like Kyle now amdbros?

Kyle got over his butthurt maybe.

He was pretty salty during 480 release.

>That's a terrible game to gauge performance.
Eh, depends on what you mean. It's EXTREMELY well optimized, and manages to get the most out of your hardware. But that's obviously not indicative of games as a whole. For example, the Fury X gets higher FPS than the 1070 in Doom but that's not the case in literally any other game.

You're right, until we see other games this doesn't mean anything but that's not to say Doom doesn't run on magic.

As we expected, most of the professional and industry people agreed that the 1080ti system was faster.

AMD's own attemps at pulling the wool over our eyes might work for fanboys but not for experienced gamers

Why are people bitching about Doom performance when Vega FE runs doom like shit compared to Nvidia?

Might have worked for Fury and older cards, but not for Vega

That's not what I just saw.

Did we actually watch the same video?
Out of the pros only one claimed System 1 to be faster, was that tourney dude, the rest stuck with System 2.

>I substitute this reality with my own!

(you)

>It "feels" better
They did the same with bulldozer you know?

AMDrones are driven by feelings.

You can't expect any logic out of anti-AMD shitposters. They live to act like stupid pieces of shit and constantly attack AMD like mindless zombies.

They've done it

Freesync > gsync

That's because Bulldozer was so warm it made players feel comfy during the winter,

I don't even doubt poojets plug the Vega one with freesync and the other without gsync. Even that way doom is such a smooth game it runs like 200hz at 60hz it's very fluid.

Intel fucked up, why release Skylake-X in summer instead of winter?

He watched the video where he sits there with his dick in his hand masturbating to the thought of AMD going bankrupt so NVidia can fuck him in the ass for all eternity.

PCB-melting hot, and as fast as a 1080 at best.
Does AMD marketing actually buy the bullshit they feed to consumers? Do they actually think people are that stupid?

kyle was never an amd fan... whoever was alive during the 8800gts era and remember the lies he was trying to pass on the consumers knows...

Just look at amdtards vehemently defending their cult.

cause intel managed to make the literal housefire the fx 95xx to look decent

Why would you use DOOM for an xsync test? The main benefit of gsync and freesync is it minimizes the feeling of framerate drops due to dynamically changing refresh rate, it doesn't matter if you're playing a well optimized game at 200fps. They should have played something with lot's of variable amount of stress on the CPU like Battlefield 1 or CoD, since those types of games stand to benefit more.

No single GPU can run fucking doom at 200FPS at 4k you dumbo.

because it would make amd look bad

>4k
did you even watch the video retarded drone

The monitors are 3440 x 1440

>both @ 100hz
>1080ti
>DOOM
>100hz

No wonder Vegan performed similarly.

what would Sup Forums look like if pic related turns out to be real?

1080ti is 50% faster than Vega

If nvidia had done this, the thread would already be in the 200's posts, but it's ok when amd does it.

They also did it with Eyefinity but whatever.

Lots and lots of butthurt nVidia fanboys and lots and lots of shitposting AMD fanboys.
Fuck GPUs.

I would contemplate selling my Fury X and getting one instead of waiting for Navi.
But no, absolutely impossible for that to be true, I would personally fly to Thailand and pay a ladyboy to fuck me in the ass,

Literally impossible because then AMD would have actual marketing instead of "muh feels"

I was implying they are gimping the Ti by limiting it to 100hz.

Why would they do actual marketing?
To let jacketman cut the prices on anything GP102 several month before Vega even launches?
Come the fuck on it would be classical ATi mistake.

Sup Forums absolutely BFTO what's your excuse now to not get an AMD GPU?
youtube.com/watch?v=TuOQ-E2czNE

Fake news.

RX 480 gets ~36 with vulkan. Assuming the shop is "real", it would mean a hefty 400% performance jump over RX 480. Impossible or rather very improbable.

Vega FE gets about ~80 FPS in Vulkan. That would be a near 80% performance increase over their FE version. Impossible unless they have something new to release.

This feels like the xbox 360 vs. ps3 console wars all over again.

To build actual hype and public interest, which drives sales. Which is what you do when you have a product worth a damn.

>Radeon
>sales
Come the fuck on.
People would eat shit with Nvidia stickers.
No amount oh hype would sale Radeons.

rx 480 has dogshit clockspeeds compared to vega tho

>they made those teasers that got people hyped
>after that they just dropped the ball with literally nothing and then the Vega FE which sucked
I don't get it

the 360 won that gen and nvidia is winning this gpu gen. the only people who argue are just asshurt fanboys at the end of the day

>Why would they do actual marketing?
You're telling me #RadeonRebels wasn't good enough for you?

>As we expected, most of the professional and industry people agreed that the 1080ti system was faster.
Nice bait.

Consoles are more about exclusives than hardware. Weebshit was on PS3, 12yo gaylo players were on 360.

>"Experiential tests" while frame limited to 100hz in a title even a 1080 shouldn't have a problem driving at 120fps 1440p
Clever. But not really.

Yeah-yeah youve read that plebbit post.

>tfw get to now play those games but with actual hardware.
youtube.com/watch?v=-Ze5FLQtV94

Be a rebel
Pay double for half performance
Potato brigade

I don't go to reddit, but it's not surprising that the "test" restrictions meant to present AMD's abortion of a GPU in a positive light are obvious to others.

I was hitting triple digit framerate with my RX 480 all settings maxed @ 1080p. Very well optimized game it was amazing.

"all settings maxed"

So 2xAA? Good joke, maxed means maxed, so 16x SSAA or fuck off.

>Consoles are more about exclusives than hardware

no. consoles for the last 2 generations have been popular as budget ways to play all the big AAA games without having to dish out big bucks for a PC. when the 360/ps3 came out they had hardware which was better than the vast majority of home PC's at the time and iirc the 360 gpu was the first with a unified shader model. all the best selling games of that gen and this current gen are multiplat AAA games because a console is an affordable, plug and play way to access this library as 99% of AAA games will come to console.
the 360 had the best versions of multiplats (i.e. the games most console players play) and was thus the best console of the gen to any non-biased person. the same goes for the ps4 this gen. ps4 players even voted that they bought their console because of the resolution advantage in a poll.

tldr; 360 > ps3

Lol a Titan xp doesn't perform like 100% over 1080.
1080ti is around 30% faster and is the same die with some cores fucked

>meant to present AMD's abortion of a GPU in a positive light are obvious to other
What obvious to me is that you and others are retarded. Wait 4 days and shitpost to your hearts content after the event.

Its a 4K benchmark shop.

All this secret is only due to failure.
Ryzen leaked numbers much earlier

shhh It's sandbagging. The 50% driver is just around the corner. Have some patience.

Man if only we had numerical benchmarks instead of feelings to evaluate hardware.

>Ryzen leaked numbers much earlier
ONLY and ONLY due to frenchies getting hands on ONE Zen ES.
These frenchies were the ones to get their hands on K8 ES btw.

where are the charts

I would buy Vega day one.

...

>Blind test with a game literally designed for AMD hardware
>Still loses to the 1080ti
Top kek

>Titan XP 100% faster than a 1080
Ya no, that bench is bullshit
Fake news.

Who needs charts when you have feels.

>Our product is slower
>Our product uses more power
>Our product is priced the same as a 1080

But you can't tell the difference!
You can save 300 bucks if you buy our GPU + monitor!

This is proof if any that VEGA is a failure and can't complete with a 1080Ti and when other games shows this (LOL Doom) AMD are going to get laughed at.

t. AMD MSI RX 480 Gaming X owner

Im stealing your pasta

>Tyler Murley
>Beat Ultra Nightmare DOOM pisol Only

holy
fucking
shit

That's literally nothing
Shitloads of people have beat the game pistol only ultra nightmare with no deaths, some have even done a no hit run

This is some Grand Slam of Autism, and I thought I was weird for doing a all white mage run on SNES FF's

[spoiler]One White Mage[/spoiler]
youtube.com/watch?v=jRsDOftpjs0&index=14&list=PLVM1qRtVEeEoFkoPmBiunxP-JEpYqv4Ps

Beating Ruby Weapon with a Level 1 Cloud Solo
youtube.com/watch?v=THeBms4QEEM&t=1569s

Don't worry man, it's more common than you think
A few buds and I did a company run with borderlands 2
Can only use equipment made by one company for the entire game no matter what, I was stupid enough to pick only snipers and jakobs, they make fucking nothing

This test is retardedly stupid.
I am sure they could have thrown in a 580 and it would have felt the same

Is a free sync display pretty essential these days?

no, but it will be harder to go back to a normal monitor after using one (I think)

t. never used a freesync monitor

>I would
I'm going to*
Ftfy :^)

Get a good freesync monitor.
Research. Some freesync implementation is shit leaving you with narrow ranges or bad ghosting.

Nope
These sync technologys are gimmicks.
They are designed to keep you from switching GPUs.
Is you have a Nvidia GPU and Gsync, you won't get a AMD card and nice versa.

You can get 99% of the benefits by using MSI afterburner to limit fps to a given value, say 60fps.
Then tailer the settings so you don't dip below 60
GPU usage stays 90-99% and you keep a solid 60fps. Nets consistent frame times and low latency.

Literally the only path to success for Vega at this point is to make a cheaper 1080. That's it. There's no way they have a 1080 Ti equivalent at this point.

>They are designed to keep you from switching GPUs.
That's why G-sync is proprietary, but I'm considering that Freesync is open-source AMD sees it as a short-term advantage at best. We could be living in a world where Freesync is standard on all monitors if Nvidia would give up trying to push their shit when there's no real advantage continuing the R&D and marketing for G-Sync when they could just adopt an open standard. It could've been to Nvidia's gain to piggyback off Freesync and give up G-Sync in the long-run, and Freesync is inevitably going to become the standard.

Freesync isn't open source. The variable refresh specification that freesync uses is a VESA standard. Freesync refers to their GPU-side technique for using the DP spec to control refresh rate as well as the brand name.

If Nvidia wanted they could come up with a competing product that works on all freesync compatible monitors without even using the freesync brand name. They could literally just come up with G-sync lite or something and use the same variable refresh tech in monitors' controllers to do the same thing as freesync.

You can find articles about it from when Freesync was first announced. forbes.com/sites/jasonevangelho/2014/05/12/amds-project-freesync-gets-momentum-as-adaptive-sync-gets-added-to-displayport-spec/#5589a0873c58

>“DisplayPort Adaptive-Sync enables a new approach in display refresh technology,” said Syed Athar Hussain, Display Domain Architect, AMD and VESA Board Vice Chairman. “Instead of updating a monitor at a constant rate, Adaptive-Sync enables technologies that match the display update rate to the user’s content, enabling power efficient transport over the display link and a fluid, low-latency visual experience.”

>AMD's Project FreeSync works in concert with this technology, providing adaptive framerate syncing on the driver side of things.

AMD encouraged VESA to port the standard to displayport (originally an embedded displayport spec used primarily by Intel to limit refresh rates to 30Hz to save battery life) but it's still a VESA standard and AMD has no claim or right to bar Nvidia from just taking advantage of what already exists in monitors. Nvidia could do whatever they want with 0 approval, not even some sort of acknowledgement, from AMD.

>never used adaptive vertical sync technology
your opinion comparing gsync and no gsync must now be discarded, user :(
at least above user admitted to not using gsync/freesync. I use gsync and it's pretty sweet you don't notice it going forwards but you notice it going back. kind of like how I didn't see just how hard my pc roflstomps my xbox until I played BF4 on xbox after playing in on my oc
this is good advice, as not all freesync is made equal. this is the advantage of gsync as it is technically better (just), but all gsync is equivalent provided nvidia hasn't come up with a better way to do it over previous years
from what I've heard top end vega is supposed to be near a 1080 performance wise

so what do you guys think is actually happening here? Most (all i think) preferred the rx vega over the 1080 ti. A hardware editor and gaming journalist both purported the rx vega was noticeably better. this really gets my hopes up for vega but what concerned me is that the video editor himself said at the end that it was a surprise to him that rx vega won the "muh feels" contest. is it possible he had pre-benchmarked the rx vega and was surprised to find people preferred it more? what's goin on here
>pic unrelated

Wait for the benchmarks. GPU performance is something we can actually benchmark and check. Heck it is what? 4 more days?

Doom, amd best case again...

It's fairly obvious how nervous they are with this video
Releases a feels vs reals video first, 4 days ahead of time, with 0 numbers
Putting the people life stories in a sub box to try and make them appear credible "40 years of gaming, ate a twix once, beat doom".

If blatantly obvious that whatever Vega is, they are really, really trying to justify it

I remember when amd pulled out blindtests for their bulldozer APUs.
>That's a terrible game to gauge performance.
Doom is 90% corridor/tunnel play. It's a terrible game to guage vulcan apu performance and/or hardware performance.

>feels vs reals
Kek

But seriously, this looks bad
Not releasing numbers says they have a dud on their hands.

Graphs or gtfo ..
Sage

>you don't notice it going forwards but you notice it going back
I did a blind test with my roommates
Noone was able to determine the difference between (GSync on + no fps cap) vs (gsync off + MSI AB fps limit)

We were able to determine the difference when there was (no fps cap +Gsync) vs (no Gsync + no fps cap)