AMD Radeon R9 480 (non-X) Performs Close to R9 390X

[btarunr]

In all the 16 nm NVIDIA "Pascal" GPU fervor, it would be foolish to ignore AMD's first "Polaris" GPUs, built on the more advanced 14 nm process. Hot on the heels of reports that a fully-equipped "Ellesmere" GPU based Radeon R9 490 performs close to NVIDIA's GeForce GTX 980 Ti (and AMD's own R9 Fury X), with nearly half its power-draw, new numbers from an early GFXBench run suggests that its cut-down R9 480 (non-X) sibling performs close to the Radeon R9 390X. The R9 480 succeeds the currently-$200 R9 380, and its prospect of offering performance rivaling the $400 R9 390X at half its power-draw appears to meet AMD's "generational leap" claims for the "Polaris" architecture. Similarly, the R9 490, based on a better-endowed "Ellesmere" chip, offering performance rivaling current $600 GPUs at a $350-ish price-point (succeeding the R9 390), appears to meet expectations of a generation leap.



View at TechPowerUp Main Site

 

[darkangel0504]

200 usd, so interesting

 

[T8RR8R]

Very intriguing. Hopefully we'll have a very high value price war between the 2 big teams this time around.

 

[MrGenius]

The benchmark link is busted. So we're still left in rumorsville for the time being.

EDIT: Change that to semi-busted. I tried it a dozen or so times before it finally worked.

 

[Tsukiyomi91]

R9 480 non X performs as close as R9 390X with half the power draw? sounds interesting for Polaris. I'll give it a few months time to let them (AMD) to give you guys (TPU) & some unbiase reviewers a sample card for proper, unbiased benchmarking & see if the claims are true, as per what AMD has promised. Since the link is dead, we shall wait & see for TPU's turn. =)

 

[GreiverBlade]

The benchmark link is busted. So we're still left in rumorsville for the time being.

R9 480 non X performs as close as R9 390X with half the power draw? sounds interesting for Polaris. I'll give it a few months time to let them (AMD) to give you guys (TPU) & some unbiase reviewers a sample card for proper, unbiased benchmarking & see if the claims are true, as per what AMD has promised. Since the link is dead, we shall wait & see for TPU's turn. =)

neither of the links in the original post are down, at last not at the moment seem that the R9-480 will be interesting at 200$... as if ... more 500$ and above were i am ... baharf ... (thanks greedy retailer and etailer )

R9-490 rivalling 600$ GPU? wait, it could rivale my 980? I know i know he talk about 980Ti and Fury X (which are above 700chf/740$ for me ) well were i am a 980 is 620chf/655$ (only counting "in stock" model )
gotta love pricing depending on the country ... mind you ... not all Swiss are working in a bank and crawl into a pool of 1000chf note

 

[arbiter]

"AMD 67DF:C4" just wondering how they know its a "r9 480" based on that number? Also another thing "API OpenGL" Um wonder if there is more to this then meets the eye.

Would also question how look at gtx 980 and 980 ti, both cards run about indentical numbers so would tell me the tests they used are worthless since if 980 push's same fps as a 980ti when there should be least 20-30% gap. 980ti does 60fps but a titan X which is SAME gpu does 209fps. yea there is something wrong with that.

 

[Assimilator]

There's a reason ZERO tech sites use GFXBench for any desktop GPU benchmarks: it's designed to test OpenGL performance on smartphone GPUs. So the graphics it renders are... well, let's just say they're far below previous-gen consoles in terms of quality.

The only reason someone would run this benchmark against a desktop graphics card is to get artificial performance numbers to generate hype, which is something that's never happened with AMD cards before.

tl;dr Don't even take these numbers with a pinch of salt - ignore them completely. They're BS.

 

[arbiter]

The only reason someone would run this benchmark against a desktop graphics card is to get artificial performance numbers to generate hype, which is something that's never happened with AMD cards before.

Um as you say that i will point back to Fury X launch when AMD did what could call artificial performance numbers when they turned off certain game graphic options to make fury X look 30% faster then a 980ti so saying it never happened with amd cards is a bit of a lie. When independent reviews got the cards that 30% disappeared very quick but even reviewers podcasts picked apart what AMD did with those tests and knew it wasn't gonna be pretty when they did tests in more normal settings users would use.

 

[john_]

Looking at the onscreen numbers, they look like being cut at 60fps. Looking at offscreen numbers the new chip looks more like 2/3 of 390X performance.

 

[arbiter]

Looking at the onscreen numbers, they look like being cut at 60fps. Looking at offscreen numbers the new chip looks more like 2/3 of 390X performance.

had a look at the benchmark, and that is pretty much what happened was vsync limit. which is likely same for what they claim it runs same as 980ti. But if you take offscreen the performance gap between this r9 480 and a gtx980ti is larger then the grand canyon.
R9 480 as its claimed to be vs gtx980ti
Not much better vs gtx 980
That is looking at Offscreen numbers but as you someone said these benchmarks are worthless since graphic's look like ~2000-2004 graphic's.

 

[dj-electric]

Title has a typo, and so does every other article about this.
if you ignore Vsync - it is not close, at all.
if you want a fair comperison in GFX, take the R9 380.

 

[T1beriu]

1. How do you know full Polaris 10 is going to be named 490? Source?

2. How do you know cut down Polaris 10 (67DF) is going to be named 480? Source?

3. "R9 480 (non-X) sibling performs close to the Radeon R9 390X." Have you looked at the benchmark numbers?! 67DF has 60% performance of the 390X. Real power is shown at off-screen scores, not locked to 60 fps.

4. "GPUs at a $350-ish price-point". Do you have a source for this? If not, please state it's just your own speculation.

5. Please stick to proper journalism (based on facts and source checking and speculation disclaimer when it takes place) and don't turn into a 3rd class technology site where speculation is presented as facts.

 

[Nihilus]

Looking at the onscreen numbers, they look like being cut at 60fps. Looking at offscreen numbers the new chip looks more like 2/3 of 390X performance.

Exactly. Really stupid article and poor journalism altogether.
"WCCF is skyrocketing its user base. Let's do what they do!!!!"

 

[ZoneDymo]

1. How do you know full Polaris 10 is going to be named 490? Source?

2. How do you know cut down Polaris 10 (67DF) is going to be named 480? Source?

3. "R9 480 (non-X) sibling performs close to the Radeon R9 390X." Have you looked at the benchmark numbers?! 67DF has 60% performance of the 390X. Real power is shown at off-screen scores, not locked to 60 fps.

4. "GPUs at a $350-ish price-point". Do you have a source for this? If not, please state it's just your own speculation.

5. Please stick to proper journalism (based on facts and source checking and speculation disclaimer when it takes place) and don't turn into a 3rd class technology site where speculation is presented as facts.

Welcome to the site

 

[Tsukiyomi91]

it's just a rumor mill (obviously)... so best not to add the hype when AMD's claims are somewhat dodgy as it only brings disappointment, even for me.

 

[TheinsanegamerN]

Exciting, but as always, I'll wait for the benchmarks to come out. I'm in no rush. My dual 770s are doing just fine right now.

 

[Tsukiyomi91]

a proper bench is what we need coz we all know AMD claims & unofficial "teaser" benches are quite dodgy... best we wait it out patiently.

 

[medi01]

Selling my R9 380 OC I am...

vsync limit

What are the benchmarks you linked?

 

[Frick]

Um as you say that i will point back to Fury X launch when AMD did what could call artificial performance numbers when they turned off certain game graphic options to make fury X look 30% faster then a 980ti so saying it never happened with amd cards is a bit of a lie. When independent reviews got the cards that 30% disappeared very quick but even reviewers podcasts picked apart what AMD did with those tests and knew it wasn't gonna be pretty when they did tests in more normal settings users would use.

Twas sarcasm.

 

[Slizzo]

The R9 480 succeeds the currently-$200 R9 380, and its prospect of offering performance rivaling the $400 R9 390X at half its power-draw appears to meet AMD's "generational leap" claims for the "Polaris" architecture.

And there's the 2x performance per watt figure they were talking about (I believe with this release?)

 

[GhostRyder]

If they are comparing the R9 480 to the R9 390X on an OpenGL bench then its fine because they are comparing their own cards to themselves on a bench that favors their own performance. We still need actual performance numbers in the hands of reviewers, however this still gives us a good taste of what to come (Least a bare minimum). The GFX run on the link shows it comparing the R9 480 - R9 390X so its ok, if it was say the R9 480 to the GTX 980 then we could complain because that is a biased benchmark.

Unless of course the new cards run OpenGL way better and that scenario listed is the only scenario showing its might. If that is the case, then we really need to wait it out and see for ourselves.

 

[HD64G]

As test shows at off-screen scores, this GPU that they try to compare to 390X is at 70-75% of it most of the time. So, we are talking about a GPU at the level of 280X performance or a bit above. Nothing spectacular imho for being new gen. I hope it isn't the Polaris 10, even the cut down one. I hope it is the full Polaris 11 though...

 

[MrGenius]

Does anybody know anything about this GFXBench GL? I'm trying to find out how stupid I am for downloading it and installing it. I thoroughly read through the EULA, checked it with VirusTotal, and scanned it with Windows Defender first...of course. Both scans said it was all good, and nothing looks fishy in the EULA. So after trying to run it it starts up fine. But it immediately asks if I want to download 265MB of some nameless additional files to synchronize the application data with the server. It doesn't want to work if I say no. It's sketching me out.

I'd really like to see how my 280X scores for comparison. But I've been screwed before. So unless someone can tell me it's seemingly harmless based on personal user experience...I'm not going any further with it.

EDIT: DERP! Already been done. You guys do know you can compare it to Nvidia cards right? Not that I'd give a shit. Because I don't buy their overpriced crap anyways.

And yes, 280X actually scores higher in some tests. LMFAO!!!

 

[arbiter]

You guys do know you can compare it to Nvidia cards right? Not that I'd give a shit. Because I don't buy their overpriced crap anyways.

Their over priced crap that in last 2 card releases forced AMD prices down, aka 970 release when 290 was selling for 450-500$ and forced them down to 350range. Also 980ti when fury was projected to be 800-900$ card and 980ti forced it be sell for 650$

Does anybody know anything about this GFXBench GL? I'm trying to find out how stupid I am for downloading it and installing it.

I downloaded it and ran it could test's. The graphics it has is very low i mean my 10 year old laptop probably could run it at 60fps. Its very bad mark to compare gpu's with given how old the graphic's in that demo are and probably old Opengl used.