Radeon RX 480 Competition Poll

EVERYBODY SHHHHH

y51ucwh9da5x.png

Performance in those looks to be about as expected, right up there with the 980 and 390x. Decent numbers at 1440p too. Be interesting if this little $250 card could be a viable entry level 1440p card.
 
Performance in those looks to be about as expected, right up there with the 980 and 390x. Decent numbers at 1440p too. Be interesting if this little $250 card could be a viable entry level 1440p card.


290, 390, 970, 980 and 390x were good mainstream 1440p cards. Not even entry level unless you are running 144hz.
 
The $10 AMD upcharge is on the AIB partners, not AMD. Let me know when you find a 1070 or 1080 at those MSRPs. I suggest that AMD will hit its MSRP before nVidia, with neither dropping below +$10 for a while.

Main point... AMD isnt raising the price over their $199 for the same card. The 4gb is still coming at $199 MSRP, but they have a different spec launching first... Different launch dates on the cards shouldnt exactly lead to salt imo. Its not like it is the same-specs card getting released, just at a different price.
FYI, the cheapest AIB 1080s have MSRP sitting at $10 above nV's MSRP as well. It's up to stock and availability. Though that's irrelevant. I am actually confused by the whole fuss about new Polaris arch features, because they are invisible here.
Clocks are @ 1120 mhz for some reason
It's base clock, for some reason they did not list turbo for the card here.
Rumor has it that those charts are charts with the 480 photo-shopped in place of the 390...

Division Example with 390: TEST: Sapphire NITRO Radeon R9 FURY 4G - Jedna z nejlepších grafických karet 28nm generace! - DDWorld.cz

and before ledra goes off on me, it isnt my rumor :p
Case closed, actually.
 
Clocks are @ 1120 mhz for some reason

yeah I'm thinking its fake, cause of that or something is not right with the bios. Which they guy stated it was a retail board and he bought it, so that shouldn't be the case.
 
NewEgg had the cards listed at 239 for most, 249 for the XFX and Saphire ones with nifty looking backplates... Looks like a $10 upcharge on the reference, from AIBs.

Thats a Newegg price gouge, not AIB markup. Only one with the markup would be XFX for custom bios and backplate. Everyone that keeps hitting that autonotify button may raise the price. Newegg sucks ass.
 
I'd probably put it somewhere north of a 390...closer to a 980 than a 390, but not passing a 980. Even if it does turn out to be 390-level, it's still a hell of a card. I love my 390. Where a 1070 pushes 50 fps in a game, my 390 pushes 35 fps, same settings. Enough so that my freesync still keeps things smooth and it looks great. A $230 that can do that is something worth considering.

I have to admit, freesync changed a lot of things as far as enjoyable settings. As long as my 390 can keep the min over 30fps, it's a lot more comfortable now. (and why I'm not buying another card until the big guns arrive next year)
 
I have to admit, freesync changed a lot of things as far as enjoyable settings. As long as my 390 can keep the min over 30fps, it's a lot more comfortable now. (and why I'm not buying another card until the big guns arrive next year)

Looking forward to a similar experience here, though my monitor's FS minimum is 40. Still. I target 60, so that should handle all but the most extreme drop.

Just not sure yet when I'll grab an AMD card. 480 is a side grade to my 970. Probably wait until Vega, unless I can get enough from my 970 to cover most of a 480 purchase.
 
Yeah, if their sole product for above 480 performance is multi-gpu solutions then they are done, because whether multi-gpu is on a single board or on separate video cards it still requires crossfire, which sucks teh serious donkey balls.

Multi-gpu is only ever to be used if the fastest single GPU solution on the market (from any vendor) isn't fast enough, because the compromises are just too great.
As someone who has used 3-4 GPU's for years and am finally down to 2, I frequently find myself turning the 2nd off for compatibility. I'd have to agree that multi gpu just isn't supported well enough and fast enough in driver support to be a good experience next to a single gpu.
 
RX 480 is for surely a hot and debatable topic, no doubt it is sparking controversy across the web.

Knowing what is out there, with the price and specs announced already in regards to RX 480, In your opinion list the GPUs you think compare to RX 480 based on the information available. Comparison by Performance? Comparison by Price? What do YOU think is important for RX 480 from a gamers perspective?

I'm interested in your opinions.
380 will clear house at $180 after rebates
390 will clear house at $220 after rebates
390x will clear house with $260 after rebates
Nano will be $240 will be clearing house because of Vega and 980ti pricing by October
Fury will be $280 will be clearing house because of Vega and 980ti pricing by October
Fury X will be $300 and will be clearing house because of Vega and 980ti pricing by October
970 will clear house about $200 after rebates
980 will be $240->$250 and continue to sell with stock coolers as stop gap till 1060
980ti will be $260->$280 and will clear house because of 1070/1080

If you are going to compare, compare on it's price competition. Everyone buys the fastest based on what their wallet can support. Since prices aren't stabilized you'll also have to look at it's closest competitor speed wise at 1080p resolution. And say "If this product X drops at or below RX480 $240's, then get X"

As a side note: This whole AMD pricing strategy is going to lead to a very LARGE inventory write down by both sides.
 
Last edited by a moderator:
Anytime I've seen a site reinvent the wheel by changing their review process for a product it's because they want to highlight something, good or bad, about said product. If you want to be fair then test it the same way you always have. And this card is targeted at 1080p gamers. Not testing there makes no sense. Most 1440p owners already have cards as fast or faster then the 480, I would imagine. Unless the 480 turns out to be faster than advertised?
 
Anytime I've seen a site reinvent the wheel by changing their review process for a product it's because they want to highlight something, good or bad, about said product. If you want to be fair then test it the same way you always have. And this card is targeted at 1080p gamers. Not testing there makes no sense. Most 1440p owners already have cards as fast or faster then the 480, I would imagine. Unless the 480 turns out to be faster than advertised?

I've accused Kyle of being biased. But his reasoning for these changes makes sense even to me. I don't think he's trying to highlight anything good or bad about the card. I think he's trying to strengthen the focus on the card being tested. Rather than doing best playable settings on ALL cards, he's suggesting (if I understand this correctly, as I can be dense at times) that they will test the reviewed card at max playable settings, and then test the other cards in the comparison at the SAME settings so that you get an apples to apples comparison. IE, is the other card more playable at those settings? Less playable?
 
I've accused Kyle of being biased. But his reasoning for these changes makes sense even to me. I don't think he's trying to highlight anything good or bad about the card. I think he's trying to strengthen the focus on the card being tested. Rather than doing best playable settings on ALL cards, he's suggesting (if I understand this correctly, as I can be dense at times) that they will test the reviewed card at max playable settings, and then test the other cards in the comparison at the SAME settings so that you get an apples to apples comparison. IE, is the other card more playable at those settings? Less playable?

For initial review that would be great for getting a feel for a card compared to other available options. It still doesn't let you know though if even the faster card can gain extra settings and still be playable. This is also unique in getting viewers inputs for testing/reviews - rather awesome HardOCP!
 
Anytime I've seen a site reinvent the wheel by changing their review process for a product it's because they want to highlight something, good or bad, about said product. If you want to be fair then test it the same way you always have. And this card is targeted at 1080p gamers. Not testing there makes no sense. Most 1440p owners already have cards as fast or faster then the 480, I would imagine. Unless the 480 turns out to be faster than advertised?

This review is going to be interesting, no doubt. The rub is there's no new competition for this card at this price point. Compare a 960/380? At the RX 480 specs (and discounting hype) this card will obliterate them. Pretty much anything under a 980/390 is going to fall here or, at best, be a tie. And there's nothing wrong with that.
 
Anytime I've seen a site reinvent the wheel by changing their review process for a product it's because they want to highlight something, good or bad, about said product. If you want to be fair then test it the same way you always have. And this card is targeted at 1080p gamers. Not testing there makes no sense. Most 1440p owners already have cards as fast or faster then the 480, I would imagine. Unless the 480 turns out to be faster than advertised?

According to AMD the RX 480 is actually targeted at "Beyond HD Gaming and Premium VR."

RX 470 is targeted at 1080p.

http://cdn.wccftech.com/wp-content/...is-11-Radeon-RX-480-RX-470-RX-460-GPUs_6.jpeg
 
There's no new competition for the 1070/1080 either, I'm sure they'll manage.
 
Mentioned it in the other thread, but probably more relevant here. RX480 is considered a side grade to the R9 390, so it probably won't even perform like a 390x.

Sapphire: Upgrade von R9 390 auf RX 480 lohne sich nicht

May be you should read through the threads and see that about 99% of the people have already said it unless you are a miner and you want to waste less power.

I see you are busy beating up that point lol.

It is not designed for 390/390x owners unless ofcourse they want to reduce their power usage by half. Thats all. .It's geared towards bringing that performance to people to 200 dollar price point
 
It is not designed for 390/390x owners unless ofcourse they want to reduce their power usage by half. Thats all. .It's geared towards bringing that performance to people to 200 dollar price point

This is the point that a lot of people are struggling to grasp. GTX 770 owners didn't look at the GTX 960 as a potential upgrade either. Different markets.

The RX 480 is MOSTLY for GTX 660/760/960 and Radeon HD 7800/R9 270/380 owners looking to upgrade.

I'm only making the switch because I got good value in my 970 sale and I want to use FreeSync. Most 970 owners, however, should look at the 1070 or wait for the 1170.
 
Generally speaking though, performance levels trickle down to a lower price point with a process change. So we should have gp104 performing like gm200, which it doesn't, it's faster.

On the other hand, Polaris 10 is slower than Hawaii. Just saying. It's disappointing, but the price is good, also the price is determined entirely by the disappointing performance. If it was faster, as I think it should have been, it would have been priced higher. While this has no real bearing on the consumer, who finds themselves paying less for a less performant product, it's bad for AMD, who find themselves charging less for the exact same die they intended to charge more for
 
This is the point that a lot of people are struggling to grasp. GTX 770 owners didn't look at the GTX 960 as a potential upgrade either. Different markets.

The RX 480 is MOSTLY for GTX 660/760/960 and Radeon HD 7800/R9 270/380 owners looking to upgrade.

I'm only making the switch because I got good value in my 970 sale and I want to use FreeSync. Most 970 owners, however, should look at the 1070 or wait for the 1170.

Freesync is a beautiful thing.
 
This is the point that a lot of people are struggling to grasp. GTX 770 owners didn't look at the GTX 960 as a potential upgrade either. Different markets.

The RX 480 is MOSTLY for GTX 660/760/960 and Radeon HD 7800/R9 270/380 owners looking to upgrade.

I'm only making the switch because I got good value in my 970 sale and I want to use FreeSync. Most 970 owners, however, should look at the 1070 or wait for the 1170.

I bought an amplifier for my car yesterday and contemplating the merits of ordering a component set today if I can decide on which I want. That's how excited I am to upgrade my video card to a 480 from a 290.
 
What time will the review be put up? All I've been able to find out is embargo lifts the 29th, figured with less than 24 hours somewhere might've been a little more specific now..
 
I know [H] isn't usually one to do things this way and it'd be a time sink, but what about a comparison of similar launch price cards going a couple generations back? I'm the type of buyer who buys a video card for 2+ years of usage and would be really intrigued about what gains I would actually see from purchasing this coming from a 7870 that was in the same price bracket.

I'd like to see this too, or something similar to Anandtech's GPU bench. You can basically run an ad hoc comparison of various graphics cards, granted they benchmarked them. Of course, since [H] doesn't really do canned benchmarks, it would really be down to the apples-to-apples comparison data.
 
I don't think it would be worth the effort, to be frank. If you want to compare to an older card, say 780 ti, you can always go back to prior reviews that pit the card against one in the 480 review. They're alreay including a 380x, 390 and 970 - that should take you back pretty far.

I'd much rather see time spent on a couple more games and/or frame variance analysis.

Also, most of those 15-card comparisons are seriously flawed in they are using old data for the older cards i.e. the cards were tested on older drivers and software versions.
 
I know [H] isn't usually one to do things this way and it'd be a time sink, but what about a comparison of similar launch price cards going a couple generations back? I'm the type of buyer who buys a video card for 2+ years of usage and would be really intrigued about what gains I would actually see from purchasing this coming from a 7870 that was in the same price bracket.
We do not have the resources to do this. We would have to retest with current game and driver versions to make the results true. We do not regurgitate benchmarks from years ago as I just do not think that is fair to to AMD, NVIDIA, or the reader.

That said, you likely know where your 7870 stands in relation to say a 960, and that will be covered with current games and drivers.
 
This is also why you don't limit yourself to reviews from one place.

If I want to see how a card will run some of today's popular games in terms of settings at a playable framerate, I come to [H].
If I want to see where a card slots among its peers and the prior generation, I read any recent review (Performance Summary) from TPU.
And if I want to see how a card compares to the last few generations, I go to Tom's GPU charts. Example below:

GeForce GTX 560, 660, 760,and 960

So it really bothers me when people ask a reviewer to change their method. We already have virtually every method covered. I'd rather have one of each than two of one, and none of the other. Read multiple sources.
 
Back
Top