• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

AMD Polaris Thread: Radeon RX 480, RX 470 & RX 460 launching June 29th

Page 151 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Status
Not open for further replies.
So once again we I see a thread turned in to national energy charge thread. Amazing how less and less is being leaked the closer we get to launch lol. Amd did the right thing not releasing drivers week before launch. Everyone seems to be busy testing lol.
 
Mopetar said:
If you're only using the card for mining (some people do this) than power cost is a factor in the profitability of said card.

If you're just gaming ~20 hours per week you're right in that it really doesn't matter that much or would only amount to a few dollars in difference for the average person.
Click to expand...
So what you're saying is that AMD performs worse on mining BTC/watt metric :hmm:
 
R0H1T said:
So what you're saying is that AMD performs worse on mining BTC/watt metric :hmm:
Click to expand...

Don't know how you read that from what I wrote.

I'm saying that if you're doing 24/7 mining with your card, you'll probably care about the power usage relative to other cards, whether they're from the same manufacturer or not.

Also, no one does bitcoin mining on GPUs any more since people started widely deploying FPGAs and ASICs that are far more efficient making it pointless to use a GPU at all. If you meant Ethereum mining, I don't know of anyone doing dedicated mining who uses Nvidia hardware (unless they're just casually mining with their gaming GPU) since AMD tends to have a massive performance advantage for that particular application.
 
nkdesistyle said:
So once again we I see a thread turned in to national energy charge thread.
Click to expand...

What do you expect when the only improvement vs a 390X is in performance per watt? That is all there is to talk about. Introducing the The AMD RX 480 LED light bulb! Replace your light bulb now and save 2.5X power.
 
sm625 said:
What do you expect when the only improvement vs a 390X is in performance per watt? That is all there is to talk about. Introducing the The AMD RX 480 LED light bulb! Replace your light bulb now and save 2.5X power.
Click to expand...

Some would say that $100+ cheaper for that same performance is also a real and tangible improvement. And those are costs now--not reflected in 390X release price which was like $150+ more, right?
 
sm625 said:
There must be an awful lot of those people because cents per hour saved is the only rational benefit of buying a GTX 960 vs its closest priced AMD competitor. Yet lots of people did just that. Same with the 970. The only benefit it offered vs the competition was cents per hour saved at the cost of lower performance per dollar. The fact that the savings were only $0.0002 per hour didnt seem to matter.
Click to expand...

That's the one really depressing thing I've noticed since I came back into the hardware fold after a couple years: Nvidia has built an enormous lead in the fanboy department. It's really pretty ridiculous.
 
SSBrain said:
A guy on TPU Forums received an RX480 in advance and while it does seem to perform well in Batman Arkham Knight (390X level, at least at 1080p), it doesn't seem to exactly have thrilling VR performance. Check out the posts on the last few pages by user gasolina:

http://www.techpowerup.com/forums/t...leaked-and-tested.223351/page-22#post-3479235
Click to expand...

Weird that it scored higher than AMD's stated score (6.8 vs. 6.3) but it's still on the low side for what you'd want if you were getting a card strictly for VR, especially if you want something slightly more future proof.
 
SSBrain said:
A guy on TPU Forums received an RX480 in advance and while it does seem to perform well in Batman Arkham Knight (390X level, at least at 1080p), it doesn't seem to exactly have thrilling VR performance. Check out the posts on the last few pages by user gasolina:

http://www.techpowerup.com/forums/t...leaked-and-tested.223351/page-22#post-3479235
Click to expand...
390x level is good. He says he can also max stuff out at 1440p which is nice. But for VR this is just entry level. I would expect you need something more powerful for full VR.
 
mopetar said:
Weird that it scored higher than AMD's stated score (6.8 vs. 6.3) but it's still on the low side for what you'd want if you were getting a card strictly for VR, especially if you want something slightly more future proof.
Click to expand...
sze5003 said:
390x level is good. He says he can also max stuff out at 1440p which is nice. But for VR this is just entry level. I would expect you need something more powerful for full VR.
Click to expand...

I've used a non-X R9 390 for a short period on a weaker CPU than user gasolina, and I had better scores on the same benchmark with stock clocks. Overclocking would bring the card to R9 390X level. There are suggestions that the RX480 might indeed only have 32 ROPs, I wonder if this is having an effect? The rendering resolution target with the HTC Vive is 3024x1680 pixels (1.4x higher on both dimensions than the actual overall display resolution).

On a related note, the TPU database has been updated today to list the RX460 as having 16 ROPs, the RX470 as 24 ROPs and the RX480 as 32 ROPs, so perhaps the rumors were true.

https://www.techpowerup.com/gpudb/2849/radeon-rx-460
https://www.techpowerup.com/gpudb/2861/radeon-rx-470
https://www.techpowerup.com/gpudb/2848/radeon-rx-480
 
Last edited:
sm625 said:
What do you expect when the only improvement vs a 390X is in performance per watt? That is all there is to talk about. Introducing the The AMD RX 480 LED light bulb! Replace your light bulb now and save 2.5X power.
Click to expand...

it should be compared to a 380, not a 390x
 
Here is some bench result I saw it. the 480 had better CPU with higher score, others seens like i5 score.
ELCAcST.png
 
Mopetar said:
Don't know how you read that from what I wrote.

I'm saying that if you're doing 24/7 mining with your card, you'll probably care about the power usage relative to other cards, whether they're from the same manufacturer or not.

Also, no one does bitcoin mining on GPUs any more since people started widely deploying FPGAs and ASICs that are far more efficient making it pointless to use a GPU at all. If you meant Ethereum mining, I don't know of anyone doing dedicated mining who uses Nvidia hardware (unless they're just casually mining with their gaming GPU) since AMD tends to have a massive performance advantage for that particular application.
Click to expand...
I know mining BTC isn't profitable, for at least the last couple of years, & was using BTC just for reference. However you did say this ~
If you're just gaming ~20 hours per week you're right in that it really doesn't matter that much or would only amount to a few dollars in difference for the average person.
Click to expand...
in response to~
If people are really trying to decide what GPU to get based on how many cents per month they may save in power, you need to take up another hobby.
Click to expand...
when he was saying that choosing between Nvidia & AMD simply to save some cents on power costs is futile, & it is, however what you inferred doesn't make sense since the original argument was in regards to different GPU vendors & gaming. If you're talking purely about mining then it's a whole different ballgame altogether.

As for mining cryptocoins, the newer gen GCN's are better & more efficient & will continue to be with the 4XX series IMO.
 
TeknoBug said:
Well from what I've read around the internet, the RX 480 performance is similar to in between a 390 and 390X, I guess when it comes out we'll see by actual users.
Click to expand...

This is all based on pre-release drivers, though. The people leaking results did not get the cards through official channels and do not have the final drivers.
 
zinfamous said:
Some would say that $100+ cheaper for that same performance is also a real and tangible improvement. And those are costs now--not reflected in 390X release price which was like $150+ more, right?
Click to expand...

Yeah, how can people not be excited about having a high performance gpu for $200 now? Instead of crap price to performance cards like the GTX 960 and R9 285.
 
SSBrain said:
A guy on TPU Forums received an RX480 in advance and while it does seem to perform well in Batman Arkham Knight (390X level, at least at 1080p), it doesn't seem to exactly have thrilling VR performance. Check out the posts on the last few pages by user gasolina:

http://www.techpowerup.com/forums/t...leaked-and-tested.223351/page-22#post-3479235
Click to expand...


He's not using official drivers. That said, 99% of stuff we are seeing is with wrong drivers from ppl who got their cards unofficially and thus do not have access to the NDA drivers.


http://www.techpowerup.com/forums/t...leaked-and-tested.223351/page-23#post-3479239
 
thesmokingman said:
He's not using official drivers. That said, 99% of stuff we are seeing is with wrong drivers from ppl who got their cards unofficially and thus do not have access to the NDA drivers.

http://www.techpowerup.com/forums/t...leaked-and-tested.223351/page-23#post-3479239
Click to expand...

It looks like gasolina is using the ones that were bundled with the GPU. A different user tried them out and couldn't make them work with his AMD Fury X: http://www.techpowerup.com/forums/t...leaked-and-tested.223351/page-22#post-3479028

It's true that the driver version is not same as the ones the reviewers are using (or supposed to be using, i.e. 16.6.2).

antihelten said:
The 32 ROP rumor was based on GPU-Z.

GPU-Z has the exact same numbers as TPU, since GPU-Z is made by TPU!
Click to expand...
Don't the guys at TPU also have samples for the review(s) they will be releasing on June 29th? My guess is that they have legitimately acquired correct technical information for these cards.
 
qookap said:
Here is some bench result I saw it. the 480 had better CPU with higher score, others seens like i5 score.
ELCAcST.png
Click to expand...

Any reason why the 480 is getting a 14.4% (12.5% for Extreme score) improvement in performance for a ~9% OC?

Edit: The 980 is getting 9.4% (8.4% for Extreme score) improvement on a ~16% OC. This seems counter-intuitive or almost as though the OC results should be flopped.

R0H1T said:
However you did say this ~in response to~when he was saying that choosing between Nvidia & AMD simply to save some cents on power costs is futile, & it is, however what you inferred doesn't make sense since the original argument was in regards to different GPU vendors & gaming. If you're talking purely about mining then it's a whole different ballgame altogether.
Click to expand...

Fair enough. Like so many others I lost the context of the discussion and invented my own. Let this post serve as a monument to my shame.
 
Last edited:
SteveGrabowski said:
Yeah, how can people not be excited about having a high performance gpu for $200 now? Instead of crap price to performance cards like the GTX 960 and R9 285.
Click to expand...

It's common here to base success on expectations rather than current market which is not a bad thing since AMD and Nvidia keeps track of vocal minority. That being said I had a GTX 960 and upgraded to R9 290 and performance was really noticeable so definitely looks like RX 480 will be a worthwhile upgrade for GTX 960/R9 280 owners. In terms of VR I think 480 will play most VR Games that don't have complex 3D engines like Project Cars fine which seems like >90% of VR games/apps.
 
SSBrain said:
Don't the guys at TPU also have samples for the review(s) they will be releasing on June 29th? My guess is that they have legitimately acquired correct technical information for these cards.
Click to expand...

My point is simply that the numbers in the TPU database doesn't in any way confirm the earlier 32 ROP rumor, since it's the same source. That doesn't mean it's wrong either of course.
 
Mopetar said:
It's almost always nice outside year round and they have a pretty stable average temperature compared to most of the US, so the summer doesn't feel that much hotter than any other time of year. Once you get used to it you don't feel like you need AC.
Click to expand...

Spot on about the weather... And the amount of solar radiation they get makes solar a huge incentive.

What really makes no sense is New York, that's ridiculous.
 
Status
Not open for further replies.

TRENDING THREADS

Back
Top