• 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.

Rumour: Radeon R9 480/480X 3DMark 11 Scores

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
BTW those crossfire scores are really horrible, so I'm not really sure what to take from this.

the 390x @CF (with a 4770k) get a graphics score of 33,000
http://www.3dmark.com/3dm11/10099998

To people who wanted a 390 and we told them to wait, did they really just wait for 20% performance and much lower power consumption?

What did you expect?
It also has better video codec support, which is also important for some.
 
Last edited:
Underwhelming IPC improvement if this is accurate. Well, compare it to Fury and Fury X and it's a bit more favorable, but we know those cards are fundamentally flawed.

If this is accurate, then power consumption is mostly the winner here with a minor performance-per-dollar improvement. You can already get a 390 for $300, so if 2304SP P10 is $300 then it is what, about 20% more performance for the same price (assuming 390 matches 970 in this benchmark)?

It's nice that it's nearly Fury Air speed, but that card never had the best performance per dollar compared to Hawaii.

To people who wanted a 390 and we told them to wait, did they really just wait for 20% performance and much lower power consumption?
if similar price, still not bad.
 
Underwhelming IPC improvement if this is accurate. Well, compare it to Fury and Fury X and it's a bit more favorable, but we know those cards are fundamentally flawed.

If this is accurate, then power consumption is mostly the winner here with a minor performance-per-dollar improvement. You can already get a 390 for $300, so if 2304SP P10 is $300 then it is what, about 20% more performance for the same price (assuming 390 matches 970 in this benchmark)?

It's nice that it's nearly Fury Air speed, but that card never had the best performance per dollar compared to Hawaii.

To people who wanted a 390 and we told them to wait, did they really just wait for 20% performance and much lower power consumption?

At least now that appears to be a worst-case scenario. It would stand to reason that these numbers aren't overly optimistic or showing full potential at this point (assuming $299 retail).
 
Last edited:
Well, GTX 1080 is 30% over 980 Ti for about the same price. See all the flak it is getting for not passing the savings onto the consumer? At least that's 30%. If this leak and the $300 rumour is accurate, then AMD's performance-per-dollar improvement on the node shrink is even smaller, despite the shrink being more significant.

I guess I hope the price is less. At $250 now the 20% is fine since you are now talking about significant price savings over 390/970 vs basically the same.
 
Well, GTX 1080 is 30% over 980 Ti for about the same price. See all the flak it is getting for not passing the savings onto the consumer? At least that's 30%. If this leak and the $300 rumour is accurate, then AMD's performance-per-dollar improvement on the node shrink is even smaller, despite the shrink being more significant.

I guess I hope the price is less. At $250 now the 20% is fine since you are now talking about significant price savings over 390/970 vs basically the same.

Yes but $300 is $300 and $699 is $699. Expectations rise exponentially as the asking price rises.
 
Underwhelming IPC improvement if this is accurate. Well, compare it to Fury and Fury X and it's a bit more favorable, but we know those cards are fundamentally flawed.

If this is accurate, then power consumption is mostly the winner here with a minor performance-per-dollar improvement. You can already get a 390 for $300, so if 2304SP P10 is $300 then it is what, about 20% more performance for the same price (assuming 390 matches 970 in this benchmark)?

It's nice that it's nearly Fury Air speed, but that card never had the best performance per dollar compared to Hawaii.

To people who wanted a 390 and we told them to wait, did they really just wait for 20% performance and much lower power consumption?
Outside of pure guessing here, do we know the price?

20% performance.
Much lower power.
Price???

Perf/$ depends 1st and 3rd values.
 
Last edited:
Maybe it's just me, but a bit slower than a 390x (non-oc'd) for $299 if these #'s are right is incredibly underwhelming for the 480. Once the 1070's hit any 390x new stock left in retail will likely be selling for what...$250? Used 390x's on ebay are already creeping towards $250. I fully understand the lower heat/power as a selling point, but that seems to be really grasping at trying to compete. And this is as a current AMD owner, not nv fanboying.
 
Maybe it's just me, but a bit slower than a 390x (non-oc'd) for $299 if these #'s are right is incredibly underwhelming for the 480. Once the 1070's hit any 390x new stock left in retail will likely be selling for what...$250? Used 390x's on ebay are already creeping towards $250. I fully understand the lower heat/power as a selling point, but that seems to be really grasping at trying to compete. And this is as a current AMD owner, not nv fanboying.

early scores performs like a Fury in a bring salt benchmark atm.
how is that for a mid range card, bad?
 
I'll wait for official launch with real specs and gaming benchmarks. Silly synthetic benchmarks won't really give an accurate picture of true performance....Or lack of.
 
Still not sure these clocks are close to final. At some point you have to assume AMD isnt retarded and that Polaris 10 would actually fully utilize GDDR5 and need GDDR5X on some SKUs. Polaris 10 is rumored to include both memory controllers. If full 2560 card exists it could be GDDR5X which could be why it has been distributed yet.

I seriously doubt AMD doesnt have a $400-500 1070 market competitor until VEGA. It wouldnt even be cost effective initially to push HBM2 down into that market, so it has to come from Polaris 10. Someone calculating backwards from known bandwidth restrictions could probably better predict final MHZ for these cards assuming AMD didnt botch something along the way.
 
Still not sure these clocks are close to final. At some point you have to assume AMD isnt retarded and that Polaris 10 would actually fully utilize GDDR5 and need GDDR5X on some SKUs. Polaris 10 is rumored to include both memory controllers. If full 2560 card exists it could be GDDR5X which could be why it has been distributed yet.

I seriously doubt AMD doesnt have a $400-500 1070 market competitor until VEGA. It wouldnt even be cost effective initially to push HBM2 down into that market, so it has to come from Polaris 10. Someone calculating backwards from known bandwidth restrictions could probably better predict final MHZ for these cards assuming AMD didnt botch something along the way.


GDDR5X only on nvidia
 
why does everybody assume that the 18060 score is at stock?...did they underclock it for the 15524 score? that doesnt make sense.

15524(stock) is pretty bad score, on par with a r9 390 as rumors suggested from sometime...at $299? thats to high...
 
There is a source over the internet and it says that "Polaris has both GDDR5 and GDDR5X compatible IMC".
 
How did you come to that conclusion?



Do we know if Polaris got a GDDR5X compatible controller?

*shrug*

I haven't been following as well as I'd like so I was just curious if this was the case. Does it really require a new memory controller?
 
Maybe it's just me, but a bit slower than a 390x (non-oc'd) for $299 if these #'s are right is incredibly underwhelming for the 480. Once the 1070's hit any 390x new stock left in retail will likely be selling for what...$250? Used 390x's on ebay are already creeping towards $250. I fully understand the lower heat/power as a selling point, but that seems to be really grasping at trying to compete. And this is as a current AMD owner, not nv fanboying.

One would hope that the 480 wouldn't be priced at $299, but more along the lines of the 380 at ~$199.
 
Pretty sure most expected a regression of IPC in Pascall and an improvement in IPC for Polaris. Nothing really new here..
IPC of Pascal CC (GP104) is aprox. 10% lower than IPC of Maxwell CC clock to clock.
 
Last edited:
Yes but $300 is $300 and $699 is $699. Expectations rise exponentially as the asking price rises.
A 300$ GPU like this is bad for AMD. This is a big issue for them. I hope that it is cheaper because the brand struggles overall.

My thinking is simple: the majority of AMD buyers are brand loyalists, just like the majority of nvidia's buyers. They(AMD) are betting too much on efficiency alone. This is not the right time to do this(market share wise): expect the 970 to sell even better.

While nvidia strengthens its brand further, increases the price and drops a bomb at the high end in order to sell over-priced mid to low end hardware, AMD tries to rise from the bottom with better price/$ and efficiency. This will not gain them market share today.

Increasing the price at the high end this day and age is the smarter choice, if you get the irony.
This is the natural way to go: nvidia is smarter, AMD is dumber. AMD should have never let go the high end. This is not the time to do this.
 
wow, if this turns out to be true, my 290 will be king for 2 more years since both amd and nvidia are only releasing big dies next year.

Vega was pushed up to this year according to several rumors. So if it launches is October, thats not far off. Sooner than GP100/102 is currently expected to launch.
 
Back
Top