Radeon RX 480 "Polaris" Launched at $199

CAT-THE-FIFTH · 15 Jun 2016 at 13:57

Starbukc said:
Simply adding on the price of a 730 might not be accurate. The prices of the 960 are (for 2/4GB) +192/229 for Asus, 195/215 for MSI. No way are they prices on top of a 730.

Never considered that - thanks for the spot!!

CAT-THE-FIFTH · 15 Jun 2016 at 13:59

LOL:

https://www.youtube.com/watch?v=YNgTR7SIPXA&feature=youtu.be&t=1m20s

Alienware guy talks of a 490?? :confused:

fs123 · 15 Jun 2016 at 14:00

CAT-THE-FIFTH said:
Why is it misleading - the R9 270X was one of the most efficient cards AMD made and the RX470 cards are replacing that series??

The R9 285, R9 380,R9 280 and R9 280X have worse performance per watt. This is why the Pitcairn GPU in the HD7870/R9 270 and HD7850/R7 265/R7 370 series has lasted so long in mobile too. Only the Fury range is better and is £350+ and not sub £200 like all of these cards.

This has been known for years.

Yes I think AMD is talking about performance per watt improvements going from 28nm to 14nm so it makes sense that they would reference the best performance per watt card on 28nm.

Not sure how it's all calculated but I don't think they claimed the same 2.8X improvements across all their new cards.

bru · 15 Jun 2016 at 14:07

CAT-THE-FIFTH said:
Why is it misleading - the R9 270X was one of the most efficient cards AMD made and the RX470 cards are replacing that series??

Ok if NVidia advertised the new 1060 as having over twice the performance of NVidia technologies and it turned out they meant versus the 660, there would be an absolute uproar from the red team.

The sooner people realise that both companies are happy to use slides or make comments that get message across, regardless of how misleading they might turn out to be, the better.

Edit: I have removed my baiting comment as I don't want to start an argument.

CAT-THE-FIFTH · 15 Jun 2016 at 14:11

bru said:
Ok if NVidia advertised the new 1060 as having over twice the performance of NVidia technologies and it turned out they meant versus the 660, there would be an absolute uproar from the red team.

But NVidia bad, AMD good. the sooner people realise that both companies are happy to use slides or make comments that get message across, regardless of how misleading they might turn out to be, the better.

Sorry but why are so desperately trying to warp this into some us vs them war in an AMD thread??

I had a GTX660 and GTX960 - try harder next time and stop twisting things out of desperation without thinking.

The R9 270X and R9 270 are more efficient than the sub £200 R9 285,R9 380,R9 380,R9 380X.

So using your warped way of thinking AMD should have compared this to the R9 380 and R9 380X with worse performance per watt or the rebadged R9 370 which is the only 70 series part AMD sells in the 300 series part.

So basically you are saying AMD should have compared it to worse performance/watt parts.

Ah,OK so you could show your false concern and then complain AMD was using the least efficient parts on purpose to overstate the improvements.

Edit!!

Next time think about what you post:

http://i.imgur.com/Sbm67wu.gif

AMD could easily have used the R9 285,R9 380,HD7790,R7 260X and any other card apart from the R9 270X.

The GCN cards since Pitcairn and Cape Verde had a decline in performance/watt in desktop SKUs until Fury was launched and that is nowhere near the price of the sub £200 SKUs like the RX470 and the performance either.

bru · 15 Jun 2016 at 14:17

CAT-THE-FIFTH said:
Sorry but why are so desperately trying to warp this into some us vs them war in an AMD thread??

I had a GTX660 and GTX960 - try harder next time and stop twisting things out of desperation without thinking.

The R9 270X and R9 270 are more efficient than the sub £200 R9 285,R9 380,R9 380,R9 380X.

So using your warped way of thinking AMD should have compared this to the R9 380 and R9 380X with worse performance per watt or the rebadge R9 370 which is the only 70 series part AMD sells in the 300 series part.

So basically you are saying AMD should have compared it to worse performance/watt parts.

Ah,OK so you could show your false concern and then complain AMD was using the least efficient parts on purpose to overstate the improvements.

Not at all, all I'm saying is that in response to the comments about Twice as fast as a TitanX being deliberately misleading, Both companies will do that sort of thing if it gets their message across. These things are normally true in specific circumstances, but can seem to be very misleading at first look.

fs123 · 15 Jun 2016 at 14:22

CAT-THE-FIFTH said:
Sorry but why are so desperately trying to warp this into some us vs them war in an AMD thread??

I had a GTX660 and GTX960 - try harder next time and stop twisting things out of desperation without thinking.

The R9 270X and R9 270 are more efficient than the sub £200 R9 285,R9 380,R9 380,R9 380X.

So using your warped way of thinking AMD should have compared this to the R9 380 and R9 380X with worse performance per watt or the rebadged R9 370 which is the only 70 series part AMD sells in the 300 series part.

So basically you are saying AMD should have compared it to worse performance/watt parts.

Ah,OK so you could show your false concern and then complain AMD was using the least efficient parts on purpose to overstate the improvements.

Edit!!

Next time think about what you post:

http://i.imgur.com/Sbm67wu.gif

AMD could easily have used the R9 285,R9 380,HD7790,R7 260X and any other card apart from the R9 270X.

The GCN cards since Pitcairn and Cape Verde had a decline in performance/watt in desktop SKUs until Fury was launched and that is nowhere near the price of the sub £200 SKUs like the RX470 and the performance either.

That's a spot on analysis. AMD could have used the nano as the reference but that is not in the same market segment at all and also uses HBM which would skew the results.

meldarthx · 15 Jun 2016 at 14:23

bru said:
Not at all, all I'm saying is that in response to the comments about Twice as fast as a TitanX being deliberately misleading, Both companies will do that sort of thing if it gets their message across. These things are normally true in specific circumstances, but can seem to be very misleading at first look.

No its not - the fact it they've used their most efficient chip in comparing - it was part of the 300 series also. Now if they'd used tonga used a higher number and someone came out with the figured from Pitcairn - found they were lower you'd have people screaming high heaven. Now you have some trying to twist they are being misleading........its not

CAT-THE-FIFTH · 15 Jun 2016 at 14:25

bru said:
Not at all, all I'm saying is that in response to the comments about Twice as fast as a TitanX being deliberately misleading, Both companies will do that sort of thing if it gets their message across. These things are normally true in specific circumstances, but can seem to be very misleading at first look.

Right you stated this:

Just as some companies say 2.8 times the performance/watt with AMD technologies, which turns out to be versus a 'three' generation old product.(GCN 1.0 Pitcairn GPU)

In the end I won't take the actual figures until I see reviews as all of it is best case scenarios,but at the same time AMD using its best performance/watt £100 to £200 part to compare a similar price-range SKU is not really very good PR IMHO.

If they wanted to really warp that metric,they could easily have used an R7 260X,R9 285 or R9 380 which all went backwards in some ways. So in that case, I can't fault AMD on using that card, for performance/watt comparisons.

We all know AMD is still using elderly GCN1.0 parts in cards like the R9 370.

It would be different if they tried to compare Vega to a R9 270X which would be warping to make it look much better peformance/watt over Fiji and then I would agree with you.

humbug · 15 Jun 2016 at 14:28

So now we have wccf claiming to have RX 480 speeds and specs.

All the specs match the Chinese leaks and the performance is about 99% dead on.
1080Mhz Base clock, 1266Mhz Boost, as fast as a Fury-Nano

Guru3D claim to have the RX 470, its as fast as a 290 which would fall into line for its spec vs the 480.

bru · 15 Jun 2016 at 14:29

Ok so lets say it is not misleading because it is accurate, so in the same breath, the twice as fast as a Titanx is also accurate (in certain situations), so can we stop telling people how much NVidia fooled everyone when they marketed it as such.

fs123 said:
Corporations may not be your friends but some have a better attitude to their customers than others. Some take customers as fools when they market some of their products to be twice as fast as their previous top end card

fs123 · 15 Jun 2016 at 14:35

bru said:
Ok so lets say it is not misleading because it is accurate, so in the same breath, the twice as fast as a Titanx is also accurate (in certain situations), so can we stop telling people how much NVidia fooled everyone when they marketed it as such.

If you say so dude. Let me know where it's twice as fast. Any game benchmark will do.

TheRealDeal · 15 Jun 2016 at 14:36

bru said:
Ok so lets say it is not misleading because it is accurate, so in the same breath, the twice as fast as a Titanx is also accurate (in certain situations), so can we stop telling people how much NVidia fooled everyone when they marketed it as such.

Nobody said 2 x Titan X was not accurate in VR. The VR section was finished and a promo video was on then finished, he then kept saying twice the perf of Titan X which is what everyone was talking about. Amd's slide even said up to 2.8 and not a fixed 2.8. Let's be honest these are 2 different types of things going on.

Any how i think Cat summed this up well and no more needs said.

humbug · 15 Jun 2016 at 14:38

Why are we arguing about bloody power consumption again?
Its like an obsession.

Lets stick to what matters, price, performance.

r7slayer · 15 Jun 2016 at 14:38

TheRealDeal said:
Nobody said 2 x Titan X was not accurate in VR. The VR section was passed and he kept saying twice the perf of Titan X which is what everyone was talking about. Amd's slide even said up to 2.8 and not a fixed 2.8. Let's be honest these are 2 different types of things going on.

Nailed it!

CAT-THE-FIFTH · 15 Jun 2016 at 14:39

Dont get me wrong if the RRP of the RX470 is closer to the R9 380 RRP than the R9 270X or R7 370 current RRP,the performance comparisons should be made with the R9 380 too,but it is not like AMD or Nvidia(who compared the GTX960 to the GTX660 BTW in their slide decks) won't try that since they will only do 70 series to 70 series comparisons,but at the same time it also means those performance/watt improvements look much smaller.

meldarthx · 15 Jun 2016 at 14:41

humbug said:
Why are we arguing about bloody power consumption again?
Its like an obsession.

Lets stick to what matters, price, performance.

because usual suspects are up to derailing the subject as usual....

you missed where I pointed out if 470 is 290+ speeds......then 480 being at 980+ speeds has to be right and only leak that seems out of place is Videocardz

Mauller · 15 Jun 2016 at 14:42

And people kept saying that P11 would only be 950 performance.

And they kept on about it for so long.

CAT-THE-FIFTH · 15 Jun 2016 at 14:43

We can only say the RX470 matches a R9 290 in 3DMark,so we are hoping it does the same in games at a reasonable price.

We will need to wait and see OFC.

humbug · 15 Jun 2016 at 14:44

Its really simple.

The 470 has a board power of 110 Watts, so its using anything "up To" 110 Watts, its twice as fast as the 270X which has a board power of 180 Watts and uses on average 135 watts.

So the 470 has twice the performance for <80% of its predecessors power.