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

Page 171 - 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.
Feb 19, 2009
10,457
10
76
The voice of reason! Too many believe AMD's Polaris 10 is a Hawaii successor. As a result of this flawed assumption, they believe that AMD is either entirely sacrificing performance gains for next gen in favor of perf/watt and/or they think AMD's engineers failed completely and Polaris 10 was "originally" intended to compete with 1070/1080.

Actually.

5.8TFlops is according to that guy's review, matching a 390 with less TFlops.

If true, that's a regression. Not a good result.

It should at the very least match a 390X. Then we can say, they went for perf/w gains without performance regression.

However, I will wait for judgement til I see reviews from more reputable sites. Despite that I already pulled the trigger for 2x RX 480 cos it's the best bang for buck by a mile compared to what's available currently. lol
 

Tweak155

Lifer
Sep 23, 2003
11,448
262
126
Well, this thread sure has been entertaining. Just about 30mins until we see what's really going on, maybe take a break from speculation?
 

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
We still dont know transitor count.But 2048SP and 64Rops will be like 15-20% faster than 2304SP and 32Rops and probably more at 1440P(my guess).

That SKu will be like 250-260mm2 MAX.

There is no way AMD didn't do simulation on this. It's why this card is a $199-229 MSRP product with a nothing special reference cooler (albeit $239-249 launch prices are expected and 4GB version to launch later). AMD simply aimed this as a low end/mainstream card (aka 4850/7850) and they met this goal. I know it's very hard for people to accept that AMD is launching bottom-up, and forfeiting the high-end RX 490 space for now, but it's exactly what some warned/projected about months ago. It's also why GP104 will continue to sell at inflated prices (i.e., with hardly any worthwhile AIB cards anywhere close to $379/599 MSRP). That's why we need competition in the high end space.
 

Head1985

Golden Member
Jul 8, 2014
1,866
699
136
There is no way AMD didn't do simulation on this. It's why this card is a $199-229 MSRP product with a nothing special reference cooler (albeit $239-249 launch prices are expected and 4GB version to launch later). AMD simply aimed this as a low end/mainstream card (aka 4850/7850) and they met this goal. I know it's very hard for people to accept that AMD is launching bottom-up, and forfeiting the high-end RX 490 space for now, but it's exactly what some warned/projected about months ago. It's also why GP104 will continue to sell at inflated prices (i.e., with hardly any worthwhile AIB cards anywhere close to $379/599 MSRP). That's why we need competition in the high end space.
Yeah just like they runs simulation on FIJI right?
 

flash-gordon

Member
May 3, 2014
123
34
101
There is no way AMD didn't do simulation on this. It's why this card is a $199-229 MSRP product with a nothing special reference cooler (albeit $239-249 launch prices are expected and 4GB version to launch later). AMD simply aimed this as a low end/mainstream card (aka 4850/7850) and they met this goal. I know it's very hard for people to accept that AMD is launching bottom-up, and forfeiting the high-end RX 490 space for now, but it's exactly what some warned/projected about months ago. It's also why GP104 will continue to sell at inflated prices (i.e., with hardly any worthwhile AIB cards anywhere close to $379/599 MSRP). That's why we need competition in the high end space.
Yeah, if they went for 48-64ROPs they would probably move the bottleneck to the bandwidth.
 

Flapdrol1337

Golden Member
May 21, 2014
1,677
93
91
Actually.

5.8TFlops is according to that guy's review, matching a 390 with less TFlops.

If true, that's a regression. Not a good result.

It should at the very least match a 390X. Then we can say, they went for perf/w gains without performance regression.

However, I will wait for judgement til I see reviews from more reputable sites. Despite that I already pulled the trigger for 2x RX 480 cos it's the best bang for buck by a mile compared to what's available currently. lol

The 390 has 1.7x the memory bandwidth. Not very strange it performs slightly better per tflop.
 

jj109

Senior member
Dec 17, 2013
391
59
91
However, I will wait for judgement til I see reviews from more reputable sites. Despite that I already pulled the trigger for 2x RX 480 cos it's the best bang for buck by a mile compared to what's available currently. lol

If you "upgrade" from 1 faster card to two marginally slower cards, you'll end up with higher input latency than before in games that scale in AFR because maximum scaling requires 3 frames in flight for 2 GPUs. You're basically betting that non-AFR takes off in a big way.
 
Last edited:

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
Actually.

5.8TFlops is according to that guy's review, matching a 390 with less TFlops.

If true, that's a regression. Not a good result.

It should at the very least match a 390X. Then we can say, they went for perf/w gains without performance regression.

However, I will wait for judgement til I see reviews from more reputable sites. Despite that I already pulled the trigger for 2x RX 480 cos it's the best bang for buck by a mile compared to what's available currently. lol

No. They didn't go for perf/watt gains without a performance regression. They went for performance gains AND performance/watt gains over R9 380/380X. It's the complete lack of understanding of what Polaris 10 replaces which has led to people thinking that a $199-229 card replaces AMD's $329/429 tiers. Tiers is the key word here What AMD 14nm cards will replace last generation $329-429 cards? $300-500 cards! It's common sense.

$329 970 -> $379-449 1070
$549 980 -> $599-699 1080

How is it you have no issues seeing NV replacements and price hikes but you incorrectly position AMD's next gen?

$199 R9 380 -> $199 R (10) = RX 480 4GB
$229 R9 380X -> $229 RX 480 8GB

The x80 name, the pricing tier, the 32 ROPs all add up.

You also know AMD confirmed Vega 10 and Vega 11. You are only setting yourself up for disappointment if you think RX 480 was a Hawaii 390X successor. This is the exact false comparison of HD5770 succeeding HD4890. It's wrong.

If you actually believe what you are saying, then Fury X successor would only match Fury X and use less than 200 watts and AMD would call it a generation. Face it, RX 480 is low-end/mainstream, not even mid-range segment, 3rd tier AMD product line in the new 14nm stack. Since I don't work for AMD, I am of course projecting that they will have 180-200W cards and also 250-275W flagship. If they don't have cards 40-60% faster than 390/390X and Fury/X, this entire generation is a fail for them.
 
Last edited:
Feb 19, 2009
10,457
10
76
The 390 has 1.7x the memory bandwidth. Not very strange it performs slightly better per tflop.

Tonga vs Tahiti. Less bandwidth. 380X matched the 280X and beat it in newer games. It didn't just match the 280.

If RX 480 doesn't match 390X with that TFlops, it's a regression.
 

NTMBK

Lifer
Nov 14, 2011
10,269
5,134
136
^--what that guy said. This is straight out of the Pitcairn playbook, and that chip has been an absolute trooper for AMD- there's a reason they still sell it today. I have no doubt we'll be seeing Polaris 10 rebadges (maybe with GDDR5X) for years to come.
 
Feb 19, 2009
10,457
10
76
No. They didn't go for perf/watt gains without a performance regression. They went for performance gains AND performance/watt gains over R9 380/380X. It's the complete lack of understanding of what Polaris 10 replaces which has led to people thinking that a $199-229 card replaces AMD's $329/429 tiers. Tiers is the key word here What AMD 14nm cards will replace last generation $329-429 cards? $300-500 cards! It's common sense.

$329 970 -> $379-449 1070
$549 980 -> $599-699 1080

How is it you have no issues seeing NV replacements and price hikes but you incorrectly position AMD's next gen?

$199 R9 380 -> $199 R (10) = RX 480 4GB
$229 R9 380X -> $229 RX 480 8GB

The x80 name, the pricing tier, the 32 ROPs all add up.

You also know AMD confirmed Vega 10 and Vega 11. You are only setting yourself up for disappointment if you think RX 480 was a Hawaii 390X successor. This is the exact false comparison of HD5770 succeeding HD4890. It's wrong.

I know it's mainstream, I've been one of the vocal ones about this, small die size, pitcairn like.

The point is per shader and clocks, 5.8 TFlops should be not only matching a 390, with such a big lead in shader performance. Bandwidth isn't the issue either. 256GB/s with memory compression is higher than 290X's 320GB/s.

However, let's just wait for proper reviews before talking about the same thing again and again.
 

sm625

Diamond Member
May 6, 2011
8,172
137
106
What? What about the 750Ti, NV launched Maxwell bottom first... nobody made an issue out of it?

The funny thing about the 750ti is that it really didnt offer anything in terms of increased performance per dollar. All it brought was power efficiency. And yet it outsold pretty much every AMD product that ever existed. This is why I think if Nvidia was launching the exact same gpu as the RX480, it would be called the GTX1060 and would be $299 for the 8GB version, and everyone would be ranting and raving about it and it would sell twice as many units as the RX480 is going to sell.
 

mohit9206

Golden Member
Jul 2, 2013
1,381
511
136
970 performance for 970 price.That's what i am getting so far from that ToT video.Was hoping for more.Maybe other review sites will show better performance using launch drivers.
 

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
I know it's mainstream, I've been one of the vocal ones about this, small die size, pitcairn like.

The point is per shader and clocks, 5.8 TFlops should be not only matching a 390, with such a big lead in shader performance. Bandwidth isn't the issue either. 256GB/s with memory compression is higher than 290X's 320GB/s.

However, let's just wait for proper reviews before talking about the same thing again and again.

You are falling into the Perf/TFlops fallacy. Videocard performance is not directly proportional to ALU/shader speed only. You only have to look at 1070 vs. 1080 to see this in recent launches. It's very hard to make a videocard with a balanced design. The modern designs have a tendency for having excessive ALU horsepower as historically from X1900XT/G80 era shaders became the biggest bottleneck, until later the bottleneck moved to pixel shading power due to higher resolutions.

You are comparing perf/TFlops and ignoring the 32 vs. 64 ROPs and 256GB/sec vs. 384GB/sec differences. Even if the memory compression offsets Hawaii's inefficient/excessive memory bandwidth paper spec, the 32 vs. 64 ROP spec is a big deal when the clock speed differences are less than 30% between P10 and 390X. It's actually amazing RX 480 is even in the ballpark.

It's also why I think RX 470 4GB could be a sleeper mainstream gaming chip, even better value than RX 480 for 1080p gaming. If 470 can be overclocked as well and it maintains the same 32 ROPs and has close to 2000 shaders, it shouldn't be too far behind.
 
Last edited:
May 13, 2009
12,333
612
126
970 performance for 970 price.That's what i am getting so far from that ToT video.Was hoping for more.Maybe other review sites will show better performance using launch drivers.
That's what I'm afraid of. 970 has been around $250 a few times. If all the 480 can offer is 970 performance for $240 then I think I'm going to pass and spend that money on something besides a video card.
 

RoarTiger

Member
Mar 30, 2013
67
33
91
There is no way AMD didn't do simulation on this. It's why this card is a $199-229 MSRP product with a nothing special reference cooler (albeit $239-249 launch prices are expected and 4GB version to launch later). AMD simply aimed this as a low end/mainstream card (aka 4850/7850) and they met this goal. I know it's very hard for people to accept that AMD is launching bottom-up, and forfeiting the high-end RX 490 space for now, but it's exactly what some warned/projected about months ago. It's also why GP104 will continue to sell at inflated prices (i.e., with hardly any worthwhile AIB cards anywhere close to $379/599 MSRP). That's why we need competition in the high end space.

4GB cars launch today and are stocked in some stores
 

jj109

Senior member
Dec 17, 2013
391
59
91
If you "upgrade" from 1 faster card to two marginally slower cards, you'll end up with higher input latency than before in games that scale in AFR because maximum scaling requires 3 frames in flight for 2 GPUs. You're basically betting that non-AFR takes off in a big way.

Actually now that I think about it some more, AFR scaling will always wreck your input latency since it's impossible get 3x scaling on 2 cards, and lowering frames in flight to two will decrease scaling.

So yeah, bad idea to replace 290X with 2xRX480.
 

ShintaiDK

Lifer
Apr 22, 2012
20,378
145
106
Wow, 163W. Talk about a lot of hype and BS. 110W my rear end!



15W idle as well.



Power consumption is a complete disaster! So much for all the LIES and HYPE!
 
Status
Not open for further replies.
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |