[Benchlife] R9 480 (Polaris 10 >100w), R9

jpiniero

Lifer
Oct 1, 2010
14,510
5,159
136
Them calling it the 480(X) is a bit of a surprise, especially they won't have Vega any time soon. Is Polaris 11 mobile only then?
 

Tuna-Fish

Golden Member
Mar 4, 2011
1,324
1,462
136
One thing that all these rumors forget: AMD has publicly stated that this generation there will very likely be more SKUs based on the same chips than usual.

A simple explanation why this would be could be that they have designed the chip for gddr5x, but no gddr5x chips will be available in June. 490/x might well be the same chip but fully enabled, with higher clocks and with gddr5x.
 
Feb 19, 2009
10,457
10
76
Where's the source URL?

Around Fury X performance is outstanding for a 2,560 SP chip.

That power gating and SIMD clock boosting patent tech has to be involved for that kind of IPC gains.
 

raghu78

Diamond Member
Aug 23, 2012
4,093
1,475
136
One thing that all these rumors forget: AMD has publicly stated that this generation there will very likely be more SKUs based on the same chips than usual.

A simple explanation why this would be could be that they have designed the chip for gddr5x, but no gddr5x chips will be available in June. 490/x might well be the same chip but fully enabled, with higher clocks and with gddr5x.

excellent reasoning. Both AMD and Nvidia are waiting on GDDR5X but they could definitely build a chip which can support both GDDR5 and GDDR5X memory. This would allow GDDR5X SKUs to be priced up the stack as they would perform better than GDDR5 based SKUs.
 

raghu78

Diamond Member
Aug 23, 2012
4,093
1,475
136
Where's the source URL?

Around Fury X performance is outstanding for a 2,560 SP chip.

That power gating and SIMD clock boosting patent tech has to be involved for that kind of IPC gains.

AMD GCN resource utilization was a major problem which we saw get to really bad level in Fiji. Actually I think if we go by the 2.5x perf/watt gains then the slower SKUs of Polaris 10 will match Fury X(275w) at around 110w (275/2.5 = 11ow) and the faster SKUs at 130w will easily beat it.
 
Feb 19, 2009
10,457
10
76
I dont think it will beat it on average by much if any.

But minimum FPS, it should. The Primitive Discard Accelerator removes scene complexity before the rendering engine wastes resources on it. In theory, that should yield very high min FPS for the performance class that it falls in. ie. it can punch above it's class for MIN FPS.
 

Saylick

Diamond Member
Sep 10, 2012
3,084
6,184
136
Where's the source URL?

Around Fury X performance is outstanding for a 2,560 SP chip.

That power gating and SIMD clock boosting patent tech has to be involved for that kind of IPC gains.

Fury X @ 130W and 230 mm^2 die is not bad at all.

AMD can easily produce an SKU with double those specs, possibly Vega 10?

>Radeon Pro Duo performance @ 250W and 450 mm^2?
 
Feb 19, 2009
10,457
10
76
Can anyone give me the short about the difference between GDDR5 and GDDR5x? Haven't read much on the GPU game since last year.

5X is 20-50% faster vram. Not as expensive as HBM2. Good for mid-range chips that need more bandwidth to reach their potential. Low-end will stick with cheaper GDDR5.
 

swilli89

Golden Member
Mar 23, 2010
1,558
1,181
136
Where's the source URL?

Around Fury X performance is outstanding for a 2,560 SP chip.

That power gating and SIMD clock boosting patent tech has to be involved for that kind of IPC gains.

https://m.reddit.com/r/Amd/comments/4et8rh/benchlife_a_very_reliable_site_for_leaks_just/
 

raghu78

Diamond Member
Aug 23, 2012
4,093
1,475
136
I dont think it will beat it on average by much if any.

But minimum FPS, it should. The Primitive Discard Accelerator removes scene complexity before the rendering engine wastes resources on it. In theory, that should yield very high min FPS for the performance class that it falls in. ie. it can punch above it's class for MIN FPS.

Silverforce i think we are going to see GDDR5 versions of Polaris 10 at 100-110w match Fury X and GDDR5X versions which might launch later and beat Fury X by 10-20% at 135w. I think AMD has a lot of options just as Nvidia to equip the higher priced SKUs to be GDDR5X based and the lower price higher volume parts to be GDDR5 based. vr-world also mentioned GDDR5X SKUs to be available.

http://vrworld.com/2016/04/13/amd-polaris-10-gpu-beat-competing-pascal/

I think GDDR5 based SKUs could hit USD 249-USD 269 and the GDDR5X SKUs could be priced at USD 349/USD 449.
 
Feb 19, 2009
10,457
10
76
Fury X @ 130W and 230 mm^2 die is not bad at all.

AMD can easily produce an SKU with double those specs, possibly Vega 10?

>Radeon Pro Duo performance @ 250W and 450 mm^2?

Not possibly, it's almost assured now as they reveal their roadmap and show even more gains with Vega.

Logically,

Polaris 11 = 1,280 SP
Polaris 10 = 2,560 SP

These are from rumors though, but it's realistic enough.

Vega 11 would be Polaris 10 + 1.5, or 3,840 SP. Vega 10 would be Polaris 10 x 2, 5,120 SP.

If Polaris 10 = Fury X...

Vega 10 would be better than Fury X CF, cos it won't suffer CF scaling issues, and it's got higher capacity HBM2 with higher bandwidth.
 

Atreidin

Senior member
Mar 31, 2011
464
27
86
Just a quick question, do we know for sure that AMD and/or Nvidia knew about GDDR5X far enough ahead of time to integrate it into their upcoming designs? I mean, you can't just switch it with GDDR5 without designing for it explicitly, right? And adding that capability is going to take a certain amount of time. I'd hope that Micron was smart enough to give everyone a heads up since I'm sure they want to make money selling it, but I'm wondering if people are expecting GDDR5X will arrive sooner than is realistic. This is an honest question, I haven't followed the development timelines of any upcoming chips very closely.
 

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
Polaris 10 to be named 480, around Fury X performance, 130 watt

If true, I'll be eating crow.

Where's the source URL?

Around Fury X performance is outstanding for a 2,560 SP chip.

That power gating and SIMD clock boosting patent tech has to be involved for that kind of IPC gains.

I know. Sounds too good to be true. Fury X peaks at 280W.

power_peak.gif


AMD GCN resource utilization was a major problem which we saw get to really bad level in Fiji. Actually I think if we go by the 2.5x perf/watt gains then the slower SKUs of Polaris 10 will match Fury X(275w) at around 110w (275/2.5 = 11ow) and the faster SKUs at 130w will easily beat it.

After Hawaii and Fiji launch, I am going to stay on the side of cautiously optimistic. Similar hype was created around Hawaii and even more for Fury X and we know how that went.

Vega 10 would be better than Fury X CF, cos it won't suffer CF scaling issues, and it's got higher capacity HBM2 with higher bandwidth.

Even if CF worked, Fury X isn't that much faster than an R9 290X. By extension, Fury X CF isn't that much faster than an R9 295X2 (non-thermal throttling 290X).

perfrel_3840.gif


I have no doubts that AMD's flagship 14nm card and NV's flagship Big Pascal (Full die) will easily outperform Fury X CF at 4K. Besides, Fury X has very little overclocking headroom which means if those 14nm/16nm cards have another 15-20% OCing headroom Fury X CF stands no chance. Hopefully Vega and Big Pascal will finally be a good upgrade for those still clinging to 780Ti SLI / 970 SLI/290 CF/ 290X CF. But man it's going to be almost a year wait :(.
 
Last edited:
Feb 19, 2009
10,457
10
76
Some interesting stuff from Neogaf (!) lol.. it's related to Polaris, but in the Nintendo NX (dev kits already out for games in the making).

http://www.neogaf.com/forum/showpost.php?p=201100330&postcount=4778

Nintendo and devs are hyped about the hardware, particularly the Primitive Discard Accelerator.

What DICE does with Frostbite & AMD's GPUOpen GeometryFX (http://gpuopen.com/gaming-product/geometryfx/) do in software/shaders, Polaris will do in hardware much faster. Automatically, game engine independent.
 
Last edited:

maddie

Diamond Member
Jul 18, 2010
4,723
4,628
136
If true, I'll be eating crow.



I know. Sounds too good to be true. Fury X peaks at 280W.
Why say this.

The performance/watt ratio takes into account all the variables. So at 130W Polaris 10,
If 2X, then a present 260W AMD card. If 2.5X then a present 325W AMD card. First equals Fury - Fury X. Second easily surpasses Fury X
 

tential

Diamond Member
May 13, 2008
7,355
642
121
We have no idea the level of updates in these GPU architectures but if Polaris is in the new NX console, that's huge... I mean, we've seen the console effect before and how that has benefited Hawaii, having that for a NEW product like Polaris in which the Xbox One and PS4 also refresh with potentially Polaris chips as well, that would be massive.

As for these Vega rumors....
Well, if these Polaris 10 = Fury X rumors are true....
We should probably start fighting to pre order Vega 10 and Big Pascal now.... Because the performance increase over that would be monsterous, they could definitely price these new flagships at $1000.... I still will see every single one sell I bet.

Polaris 10 will not be 250 based off GDDR5 skus, that's way too little for a LOT of performance. Definitely over 300. You're way too optimistic
 

tviceman

Diamond Member
Mar 25, 2008
6,734
514
126
www.facebook.com
Sounds too good to be true to have Fury X performance with 192-224gb/s bandwidth given how GCN is very, very bandwidth hungry but I'll definitely be happy if it is true! However,, given that the rumor said "around Fury X performance" that could mean a lot of things. Tied Fury X at 1080p? 5% slower at 1440p? 10% slower at 4k? Any of those could either be fantastic news or downright disappointing.

Either way, bring on the new GPUs!
 
Last edited:

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
Sounds too good to be true to have Fury X performance with 192-224gb/s bandwidth given how GCN is very, very bandwidth hungry but I'll definitely be happy if it is true! However,, given that the rumor said "around Fury X performance" that could mean a lot of things. Tied Fury X at 1080p? 5% slower at 1440p? 10% slower at 4k? Any of those could either be fantastic news or downright disappointing.

Either way, bring on the new GPUs!

It's reminiscent of Zen hype in the other sub-forum -- 8 core Zen with Haswell IPC with 4.2-4.4Ghz boost, and while at it in a 95W TDP and priced under $400. This is how AMD products are set up to fail on this forum. A certain minority will keep hyping something up and when data comes out that shows it may not live up to their expectations, they disregard it and hype up the expectations even more.

Thus far, I have not seen 1, not 1 reliable source, that has been able to produce any evidence to support the claim that Polaris 10 will = Fury X @ 130W TDP.

It's all like statements that someone makes online, the statements themselves aren't verifiable. But then it magically becomes the truth and people go with it.

For example, in AMD's own slides, they claim 2.5X perf/watt compared to 2014 GCN parts. What does that tell me as a reader? Well it 100% means the 2.5X improvement had nothing to do with it over Fiji. And yet, here we are with people going along with a 2560 shader 130W chip ~ 280W Fiji. So this forum magically went from having 390X +5-10% performance in Polaris 10 (280W vs. 130W), to now claiming that nope, it's not good enough. Let's just bump that up to Fury X. And then others on other forums (WCCFtech or Videocardz) even go on as far as to say Polaris 10 is a fail if it cannot easily beat 980Ti/Fury X. It's because people cannot accept that NV's GP104 will smash Polaris 10 into the ground and that AMD is 6 months behind with Vega 11.

I personally will not believe that Polaris 10 with 8Gbps 256-bit memory bus and rumored 230-240mm2 die size will match the Fury X.

Like think about it, on April 26, AMD is launching Radeon Pro Duo with an expensive water cooler for $1500 US, but just 1.5 months later here comes Polaris 10, with 8GB of VRAM, using 1/2 the power and costing way less than a Fury X? So why wouldn't AMD just cancel that Radeon Pro Duo and release a dual-Polaris 10 card without expensive water cooling, more VRAM, better features (HDMI 2.0a, DP1.3, etc.)

Logically, this tells me Polaris 10 is being overhyped everywhere online. It's a staight up midrange product that's now being hyped way way up since it looks like AMD will have nothing to counter 1080. It also contradicts AMD's own strategy of bringing 290/290X spec to more affordable (sub-$349 price levels: AMD's own words, own presentation). So now I am supposed to believe that Polaris 10 will cost less than $349 and be as fast as the Fury X? I am just following along all of this logic.

Sorry, but I don't buy it. AMD's track record speaks for itself -- 290X, Fury X, massive hype and both failed to deliver. I'd rather be cautious than hype up another midrange AMD chip.

Even sites like Videocardz are starting to post pure garbage like Polaris 10 replacing Fury X, the same site that posted AMD's roadmap showing Vega as a Fury X replacement. Facepalm!
 
Last edited:
Feb 19, 2009
10,457
10
76
RS, 390X + 10% is within range of Fury X, especially at 1080p.

I don't agree with your regarding the 290X failing to deliver. Are you thinking straight? It beat Titan as it was designed to do, at nearly half the price. Even with that crap reference cooler.
 

Qwertilot

Golden Member
Nov 28, 2013
1,604
257
126
Tying/nearly doing so with Fury X at 1080 seems relatively plausible? The 390X nearly does that. Imagine the bandwidth difference will show further up the resolution scale.

Plain old Fury is a fairer baseline comparison mind. Unless they do a very heavily clocked/water cooled P10!
 

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
RS, 390X + 10% is within range of Fury X, especially at 1080p.

I don't agree with your regarding the 290X failing to deliver. Are you thinking straight? It beat Titan as it was designed to do, at nearly half the price. Even with that crap reference cooler.

Hold on, so now we are supposed to care about 1080p performance? Come on Silver. 1080p is a CPU limited resolution. Using it to test high end GPUs without SSAA/VSR/DSR is worthless.

Check 980Ti vs. 970/980/780Ti at 1080/1440/4K. The performance delta grows in favour of 980Ti as resolution goes up since there are A LOT of games which are CPU limited at 1080p.

I am not going to spend $500-650 for a next gen flagship card to play at 1080p. A single $225 290 is more than enough for that. Please, don't start pulling out the 1080p stuff since then I'll pull out i7 6700K @ 4.6Ghz gaming benchmarks that will show that anyone on here with i7 4790K @ 4.9Ghz and below is bottlenecking their GPU.

That's a FACT:
https://m.youtube.com/watch?v=f5lfMogcrPU

If people want to waste $350+ on a next gen card with Fury X level of performance or higher to play at 1080p, that's their business. However, when we want to compare GPUs in GPU-limited scenarios, 1080p 60Hz results are far less relevant.
 
Last edited:

RussianSensation

Elite Member
Sep 5, 2003
19,458
765
126
Tying/nearly doing so with Fury X at 1080 seems relatively plausible? The 390X nearly does that. Imagine the bandwidth difference will show further up the resolution scale.

Plain old Fury is a fairer baseline comparison mind. Unless they do a very heavily clocked/water cooled P10!

That's because 1080p is a peasant resolution and CPU limited mess. I am pretty sure a single 980Ti and my Tri-Hawaii setup are almost as fast at that resolution. That's because 2 of my GPUs are sitting there and enjoying the sauna while they are waiting for Mr. Skylake to bring in more hot water to pour on the rocks. Until then, they aren't even breaking a sweat. But guess what, I didn't buy that level of GPUs to game at 2002 resolution. No sir, I already had 1600x1200 in 1998.

If someone wants to have 70-80% GPU usage on 1080p 60Hz after spending $500+, they can knock themselves out. The only benchmarks that matter to me are 1440/1600p and 4K (or similar). If Polaris 10 is as fast as the Fury X at 1080p, but is 20% slower at 1440p, it's not a high end card. End of story.

It's hurting the mind how in the last 5 years we looked up to 1440p and 4K but now that Keplet and Maxwell perform worse at those resolutions, suddenly 1080p 60Hz is the benchmark? If we are going to discuss 1080, lets talk 120-144Hz or 1080p with DSR/VSR, etc. Besides, the vast majority of 1080p monitors are cheap junk and tiny 24" and below eye sore. The only difference is most gamers still using them are too stubborn and ignorant to realize it.

I've never met 1 person who said, no that 27-34" 1440p/4K monitor is junk; I'll go back to my epic 23 inch 1080p 60Hz one. While at it, why not get an XB1? 720p @ 30 FPS, more cinematic, doesn't require anything above a 380X. EPIC.
 
Last edited:

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
Why say this.

The performance/watt ratio takes into account all the variables. So at 130W Polaris 10,
If 2X, then a present 260W AMD card. If 2.5X then a present 325W AMD card. First equals Fury - Fury X. Second easily surpasses Fury X

Figuring in the difference in HBM vs. GDDR5 could make the overall card a bit less than 2.5 even if the GPU does it.