Regarding the RX 480 AOTS benchmark at Computex

[Fulgrim]

2 minutes ago, Demonking said:

because AMD makes their own CPU's which compete with intel, so by using their competition product instead of their own to test another of their own product that just shows they have no faith in said product, but expect people to buy it. how is this so hard to understand.  

It's not a matter of faith. It's a matter of their FX chips being inferior to the current high-end segment of the desktop CPU market. Denying it is plain stupidity. And so is using faith as a reasoning for them to use FX chips in benchmarks.

[Trixanity]

5 minutes ago, Demonking said:

because AMD makes their own CPU's which compete with intel, so by using their competition product instead of their own to test another of their own product that just shows they have no faith in said product, but expect people to buy it. how is this so hard to understand.  

We've all understood it, told you why it is so, but you just keep on digging. I'd say this is more embarrassing than AMD throwing Bulldozer under the bus. AMD is watching this exchange and feeling fremdscham. I guarantee it.

[ivan134]

1 hour ago, That Norwegian Guy said:

AoTS is the only natively developed DX12 title. Hitman, QB and TW: Warhammer included DX12 support mid-development and aren't developed ground up to use it.

What is relevant is the implementation not the when. The argument can be made against Hitman, that async compute is basically doing nothing in that game, but not Total War or Quantum Break. Quantum Break uses it for their lighting and I don't need to tell you what it's used for in Warhammer. Maybe you want the info from the horse's mouth?

 

 

[ANewFace]

5 hours ago, DarkBlade2117 said:

So you can have a mobile oven in your backpack? Razers idea of cooling is no better than Apple's idea of cooling.

 

You think I want to carry around a 17in 7-9lb laptop? Of course not, ill take the heat for the portability 

[Doobeedoo]

Well mainly all for demo and all. Expect ton of benchmarks soon anyway.

[RagnarokDel]

5 hours ago, TidaLWaveZ said:

Actually...

 

The reference card is $450

Exactly my point.

We're talking about reference prices. We dont know what AIB partners will do for the 480, for all you know there could be a 170$ ultra basic 480.

1 hour ago, Demonking said:

because AMD makes their own CPU's which compete with intel, so by using their competition product instead of their own to test another of their own product that just shows they have no faith in said product, but expect people to buy it. how is this so hard to understand.  

Actually for benchmarking purpose, it's probably more reliable to use a competitor's CPU instead of your own. Nobody can create conspiracy theories about AMD creating voodoo magic using a CPU chip at 8 ghz or some shit

 

 

[WereCat]

1 hour ago, Demonking said:

I'm going to leave it at this before an admin gets on us. If i was the engineer behind FX i would be embarrassed by this. If i was AMD  would still use fx to this day to show i had faith in my product from 4 years ago showing it still can compete.

Come on....

If you are showing your new GPU then it makes sense to use the best available CPU so that you can show full potential of your new product.

And you want to make damn sure that your new product will appeal as best as it can to the people.

 

What good for it would be if you hindered the performance of your new GPU because you are so proud of your 5 year old CPU that is soon to be discontinued?

 

I am sure AMD would love to use their CPU for these tests but Zed is obviously not ready yet.

[Citadelen]

1 hour ago, Demonking said:

I did if you look above  would know that. This doesn't change the fact that it is embarrassing.

I think they'd rather be embarrassed than be seen as stubborn.

[DarkBlade2117]

1 hour ago, ANewFace said:

You think I want to carry around a 17in 7-9lb laptop? Of course not, ill take the heat for the portability 

You know Dell, HP, Gigabyte, MSI ALL have notebooks that weigh less then 5 pounds less. Why get a 1070m or what not when it'll thermal throttle in the Blade? I rather not have my laptop sit at 90C+

The blade is not a good laptop, especially for the price. Someone can suggest a better one for a much cheaper price. I think it is HP specifically who had one.

[TidaLWaveZ]

1 hour ago, RagnarokDel said:

Exactly my point.

We're talking about reference prices. We dont know what AIB partners will do for the 480, for all you know there could be a 170$ ultra basic 480.

Actually for benchmarking purpose, it's probably more reliable to use a competitor's CPU instead of your own. Nobody can create conspiracy theories about AMD creating voodoo magic using a CPU chip at 8 ghz or some shit

 

 

We're really talking about graphics card solutions and there's a little part of the AMD press conference where they announce that the RX 480 will be starting at $199. You could be right about third parties releasing a cheaper version, but I interpret what they said as the baseline is $199.

[Big-G]

I'm just gonna say this...

 

AotS was the worst mistake they could do. The market currently perceives it as an AMD biased benchmark, especially after the 390X vs Titan X fiasco.

 

They keep pulling these dumb-arse benchmarks that just paint them in a bad light. Why not use a benchmark considered less biased? Something to hide?

[The Benjamins]

1 hour ago, GidonsClaw said:

I'm just gonna say this...

 

AotS was the worst mistake they could do. The market currently perceives it as an AMD biased benchmark, especially after the 390X vs Titan X fiasco.

 

They keep pulling these dumb-arse benchmarks that just paint them in a bad light. Why not use a benchmark considered less biased? Something to hide

It's biased due to nvidia's architecture. It is a great example of good dx12 games having gains on amd hardware. Which you will not get on nvidia's hardware. 

[Belgarathian]

2 hours ago, GidonsClaw said:

I'm just gonna say this...

 

AotS was the worst mistake they could do. The market currently perceives it as an AMD biased benchmark, especially after the 390X vs Titan X fiasco.

 

They keep pulling these dumb-arse benchmarks that just paint them in a bad light. Why not use a benchmark considered less biased? Something to hide?

I agree with you that the results are bias, but it's because AMD push it so hard because it is by far the most well developed DX12 game. It's engine can and has demoed pretty much all the benefits of DX12 & Async compute. If another game offered that level of depth, I'm sure they would demo that too. 

 

Nvidia is hamstrung on some of the bigger benefits of DX12 thanks to their architecture. 

[Big-G]

3 minutes ago, Belgarathian said:

I agree with you that the results are bias, but it's because AMD push it so hard because it is by far the most well developed DX12 game. It's engine can and has demoed pretty much all the benefits of DX12 & Async compute. If another game offered that level of depth, I'm sure they would demo that too. 

 

Nvidia is hamstrung on some of the bigger benefits of DX12 thanks to their architecture. 

I wouldn't purely call it an uArch issue, it's also driver interpretation of compute instructions to pass to the CUDA's... The 1080 drivers are very new, AMD has had a long time to sort out their stuff since Fury X released....

[Belgarathian]

2 minutes ago, GidonsClaw said:

I wouldn't purely call it an uArch issue, it's also driver interpretation of compute instructions to pass to the CUDA's... The 1080 drivers are very new, AMD has had a long time to sort out their stuff since Fury X released....

AMD has been playing with Async compute since Mantle which was 4 years ago or there abouts. Most of the oxide engine (used in AotS) was used in the Star Swarm demo for Mantle. 

[JAKEBAB]

OZ? Does he mean Australia? If so where is he getting the reliable Internet? lol

 

Didn't they only show one benchmark though? Why not more games? I don't believe there lying, but I definitely think there hiding something if this was the only game benchmarked.

[evilarceus]

I was going to put my money on the GTX 1070...

 

but I guess not...

[The Flying Sloth]

@App4that What I meant about 480

[App4that]

41 minutes ago, The Flying Sloth said:

@App4that What I meant about 480

http://www.fudzilla.com/news/graphics/40808-amd-radeon-rx-480-crossfire-rendering-differently-vs-gtx-1080

[MoonSpot]

13 hours ago, -BirdiE- said:

Yeah... Well if the shaders working improperly means the game looks better and is less taxing on a GPU, then please sign me up for the broken shading...

Looking at the lack of snow on the path/dirt road for the comparison during AMDs event lends credit to it being broken on the 1080.  I expected more from you than the level of pettiness that statement makes.

15 hours ago, TidaLWaveZ said:

 

~$140 price difference.  May be a deal breaker for certain budgets, but IMO justifies a single card setup.

Interesting maths, might I ask how did you arrive at that?  I assume you meant to say $240 since 700-230-230 =/= 140

[DXMember]

I guess that application specific optimization didn't pay off after all (:

http://www.geek.com/games/futuremark-confirms-nvidia-is-cheating-in-benchmark-553361/

http://www.geek.com/games/nvidia-still-cheating-even-with-latest-3dmark-build-552713/

[vanished]

2 minutes ago, DXMember said:

I guess that application specific optimization didn't pay off after all (:

http://www.geek.com/games/futuremark-confirms-nvidia-is-cheating-in-benchmark-553361/

http://www.geek.com/games/nvidia-still-cheating-even-with-latest-3dmark-build-552713/

Cheating in benchmarks has been going on for countless years... something needs to change.  I want to see a $1 billion fine every time someone's found guilty.  This is serious stuff - people base their buying decisions on results like this and so it's no different than that massive VW emissions scandal.

[DXMember]

1 minute ago, Ryan_Vickers said:

Cheating in benchmarks has been going on for countless years... something needs to change.  I want to see a $1 billion fine every time someone's found guilty.  This is serious stuff - people base their buying decisions on results like this and so it's no different than that massive VW emissions scandal.

well the massive VW and their daughter companies together with the Japanese Mitsubishi are killing trees and fluffy animals with their emission scandal

but I do agree with you - they should build an assembly factory in Europe and sell the ICs and microprocessors at a discount for that

[Big-G]

2 minutes ago, Ryan_Vickers said:

Cheating in benchmarks has been going on for countless years... something needs to change.  I want to see a $1 billion fine every time someone's found guilty.  This is serious stuff - people base their buying decisions on results like this and so it's no different than that massive VW emissions scandal.

The thing is, it's well known... the only thing Nvidia guarantees is the base clock, boost clock is an estimation, but GPU Boost will always try and get the best clock range it can get within limits. So technically it is not cheating... you are getting the best you can get...

 

AMD on the other hand advertises a boost clock, but generally the card will throttle down from that to match the environment. So essentially most AMD owners always get less than they thought...

[2FA]

6 hours ago, Demonking said:

because AMD makes their own CPU's which compete with intel, so by using their competition product instead of their own to test another of their own product that just shows they have no faith in said product, but expect people to buy it. how is this so hard to understand.  

Intel has the i7 980X from 4 years ago, why aren't they using their own product?