( Updated )Star Wars Battlefront Benchmarked On Nvidia And AMD, Runs Best On Radeon – Exceptionally Well Optimized Across The Board

[patrickjp93]

Honestly, it would've been so much easier if they just wrote what editions of cards they used since we can see 1 WF card in the bench . Then I could have an argument if the cards in the benchmarks are the best or worst possible scenarios for the specific cards.

The AMD cards are all Sapphire Tri-X and ofc the Fury X is its water cooled reference design. Nvidia's are reference designs with GPU Boost disabled. Biased article is biased. <-- If you reached this point without realizing the sarcasm, I feel sorry for you.

[Mr_Troll]

The AMD cards are all Sapphire Tri-X and ofc the Fury X is its water cooled reference design. Nvidia's are reference designs with GPU Boost disabled. Biased article is biased. <-- If you reached this point without realizing the sarcasm, I feel sorry for you.

do you work at guru3d? show me the line where they say the use reference NV cards with gpu boost disabled...pls

[patrickjp93]

do you work at guru3d? show me the line where they say the use reference NV cards with gpu boost disabled...pls

I work at Alphabet. All knowledge is belong to me. No other citation required.

[don_svetlio]

The AMD cards are all Sapphire Tri-X and ofc the Fury X is its water cooled reference design. Nvidia's are reference designs with GPU Boost disabled. Biased article is biased. <-- If you reached this point without realizing the sarcasm, I feel sorry for you.

Much Fanboy, Such Salty, So Drama

[don_svetlio]

I work at Alphabet. All knowledge is belong to me. No other citation required.

"I am my own source - BELIEVE ME PEASANTS!" - you are starting to sound like a Cult Leader here.

[Khvarrioiren]

I work at Alphabet. All knowledge is belong to me. No other citation required.

 

No, but really do go ahead and provide backup to your claim.

[Prysin]

So instead of being rip graphics cards it would be rip CPU? If that's the case I'm glad I didn't stick with the i5 4440.

well, at stock yes.

If you OC the i7, it should hold out much longer

i think the only games were you could hit CPU limits these days, anyway, will be games who are heavily drawcall based with massive number of units...

games like Ashes, Total War series, civilisation etc... RTS in general, can hit CPU limits even today...

[Dabombinable]

well, at stock yes.

If you OC the i7, it should hold out much longer

i think the only games were you could hit CPU limits these days, anyway, will be games who are heavily drawcall based with massive number of units...

games like Ashes, Total War series, civilisation etc... RTS in general, can hit CPU limits even today...

I can run my 4790K safely at 4.8GHz, so it will be fine for a long time. The i5 however can only be pushed to 3.25GHz base.....and its version of turbo boost is useless unlike turbo boost on Devil's canyon-the boost speed might as well be the base clock.

[patrickjp93]

No, but really do go ahead and provide backup to your claim.

As for the consistency, already did. The card models claim was sarcasm and highlighted as such.

[don_svetlio]

As for the consistency, already did. The card models claim was sarcasm and highlighted as such.

https://en.wikipedia.org/wiki/Sarcasm

[Doughnutnator]

Silicon lottery or not, 1400 MHz on 2.0 is not happening. Please back up your statement. You were just calling out patrickjp93 for that. I don't remember ever seeing it giving more than 100-150, let alone 300 or 400 MHz (because that needs playing with voltages and afaik 2.0 can't do that).

My GTX 970 boosts up to 1430Mhz out of the box... A friend of mine has a 980 and it's almost the same, it boosts up

I guess that it's luck. Anyways AMD is stronger in this game than Nvidia. It would be interesting to see results with Maxwell overclocked, sometimes the extra 100-200Mhz makes a difference...

[IMPERIUS]

Why cant all games be as optimised as this one?

[Kinda Bottlenecked]

Why cant all games be as optimised as this one?

 

If not, Nvidia wouldn't get that 82% market share 

 

 

:lol:

[SteveGrabowski0]

Wow, the 380 just murders the 960 like they were completely different classes of GPU.

[ACatWithThumbs]

Yes for the small group of hardware enthusiasts like us we OC our GPUs you'd be surprised how many people don't overclock or give a fuck about overclocking, even when I offered to do it for them.

Those people also don't buy a $300-500 GPU's the people that buy those GPU's are mostly enthusiasts.

And almost every benchmark ignores the GPU features like MFAA which increases the performance when using MSAA by almost 10fps on Maxwell.

A benchmark should be done in the way the cards are actually used if a game as an example has Mantle it should be in the benchmark as well.

 

Remember that benchmarks turn on all the eye candy that normal people don't use. I'm pretty sure most settings are on ultra or very high and AA is at 16x.

I'm playing the game already maxed out the game also has no MSAA it only has FXAA/TAA and a resolution scale to downsample.

[typographie]

Would it be wrong to assume nVidia will release better drivers once the game is released?

 

They may. I think all of EA's Frostbite 3 engine games have been associated with AMD dating back to Battlefield 4, and I think some of the performance discrepancies got ironed out over time.

 

This isn't really anything to worry about, though. Yeah, Nvidia's cards are a few FPS behind where we'd expect them to be, but they're all achieving very good performance regardless.

[Prysin]

If not, Nvidia wouldn't get that 82% market share 

 

 

:lol:

sort of true really...

if driver "magic" wasnt ever a thing, the GTX 900 series would be battling to even get foothold over the R9 200 series...

[Humbug]

Guru3D has updated the benchmark with Titan X, Titan Black, GTX 770, GTX 950 and R9 280x.

770 beats 960.

280x is tied with 380, knowing that GPU it will probably pull well ahead of the 380 when overclocked.

950 strangely is matching 960, which suggests to me Nvidia needs to tune the 960 driver specifically.

 

[DEDRICK]

I am happy with my 290 1440p numbers, I want to buy another for 1 GPU per eye VR

[FratStar]

Those people also don't buy a $300-500 GPU's the people that buy those GPU's are mostly enthusiasts.

And almost every benchmark ignores the GPU features like MFAA which increases the performance when using MSAA by almost 10fps on Maxwell.

A benchmark should be done in the way the cards are actually used if a game as an example has Mantle it should be in the benchmark as well.

 

I agree with you to an extent I have a small sample size (because I'm one person), but I have friends with a 780ti, r9 390, 2 660tis in SLI, a 680 classified, another with a 390, 290s in CF, and none of them overclocked their GPUs. I have more friends with custom PCs, but I don't know if they overclock. Not everyone that spends money on a GPU is a hardware enthusiast they might just be gamers that want to be positive they max out their games and play at 1080p 60fps all the time.

 

Also there reason why they don't use those features like MFAA is because that is comparing Apples to oranges. AMD cards don't even support MFAA, Nvidia cards don't support Mantle etc. Benchmarks have to be in a Apples to Apples comparison otherwise we can't objectively compare the cards. Further more games have to support features like MFAA and not many do just like a lot of games don't support TXAA or AMDs EQAA.

Everyone is well aware that Nvidia does not use the same methodology with their technologies that AMD does. I'm not gonna be like "oh black boxed middleware," it's like TressFX vs HairWorks while Hairworks looks better IMO, but TressFX is open source as a programmer it much easier to optimize for code you can actually look at (At least in my experience as a developer).

[MageTank]

It is - GPU Boost 2.0 can get a Maxwell card to 1400 out of the box. That is not stock.

 

960 @ 1350 out of the box. Enjoy

http://techreport.com/review/27806/five-geforce-gtx-960-cards-overclocked/2

Sorry, but this is just not true for most Maxwell cards. You are showing factory OC'd cards that boost higher. Understand, reference GTX 960 is clocked in at 1127mhz, boost clock is 1178, but real world boost is somewhere in the ball park of 1200-1225 depending on multiple factors (thermals, lottery, etc). Each of the cards you linked, has a factory OC of 100mhz or more, and after market coolers designed to handle that extra cooling. They are not only clocked higher out of the box, but the improved coolers allow them to boost even higher too. 

 

Looking at Maxwell across the full spectrum of their lineup (from the GTX 750, all the way to the GTX Titan X) it's safe to say your 1400mhz number is not accurate for all Maxwell cards. 

[danieltoruno]

I don't see the Titan X on any of those graphs.

 

How is its performance?

IMO Titan X was a poorly designed card. Asides from that, its way overpriced. 980ti gets about the same performance for $350 cheaper. Titan X actually got worse frame rates than the 980ti running this game.

[danieltoruno]

I can't wait to get home and test this game out on my 3440x1440 monitor! I'm so hyped! Star Wars Battlefront II for PS2 was my childhood!

[typographie]

IMO Titan X was a poorly designed card. Asides from that, its way overpriced. 980ti gets about the same performance for $350 cheaper. Titan X actually got worse frame rates than the 980ti running this game.

 

Usually when a 980 Ti outperforms a Titan X, there's some thermal throttling at play. Though I suppose that is consistent with the "poorly designed" point.

[don_svetlio]

Sorry, but this is just not true for most Maxwell cards. You are showing factory OC'd cards that boost higher. Understand, reference GTX 960 is clocked in at 1127mhz, boost clock is 1178, but real world boost is somewhere in the ball park of 1200-1225 depending on multiple factors (thermals, lottery, etc). Each of the cards you linked, has a factory OC of 100mhz or more, and after market coolers designed to handle that extra cooling. They are not only clocked higher out of the box, but the improved coolers allow them to boost even higher too. 

 

Looking at Maxwell across the full spectrum of their lineup (from the GTX 750, all the way to the GTX Titan X) it's safe to say your 1400mhz number is not accurate for all Maxwell cards. 

A reference 970/980 cannot OC for shit - THERMAL GATE PARTY!

Really though - if you compare reference vs reference then the Fury X > 980 Ti as the GM200 core throttles like a boss.

If you want to compare reference vs reference then Maxwell are not worth a damn considering they MUST be OCed to stand their ground.

And a page or so back a guy got 1450 on his Maxwell GPU without touching shit. Go educate yourself, run along now.