Battlefield 1 Benchmarks (Gamegpu & the rest)

[tviceman]

I wouldn't upgrade yet, R9 290 to GTX1070 is not worth it for BF1 even for 2560x1440.

But the following Sapphire has a nice perf/$

Sapphire Fury at $309,99 + 10% off w/ promo code EMCFEFT22, ends 10/20 = $280
+
Display Port to DVD-D cable = $12

Total = $280

Edit: As you said, R9 290 will be ok at 1440p Medium. Customizing the settings more and im sure you can even use Ultra Textures if you lower other settings.

Edit 2: no DP to DVI needed

Disagree entirely. Best case 1070 is 61% faster than best case 290. That is completely worth it, IMO. Best case 1070 is 15% faster than Fury, consumes 50 less watts, has double the vram, and can be had fairly regularly now on jet.com or ebay for ~$360 or $390 pretty much anywhere with Gears of War game code. Faster, consumes less power, double the vram (future proofing); it's a much better purchase decision. Fury will be dead in a year at 1440p and resale value will be abysmal. On top of that, as Carfax said above Nvidia hasn't released optimized drivers yet for BF1. What we're seeing for Nvidia right now is worst case scenarios.

 

[Carfax83]

Disagree entirely. Best case 1070 is 61% faster than best case 290. That is completely worth it, IMO. Best case 1070 is 15% faster than Fury, consumes 50 less watts, has double the vram, and can be had fairly regularly now on jet.com or ebay for ~$360 or $390 pretty much anywhere with Gears of War game code. Faster, consumes less power, double the vram (future proofing); it's a much better purchase decision. Fury will be dead in a year at 1440p and resale value will be abysmal. On top of that, as Carfax said above Nvidia hasn't released optimized drivers yet for BF1. What we're seeing for Nvidia right now is worst case scenarios.

The amount of people trying to sell Fury is unbelievable, despite the fact that's essentially a crippled high end card with only 4GB of VRAM, plus several other negatives.. And when you bring up the issue of VRAM, it's like they don't even want to hear it, as you can see by this thread.

 

[Bacon1]

Gears of War 4 GTX 1060 is on par with Fury leave alone RX 480.

Are we not looking at the same graphs?

480 is 2% faster @ 1080p, same avg @ 4k but higher minimums than 1060. Fury X has 20% faster @ 1080p, 28% @ 4k.

In what world is 20% slower "on par"?

Forza the 480 is slightly faster than the 1060 with 7% better minimums and at 980 ti levels. All non-Pascal / Polaris do terribly in that game, as seen by the 480/1060 almost matching OC'd 980 ti and both have 70-80+% higher minimums.

GoW:UE was a DX9 -> DX12 port, pretty much the opposite of "native" DX12.

 

[Bacon1]

The amount of people trying to sell Fury is unbelievable, despite the fact that's essentially a crippled high end card with only 4GB of VRAM, plus several other negatives.. And when you bring up the issue of VRAM, it's like they don't even want to hear it, as you can see by this thread.

You mean the thread that shows that even @ 3440x1440 the limited 4GB Fury can run Ultra settings and not take much of a hit? The exact same thread where you've been asked to test and have failed to do so and been proven wrong multiple times?

Pretty great thread actually

 

[Unreal123]

Are we not looking at the same graphs?

480 is 2% faster @ 1080p, same avg @ 4k but higher minimums than 1060. Fury X has 20% faster @ 1080p, 28% @ 4k.

In what world is 20% slower "on par"?

Forza the 480 is slightly faster than the 1060 with 7% better minimums and at 980 ti levels. All non-Pascal / Polaris do terribly in that game, as seen by the 480/1060 almost matching OC'd 980 ti and both have 70-80+% higher minimums.

GoW:UE was a DX9 -> DX12 port, pretty much the opposite of "native" DX12.

I mean games ,which only has DX12 option to run always tend be better on nvidia. Games ,which has DX11 and DX12 options tends be good on AMD ,however, still cannot beat nvidia on DX11 with its counter parts in the same game. For example, total warhammer, QB , Battlefield 1 and Deus EX mankind divided.

 

[Carfax83]

You mean the thread that shows that even @ 3440x1440 the limited 4GB Fury can run Ultra settings and not take much of a hit? The exact same thread where you've been asked to test and have failed to do so and been proven wrong multiple times?

Pretty great thread actually

Until I see video footage, I don't believe it. I'm sure it would be stuttering like Porky the Pig, given your frametime raw dump, and the other footage that I linked to on YouTube with the Fury's big brother, running at 1080p ultra quality. If a Fury X couldn't handle 1080p ultra quality, I don't think your standard Fury can handle the game at 3440x1440 with ultra quality textures and most of the settings turned on..

 

[Grooveriding]

What is sad about Battlefield 1 performance is how much it highlights what a miserable upgrade the 1080/1070 are for 980ti users. 1070 is at best equal to 980ti at stock and will be slower after overclocking with 980ti's 25-30% overclock headroom. 1080 manages about 20% over a 980ti, which will shrink to 10-15% again with GM200's superior overclocking compared to GP104.

Nvidia really needs to put out their 1080ti ASAP and provide a real upgrade for 980ti users that are sitting on their cards not wanting to spend $1200 for the real GM200 upgrade. The 1080 looks like a mild refresh compared to 980ti in this game.

 

[Carfax83]

What is sad about Battlefield 1 performance is how much it highlights what a miserable upgrade the 1080/1070 are for 980ti users. 1070 is at best equal to 980ti at stock and will be slower after overclocking with 980ti's 25-30% overclock headroom. 1080 manages about 20% over a 980ti, which will shrink to 10-15% again with GM200's superior overclocking compared to GP104..

It's only that way if you look at it strictly from a performance perspective, and even then, not taking into account DX12. An overclocked 980 Ti is a beast to be sure. I know because I have one. But when you consider how much more power the 980 Ti uses at high clock speeds ie 1450 Mhz and up, the increased noise from having to boost your fan speeds, less VRAM, inferior HEVC decoding, and the lower DX12/Vulkan performance, then both the GTX 1080 and 1070 looks like something special by comparison.

I don't need to overclock my GTX 1080, or even mess with the fan speed. It automatically boosts above 2 Ghz, but will run stable at 2 Ghz for long gaming periods and remain dead silent doing so. With my GTX 980 Ti, I had to set my fan speed to around 60% to hit 1.5 Ghz with the memory at 8 Ghz, and still end up with less performance than my GTX 1080 which uses a lot less power and generates nowhere near as much noise. The improved DX12 performance, HEVC decoding etcetera are just icing on the cake.

Despite all this though, I still plan on buying a 1080 Ti next year and selling my GTX 1080.. But I never regretted buying it, as it's one helluva GPU..

 

[moonbogg]

What is sad about Battlefield 1 performance is how much it highlights what a miserable upgrade the 1080/1070 are for 980ti users. 1070 is at best equal to 980ti at stock and will be slower after overclocking with 980ti's 25-30% overclock headroom. 1080 manages about 20% over a 980ti, which will shrink to 10-15% again with GM200's superior overclocking compared to GP104.

Nvidia really needs to put out their 1080ti ASAP and provide a real upgrade for 980ti users that are sitting on their cards not wanting to spend $1200 for the real GM200 upgrade. The 1080 looks like a mild refresh compared to 980ti in this game.

Fully agree. Compared to an OC'd 980ti, its perfectly clear just exactly how mid range the 1070/80 really are. The Titan X is much better, and the 1080ti should be a little less powerful than the Titan. I'd like even more performance than a Titan X personally for me to upgrade to it. I strongly feel that this entire generation from top to bottom is basically a Maxwell refresh. I've never had a problem skipping a refresh cycle personally.
They used to admit they were refreshes. I had dual 7800GTX's and when the 7900's came out, everyone knew they were refreshes. They were more powerful, sure, but easily could be skipped if you had last gen's high end. I want a jump like my last one from 670's to 980ti's. That was just titanic. That would be quite a long wait though I think. Probably close to 2019 actually.

 

[Bacon1]

I mean games ,which only has DX12 option to run always tend be better on nvidia. Games ,which has DX11 and DX12 options tends be good on AMD ,however, still cannot beat nvidia on DX11 with its counter parts in the same game. For example, total warhammer, QB , Battlefield 1 and Deus EX mankind divided.

Except I just showed that AMD is doing fine in this titles, and even ahead of the 1060 while you were claiming the opposite...

Total War: AMD seems to be faster: http://www.guru3d.com/articles_pages/gigabyte_radeon_rx_480_g1_gaming_review,12.html

QB is a broken mess and they removed DX12 and haven't patched the windows store version in months.

BF1 shows AMD doing better in DX12 on some sites and others say there are issues, we'll have to wait for more testing.

DXMD hasn't been tested by any sites afaik since its final release of DX12, just its initial one. Its had 3 patches since then.

Feel free to provide any updated benchmarks to support any of your claims.

 

[Unreal123]

Except I just showed that AMD is doing fine in this titles, and even ahead of the 1060 while you were claiming the opposite...

Total War: AMD seems to be faster: http://www.guru3d.com/articles_pages/gigabyte_radeon_rx_480_g1_gaming_review,12.html

QB is a broken mess and they removed DX12 and haven't patched the windows store version in months.

BF1 shows AMD doing better in DX12 on some sites and others say there are issues, we'll have to wait for more testing.

DXMD hasn't been tested by any sites afaik since its final release of DX12, just its initial one. Its had 3 patches since then.

Feel free to provide any updated benchmarks to support any of your claims.

Here is the best example of DX12 Warhammer benchmark.

It is done on max setting and a GTX 1060 on DX12 is beating Fury X.

https://forums.overclockers.co.uk/showthread.php?t=18750783

User benchmark >>>>>>>>>comprised wanna be fair benchmark from 3rd party.

 

[Erenhardt]

Here is the best example of DX12 Warhammer benchmark.

It is done on max setting and a GTX 1060 on DX12 is beating Fury X.

https://forums.overclockers.co.uk/showthread.php?t=18750783

User benchmark >>>>>>>>>comprised wanna be fair benchmark from 3rd party.

LOL
Wanna see my numbers then? AMD is over 9000 while nv is between 0 and negative 1.
As this is user benchmark, I guess it is valid for conclusions.

 

[Unreal123]

LOL
Wanna see my numbers then? AMD is over 9000 while nv is between 0 and negative 1.
As this is user benchmark, I guess it is valid for conclusions.

Some poster only see the number while do not look at the settings. Look at the settings and compare it with other sites because 3rd party sites, which do not use MSAA and only use AMD tech MLAA in the benchmark in their reiview.

 

[showb1z]

Some poster only see the number while do not look at the settings. Look at the settings and compare it with other sites because 3rd party sites, which do not use MSAA and only use AMD tech MLAA in the benchmark in their reiview.

The post you responded to had Warhammer tested with FXAA, which is Nvidia tech AFAIK.
Fantastic you provided 8x MSAA benches though, so relevant and useful when comparing cards in the BF1 thread!

 

[Unreal123]

The post you responded to had Warhammer tested with FXAA, which is Nvidia tech AFAIK.
Fantastic you provided 8x MSAA benches though, so relevant and useful when comparing cards in the BF1 thread!

Your logic explains why AMD is just a mid range company with no vision. People buy high end card not to demote settings or reduce settings, therefore, they buy high end card to run game at max settings ,which why you see people CF or SLI cong.

 

[showb1z]

So what you're saying is Nvidia has no vision because they develop inferior tech like FXAA. Ok got it.
Seriously, sober up, you're making no sense and derailing the thread.

 

[PhonakV30]

I have no idea what @Unreal123 said.he better leaves this thread alone.His Posts about NV/AMD are not useful.


attacking or threatening other posters is not allowed.
Markfw900

 

[Piroko]

Some new tests by a less known site:
https://uk.hardware.info/reviews/70...testresultaten-full-hd-1920x1080-+-frametimes

Test was done in single player.
Nvidia has an edge in 1080p but loses it in 1440p and UHD. As for DX12, seems like performance only improves when you're CPU-bound.


Also, PCGH multiplayer and DX12 tests should be added within the next hours.

 

[Unreal123]

I have no idea what @Unreal123 said.he better leaves this thread alone.His Posts about NV/AMD are not useful.

My post will only make sense to you when i post benchmark where AMD is ahead. Only that is useful according to you.


Trolling is not allowed
Markfw900

 

[PhonakV30]

My post will only make sense to you when i post benchmark where AMD is ahead. Only that is useful according to you.

Like this nonsense? Really seeking alpha ?

So in summary of DX12 that the native DX12 games like Gears of War Ultimate, Forza Horizon 3, Forza Apex, Gears of War 4 GTX 1060 is on par with Fury leave alone RX 480.

http://www.pcgameshardware.de/Gears.../Specials/DirectX-12-Benchmarks-Test-1208296/
http://www.pcgameshardware.de/Gears-of-War-4-Spiel-55621/Specials/Performance-Test-Review-1209651/
http://www.pcgameshardware.de/Forza.../Specials/Benchmarks-Test-DirextX-12-1208835/

AMD only performances better on DX12 ,which involves gaming evolve ,however still do not manage to beat nvidia DX11 performance even with AMD uses DX12+ Async.

http://www.pcgamer.com/total-war-wa...md-still-cant-match-nvidias-dx11-performance/
https://forums.overclockers.co.uk/showthread.php?t=18750783&highlight=Warhammer

Nvidia is faster in this game on DX11 even faster then DX12+ Async. AMD will be minority forever until AMD community start posting logic. As seeking alpha and wall street journalist said that AMD biggest enemy is logic, fact and reality.

 

[Glo.]

To everyone showing any form of brand favourism in any gaming thread:

You have to be incredibly stupid to not see that games are optimized for specific architectures, and one games prefer AMD and others Nvidia. Arguing about one brand being better from the other shows rather your level of intellect, rather than anything more.

 

[pcslookout]

Most review sites are too chicken to do that. They are so terrified of not being able to maintain perfect repeatability that they totally sacrifice any kind of useful data in these multiplayer games by doing CPU tests in single player. But still, avoiding GPU bottlenecks is critical if you want to see relative CPU performance. Of course, most sites get this part wrong as well.

So what do you think the minimum fps would be with a GTX 1080 if they did do these benchmarks in 64 slot full multiplayer BF1 server ?

I wish they would.

 

[stahlhart]

Unreal123 and behrouz, both of you: out of this thread now, or it's getting locked and both of you are getting time off.
-- stahlhart

 

[Head1985]

What is sad about Battlefield 1 performance is how much it highlights what a miserable upgrade the 1080/1070 are for 980ti users. 1070 is at best equal to 980ti at stock and will be slower after overclocking with 980ti's 25-30% overclock headroom. 1080 manages about 20% over a 980ti, which will shrink to 10-15% again with GM200's superior overclocking compared to GP104.

Nvidia really needs to put out their 1080ti ASAP and provide a real upgrade for 980ti users that are sitting on their cards not wanting to spend $1200 for the real GM200 upgrade. The 1080 looks like a mild refresh compared to 980ti in this game.

Yeah pascal is way worse than kepler was.GTX670/680 were true upgrade even from GTX580.GTX580 didnt have much oc headroom and after both cards OC GTX670 was still 20-30% faster than GTX580.
Best strategy is just ignore these Gx104 SKu and buy only cutdown BIG SKU.NV cutdown BIG sku very very little(980TI vs TITANX only 2x SMX disabled) and they also have HUGE oc headroom unlike Gx104 parts.

 

[frowertr]

I don't understand why there is so much bias for one brand or another. I love NV but it doesn't hurt my feelings to see AMD doing well either. We need the competition in order to keep prices as low as possible.

People take this brand bashing way too seriously.