(gamegpu) Battlefield 1 Benchmarks

Final8y · 18 Oct 2016 at 19:58

southernorth said:
The 1060 when using DX11 is faster than the 480 using DX11 or DX12. It has the highest fps (overall) of any of these 4 tests. What else is there to say? the 1060 wins in this respect.

Also, my mistake for saying that the 1060 won in DX12.

And I do agree that these results are somewhat odd, but I can only go by what they say.

Its not faster in DX12 its 2 FPS slower at 1080P and 4fps slower in 1440, and 2 fps slower at 4K, its clear you don't know how to read the graph.

tyler_jrb · 18 Oct 2016 at 20:02

TheRealDeal said:
http://www.guru3d.com/index.php?ct=articles&action=file&id=26226

For the gtx980ti results can i get a

G-I-M-P-E-D

Unless they are all running it at 1ghz. Its likely

Last time I checked one of my Ti's was about 20-25% max slower than a 1080. Even max oc vs max oc.

TheRealDeal · 18 Oct 2016 at 20:04

tyler_jrb said:
Unless they are all running it at 1ghz. Its likely

If that really is the real gtx980ti performance then it's a definite Gimp. I think there is something else going on here though as the other Maxwell cards all seem to perform decent.

They do say in the conclusion that they are going to retest the Gtx980ti as the results are weird.

southernorth · 18 Oct 2016 at 20:49

Final8y said:
Its not faster in DX12 its 2 FPS slower at 1080P and 4fps slower in 1440, and 2 fps slower at 4K, its clear you don't know how to read the graph.

I'm talking about 1080p.....

And the fastest result at 1080p (overall between the 1060 and 480 across dx11/12) is the 1060 using dx11. do you need some more help? Clearly reading isn't one of your strengths. Why would dx12 even be talked about when it offers lesser performance than the 1060 dx11.

1060
dx11 - 95
dx12 - 87

480
dx11 - 94
dx12 - 89

Now, I said "The 1060 when using DX11 is faster than the 480 using DX11 or DX12. It has the highest fps (overall) of any of these 4 tests"

The 1060 at DX11 got 95 fps. This beats all other results.

Please do explain how you disagree with that, and how 95 is somehow not the largest of the 4 numbers above. I'm talking about overall performance, not dx12 v dx12 as in this case dx11 beats both (for the 1060).

Final8y · 18 Oct 2016 at 21:12

southernorth said:
I'm talking about 1080p.....

And the fastest result at 1080p (overall between the 1060 and 480 across dx11/12) is the 1060 using dx11. do you need some more help? Clearly reading isn't one of your strengths. Why would dx12 even be talked about when it offers lesser performance than the 1060 dx11.

1060
dx11 - 95
dx12 - 87

480
dx11 - 94
dx12 - 89

Now, I said "The 1060 when using DX11 is faster than the 480 using DX11 or DX12. It has the highest fps (overall) of any of these 4 tests"

The 1060 at DX11 got 95 fps. This beats all other results.

Please do explain how you disagree with that, and how 95 is somehow not the largest of the 4 numbers above. I'm talking about overall performance, not dx12 v dx12 as in this case dx11 beats both (for the 1060).

Sorry but 1FPS, your whole argument is about 1 fps just at 1080p, which is within margin of error :rolleyes:

.
On the ignore list you go.

BS Dave · 18 Oct 2016 at 23:31

I wonder if DICE will patch in some fancy dx12 mgpu mode - tiled rendering or something. Even if it's less fps for nvidia than dx11 it might be worth it for the drop in latency over alternate frame rendering...

abui1806 · 18 Oct 2016 at 23:33

I am in the market for a new graphics card. ATM I am looking for something small, ITX sized cards. Question is should I buy the Gigabyte GTX 1070 ITX now for this game or wait until Vega comes out? I am using SLI GTX 970's and wanting to get rid of them.

humbug · 18 Oct 2016 at 23:41

southernorth said:
http://www.guru3d.com/articles_pages/battlefield_1_pc_graphics_benchmark_review,6.html

Yet more benchmarks

Shows 1060 winning in dx11 and dx12 (1080p ultra). Then quite even at 1440p and 4k, with 480 winning 1440p.

The R9 390 - GTX 980 - RX 480 - GTX 1060 and GTX 980TI are all within a 10% margin, the R9 390X is faster than all of them....... the Fury-X is also holding its own very well indeed beaten only by the 1080 and Pascal TX..

Yet again all Nvidia's last generation cards seem to be lacking any sort attention in new titles, you have a GTX 970 not far off 90% as fast as a 980TI.

Well done Nvidia. my card in the legacy bin already is it?

tommybhoy · 18 Oct 2016 at 23:49

Aye, Maxwell's pumped mate.

~~Long~~ live Pascal!

Battlefield 1 is a treat to the eyes and a feast to play on both AMD and NVIDIA hardware. It is a game well-done.

To do:

Add more graphics cards (!)
Retest GTX 980 Ti (current result set is odd)
CPU scaling in-between Intel <> AMD versus DirectX 12

Doom112 · 19 Oct 2016 at 03:09

" Correct - I already made note of that. It might have been a driver issue or failed optimization. The 980 Ti is on the to-do list for a retest tomorrow, I am out of activations/swaps for today. "

http://forums.guru3d.com/showthread.php?t=410451

Too much double standard and some people do not even read or look. He has some problem testing GTX 980 Ti. There is only 5% difference GTX 980 to GTX 980 Ti in his testing.

Doom112 · 19 Oct 2016 at 03:15

humbug said:
The R9 390 - GTX 980 - RX 480 - GTX 1060 and GTX 980TI are all within a 10% margin, the R9 390X is faster than all of them....... the Fury-X is also holding its own very well indeed beaten only by the 1080 and Pascal TX..

Yet again all Nvidia's last generation cards seem to be lacking any sort attention in new titles, you have a GTX 970 not far off 90% as fast as a 980TI.

Well done Nvidia. my card in the legacy bin already is it?

Yes a legacy product beating god like hardware Fury X.

http://www.pcgameshardware.de/Battl...attlefield-1-Technik-Test-Benchmarks-1210394/

http://www.overclock3d.net/reviews/software/battlefield_1_pc_performance_review/9

http://www.purepc.pl/karty_graficzn...tlefield_1_pc_wymagania_pod_kontrola?page=0,8

http://www.purepc.pl/karty_graficzn...tlefield_1_pc_wymagania_pod_kontrola?page=0,7

on DX12 GTX 980 TI give 85 fps and on DX11 it is giving 95 fps while fury X has no difference according to PurePc.

humbug · 19 Oct 2016 at 03:34

Doom112 said:
Yes a legacy product beating god like hardware Fury X.

http://www.pcgameshardware.de/Battl...attlefield-1-Technik-Test-Benchmarks-1210394/

http://www.overclock3d.net/reviews/software/battlefield_1_pc_performance_review/9

http://www.purepc.pl/karty_graficzn...tlefield_1_pc_wymagania_pod_kontrola?page=0,8

http://www.purepc.pl/karty_graficzn...tlefield_1_pc_wymagania_pod_kontrola?page=0,7

on DX12 GTX 980 TI give 85 fps and on DX11 it is giving 95 fps while fury X has no difference according to PurePc.

So Nvidia have issues with Maxwell DX12 performance, that was the point i'm making.

The only one among those that doesn't look so bad is the GamesHardware.de one, or at least it would be if it was not already common knowledge they are using a well and truly squeed 30% factory overclocked 980TI, so take that off those results and its pretty bad again.

southernorth · 19 Oct 2016 at 03:39

Final8y said:
Sorry but 1FPS, your whole argument is about 1 fps just at 1080p, which is within margin of error .
On the ignore list you go.

Ah brilliant, you realise you were wrong then ignore me haha

I'm sure if AMD was leading by 1fps margin of error wouldn't be mentioned

Doom112 · 19 Oct 2016 at 04:17

humbug said:
So Nvidia have issues with Maxwell DX12 performance, that was the point i'm making.

The only one among those that doesn't look so bad is the GamesHardware.de one, or at least it would be if it was not already common knowledge they are using a well and truly squeed 30% factory overclocked 980TI, so take that off those results and its pretty bad again.

That is the thing OC is also a feature.If Fury X is not a overclockers dream then it is not Nvidia fault.

Doom112 · 19 Oct 2016 at 04:25

southernorth said:
Ah brilliant, you realise you were wrong then ignore me haha

I'm sure if AMD was leading by 1fps margin of error wouldn't be mentioned

Some people only like DX12 for sake of AMD not because it has useless features up till now but only because it has manage to decrease Nvidia performance. In this history of API DX12 is the most useless API ever created , with no benefit to 80% of PC users, Day 1 annoying performance, manage to provide the most broken games this year, and even AMD barely has 1 to 2 fps difference on DX12.

SiDeards73 · 19 Oct 2016 at 06:33

Doom112 said:
Some people only like DX12 for sake of AMD not because it has useless features up till now but only because it has manage to decrease Nvidia performance. In this history of API DX12 is the most useless API ever created , with no benefit to 80% of PC users, Day 1 annoying performance, manage to provide the most broken games this year, and even AMD barely has 1 to 2 fps difference on DX12.

Oh god how many times do we have to say it, it's not all about high fps numbers? How is having 100fps useful if your monitor is only 60hz? How is having 200fps useful if your adaptive sync range is 40-144hz?

What IS useful, and this is the part your struggling with, is DX12 brings up the minimums so it lessens out the variance and spikes you get.

If im getting 100fps with dips to 20fps using dx11 on a 40-144hz screen, but Dx12 gives me 90fps with dips to 40fps which should I use?

Come on Sherlock even you should be able to work that one out

Clue, it's not just about high fps numbers it's about a smoother experience

Final8y · 19 Oct 2016 at 06:51

SiDeards73 said:
Oh god how many times do we have to say it, it's not all about high fps numbers? How is having 100fps useful if your monitor is only 60hz? How is having 200fps useful if your adaptive sync range is 40-144hz?

What IS useful, and this is the part your struggling with, is DX12 brings up the minimums so it lessens out the variance and spikes you get.

If im getting 100fps with dips to 20fps using dx11 on a 40-144hz screen, but Dx12 gives me 90fps with dips to 40fps which should I use?

Come on Sherlock even you should be able to work that one out

Clue, it's not just about high fps numbers it's about a smoother experience

+1

Greebo · 19 Oct 2016 at 08:41

SiDeards73 said:
Oh god how many times do we have to say it, it's not all about high fps numbers? How is having 100fps useful if your monitor is only 60hz? How is having 200fps useful if your adaptive sync range is 40-144hz?

What IS useful, and this is the part your struggling with, is DX12 brings up the minimums so it lessens out the variance and spikes you get.

If im getting 100fps with dips to 20fps using dx11 on a 40-144hz screen, but Dx12 gives me 90fps with dips to 40fps which should I use?

Come on Sherlock even you should be able to work that one out

Clue, it's not just about high fps numbers it's about a smoother experience

+2

I think too many people get hung up on maximum framerates and even average.

I know I would much rather play a game with a card which gave a min of 40 and max of 50 fps with an average of 45 rather than another card which technically is 10% faster with an av of 50 fps but had max fps of 90 and min of 10.

Nexus18 · 19 Oct 2016 at 10:27

snips86x · 19 Oct 2016 at 10:32

+4

So long as there is an expectation that the frames are enough to keep the experience smooth when using the likes of gsync, I'm no bothered. So far, it's looking promising.