Gears of War 4 Benchmark

[Unreal123]

http://www.pcworld.com/article/3128346/software-games/tested-gears-of-wars-4-pc-benchmarks-yield-glorious-graphics-options-galore.html

http://wccftech.com/gears-war-4-performance-review/

In short, Nvidia is taking a big lead against AMD's every counter part for Maxwell or Pascal mid range. However, Nvidia is still yet to release a game ready driver for Gears of War 4.

 

[Riek]

From pcworld
We tested the game using a prerelease GeForce 373.02 driver, which are Game Ready for Gears of War 4. On AMD’s side, we used the most current Radeon driver, 16.9.2. As this article was being prepared for publication, AMD pinged me to say that new Radeon 16.10.1 drivers will be released “very shortly” with optimizations for Gears of War 4,

Difference is rather small in pcworld. On Ultra 1060 Founder is 3% faster then 480 reference.

From wccftech:
Drivers used

Radeon cards used Crimson 16.9.2
GeForce cards used Geforce 373.02

In short, seems like they used a pre-release rdy drivers from nvidia, and not game ready drivers from AMD.
Just to clarify, since you make it seem as if AMD had a driver advantage here.

 

[PowerK]

Performance guide is up at GeForce.com
http://www.geforce.com/whats-new/guides/gears-of-war-4-graphics-and-performance-guide
Game ready driver should be available sometime today according to the article.

 

[Unreal123]

Another benchmark
http://www.gamersnexus.net/game-bench/2628-gears-of-war-4-gpu-benchmark-on-pc

 

[antihelten]

In short, seems like they used a pre-release rdy drivers from nvidia, and not game ready drivers from AMD.
Just to clarify, since you make it seem as if AMD had a driver advantage here.

From the Gamersnexus review:

The forthcoming 16.10.1 update contains the same GOW4 game optimizations as 16.9.2. Performance has not changed between driver packages.

...

Note that 16.9.2 and 16.10.1 drivers both exhibit the same performance in Gears of War 4, which we've validated directly with AMD.

They didn't actually test the 16.10.1 drivers though (or at least they didn't post any numbers for 16.10.1), and it kind of contradicts what PCWorld was told by AMD, so this should probably be taken with a grain of salt

 

[ShintaiDK]

 

[Thinker_145]

So many conflicting benchmarks wtf?

Sent from my HTC One M9

 

[Carfax83]

So many conflicting benchmarks wtf?

Sent from my HTC One M9

Conflicting benchmarks? What are you on about? The benchmarks so far are fairly uniform if anything.. Gears of War 4 is an exceptionally optimized DX12 title. It even makes use of tiled resources, a highly useful feature that came out with DX11.2 but that no game developer has used to my knowledge.

 

[Thinker_145]

Conflicting benchmarks? What are you on about? The benchmarks so far are fairly uniform if anything.. Gears of War 4 is an exceptionally optimized DX12 title. It even makes use of tiled resources, a highly useful feature that came out with DX11.2 but that no game developer has used to my knowledge.

Did you look at the post right above? That does not go in line with anything we had seen before.

Sent from my HTC One M9

 

[Erenhardt]

Quite good performance. Wonder how is the CPU side of things.

 

[PhonakV30]

It's Nvidia's baby , nothing.No surprise that why RX480 performs badly compare to R9 390x, GTX 1060.about Pc world's review , dunno how RX 480 is equal to GTX 1060.

 

[Krteq]

Gears of War 4 is an exceptionally optimized DX12 title.

I already asked you in another thread but no answer from you.

EU4 is DX12 compatible now? Last time I checked, it still had DX11 resource management and no multi-engine support implemented.

So, how can this be "a true, exceptionally optimized DX12 title"? It's still DX11.2/11.3 engine/game with DX11 resource management and no multi-engine support.

 

[AtenRa]

Gears of War 4 is an exceptionally optimized DX12 title.

This seems DX-12 Optimized to you ??
GTX 970 faster than Fury X ?? GTX 960 faster than RX 470 ?? GTX 950 ~50% faster than RX 460 ???

edit: GTX 1080 = GTX 980Ti ??

 

[Piroko]

Something's off with the results from GamersNexus, they get much higher benchmark scores than even Nvidia themselves:

Their 970 scores are higher than a 980Ti and a 1070 in Nvidias benchmark run lol

 

[Sweepr]

Ouch, terrible minimum FPS from Fury X. I wonder if it's another case of VRAM limitation, looks like it won't age well.

Stepping up your game to the quickly becoming standard 1440p and it starts to get pretty rough out there. We start to see things get a fair bit more difficult to push resulting in a pretty big hit with only our GTX 980ti able to maintain that greater than 60FPS 5% minimums. Interestingly enough the VRAM usage didn’t go up by as much as expected with the GTX 980ti only up 300MB to 4.97GB of usage while the R9 Fury X remained at the 3.62GB wall.

GTX 1060 ahead of RX 480 in this DX12 title @ GamersNexus and WCCFTech. Bodes well for NVIDIA in future Microsoft titles as well.

 

[AtenRa]

I already asked you in another thread but no answer from you. So, how can this be "a true, exceptionally optimized DX12 title"? It's still DX11.2/11.3 engine/game with DX11 resource management and no multi-engine support.

Is this like the EU3 with Mantle support in Sniper Elite III Thief ??

 

[AtenRa]

Ouch, terrible minimum FPS from Fury X. I wonder if it's another case of VRAM limitation, looks like it won't age well.

Yeap, this is why GTX 970 with only 3.5GB + 0.5 is faster

 

[Sweepr]

Yeap, this is why GTX 970 with only 3.5GB + 0.5 is faster

Except that quote comes from WCCFTech (where it's faster than a GTX 1060). Let's see if ComputerBase will test the game at Ultra/Insane settings or limit if to Medium settings not to evidence Fury's limitations. Certain settings definitely use >4GB VRAM.

 

[Sweepr]

GameGPU Results Out!

Edit: They didn't use NVIDIA release drivers for the game.

Nvidia GeForce / ION Driver Release 372.90

AMD Radeon Crimson Edition 16.10.1

http://gamegpu.com/action-/-fps-/-tps/gears-of-war-4-test-gpu

 

[Spjut]

Benchmarks up at GameGPU. Compared to others, they've tested Kepler GPUs as well.
http://gamegpu.com/action-/-fps-/-tps/gears-of-war-4-test-gpu

Also CPU benchmarks, good show for the FX 8350 and i7 vs i5

 

[ShintaiDK]

So another game that bust both the DX12 and Kepler driver myth.

 

[AtenRa]

It seems GamersNexus need to retest with 16.10.1 driver and any new Driver in general for both AMD and NVIDIA.

 

[Carfax83]

I already asked you in another thread but no answer from you. So, how can this be "a true, exceptionally optimized DX12 title"? It's still DX11.2/11.3 engine/game with DX11 resource management and no multi-engine support.

It supports asynchronous compute, so how could you say it has no multi engine support? Also, according to a tech interview, UE4 has now transitioned fully to DX12:

Digital Foundry: Is the game built as a DX12-only title (with certain low-level Xbox additions console-side)?

Mike Rayner: Gears of War 4 is a DirectX 12 title on both Xbox One and Windows 10. Working with Microsoft's Silicon, Graphics and Media team and Epic, we have transitioned Unreal Engine fully to DX12. DirectX 12 has allowed us to increase performance by giving us more direct control over the hardware, simplifying the driver layer, and allowing us to make fully informed and optimal decisions on how to manage graphics resources.

Source

 

[Carfax83]

This seems DX-12 Optimized to you ??
GTX 970 faster than Fury X ?? GTX 960 faster than RX 470 ?? GTX 950 ~50% faster than RX 460 ???

edit: GTX 1080 = GTX 980Ti ??

No idea, but GameGPU and PCgameshardware.de are showing much better performance for the Radeons, so I guess it depends on who you want to believe.

And at 1080p, the GTX 1080 seems to be CPU limited in the benchmark hence the similar performance. At 1440p, it's a different story. For the 0.1% low, the GTX 1080 is 34% faster than the GTX 980 Ti.

 

[dark zero]

GameGPU Results Out!

http://gamegpu.com/action-/-fps-/-tps/gears-of-war-4-test-gpu

That is better. Fury X at 980 Ti levels and the rest of the cards in a comparable range.