Call of Duty WWII: Benchmark Performance Analysis

[W1zzard]

Call of Duty: WWII takes you back to a time without fancy fighting tech; you had to rely on your rifle and team mates. Sledgehammer's engine received optimizations in both graphics quality and performance, and there are plenty of settings to play with. We tested at up to 4K resolution, using 15 graphics cards, including the brand-new GTX 1070 Ti.

Show full review

 

[xkm1948]

Excellent summary, thx for the fast work!

Would you get a performance analysis on Wolfenstein II as well?

 

[Lightofhonor]

Always interesting to see how performance trends as resolution scales. Like Nvidia cards have loads of torque but the higher HP of the AMD cards help them catch up in the high resolution game.

Which does nothing if you are playing on a 1/4 mile track aka 1080p

 

[Dimi]

Thanks for the analysis but i have to wonder, what is the point of putting any motion blur on? Thats the first thing i turn off in ANY game including 99% of my friends.

 

[xkm1948]

Thanks for the analysis but i have to wonder, what is the point of putting any motion blur on? Thats the first thing i turn off in ANY game including 99% of my friends.

I like it. I think it helps with motion sickness a bit. I am super sensitive to motion sickness and I usually have to leave the blur on to avoid sickness issues.

 

[W1zzard]

Thanks for the analysis but i have to wonder, what is the point of putting any motion blur on? Thats the first thing i turn off in ANY game including 99% of my friends.

Agree with you completely, I do the same. "But cinematic" ...

Updated the settings page to mention you can turn off DOF and motion blur.

 

[Dimi]

Nice thank you.

 

[TheMailMan78]

This game includes such realistic elements as black female nazis. Such realism. Much progress.

 

[Mescalamba]

STEAM reviews dont look exactly positive.

 

[jesdals]

I found that the beta lacked support for Eyefinity support, is this still the case?

 

[dirtyferret]

Nice review, any chance of CPU results to be included in future tests?

 

[Melvis]

Does this game have SLi support? and if so is it any good with scaling?

 

[bug]

I like it. I think it helps with motion sickness a bit. I am super sensitive to motion sickness and I usually have to leave the blur on to avoid sickness issues.

I think it's just a "because we can" feature. Otherwise the option to turn it off wouldn't be there.
At the same time I wonder if the irritation/motion sickness isn't because motion blur is implemented poorly. I mean, I don't seem to mind it much in movies.

 

[EarthDog]

Nice review, any chance of CPU results to be included in future tests?

+1.

Would love to see this... or at least a cpu test across all the games you test at multiple resolutions woth multiple level cards..

 

[bug]

+1.

Would love to see this... or at least a cpu test across all the games you test at multiple resolutions woth multiple level cards..

Frankly, these tests will look pretty much like any others in the past 10+ years: some cores will be used, some not and some will be maxed out in titles that don't know how to offload.
According to Steam, 2 cores had almost 50% of the market in H1'17. Now quad cores have almost 70%. So games will target that for the foreseeable future.

I mean, sure, I'd like to see how each and every title scales, but realistically, it will get repetitive fast and it will probably just be a waste of reviewer's time. It may still be worth it to highlight the odd title (either way) here and there.

 

[EarthDog]

You post a lot here... its a frequent question and debate. Why not have a modern reference? I think many people would be surprised at the results.

 

[bug]

You post a lot here... its a frequent question and debate. Why not have a modern reference? I think many people would be surprised at the results.

I think this is only a debate for the uninformed that think once they got a bazillion cores, they will magically be put to work.
Games industry doesn't care about that. They need to target the least common denominator because their only benchmark is number of sales (if possible in the opening week-end).

Like I said, I'm not opposed to benchmarking CPU usage, I'm just predicting it will not reveal anything exciting. That's all.

 

[EarthDog]

I think you hit the nail on the head...

I think this is only a debate for the uninformed

The point is to inform and see what a difference a CPU makes at times, and how it makes no difference other times. It will vary wildly from title and settings to the GPU used. I would like to see:

2500K/2600K/4670K/4770K/7600K/7700K/8700K/7820X/7900X and AMD offerings of the same group (from a BD quad to octo to Ryzen quad to octo). And across a few GPUs. Say a 1050/1070, and 1080Ti as well as RX460, Vega 56, and Vega 64.

 

[bug]

I think you hit the nail on the head...


The point is to inform and see what a difference a CPU makes at times, and how it makes no difference other times. It will vary wildly from title and settings to the GPU used. I would like to see:

2500K/2600K/4670K/4770K/7600K/7700K/8700K/7820X/7900X and AMD offerings of the same group (from a BD quad to octo to Ryzen quad to octo). And across a few GPUs. Say a 1050/1070, and 1080Ti as well as RX460, Vega 56, and Vega 64.

So basically a measly 18 CPUs x 6 GPUs for each title. That can't take long to test. And then you'll want SLI/Crossfire

(Just to be clear, your request is technically legit, it's just the RoI that I'm sceptical about).

 

[EarthDog]

SLI and Crossfire also offer different results indeed and rely MUCH MORE on the CPU used for best scaling. People will be surprised how much a modern CPU performs better in many titles with mid-range+ GPUs... most have their head in the sand and continue to suckle the teet of their X58 platform and GTX 1080+ without knowing there is a difference.

Heh, yeah, LOE v ROI, I hear you... but it's incredibly common question which needs a good answer. We don't have that in one place.

 

[bug]

@W1zzard you have the stage

 

[W1zzard]

Impossible to do CPU tests with that many GPU combinations on top

 

[EarthDog]

Impossible? No. Time consuming.. oh hells yeah.

 

[bug]

Impossible to do CPU tests with that many GPU combinations on top

You don't have to use that many GPUs. I'm sure @EarthDog wouldn't mind if you added a few to those he listed

On a more serious note, if you're thinking about doing this, I think should decide how much time you're willing to allocate to the task and then, starting from @EarthDog 's list, start slashing "intermediary" combos till everything fits in your time budget.

 

[dirtyferret]

Impossible to do CPU tests with that many GPU combinations on top

Techspot uses one GPU for several CPUs. I understand it's time consuming but could be interesting if you just used an AMD 1600 and Intel 8700 and simply shut down some cores

https://www.techspot.com/review/1333-for-honor-benchmarks/page3.html