Final Fantasy XV Benchmark released for Windows

[CAT-THE-FIFTH]

I got 7928(1080p and high) and then 3507(4k and high) with my 1800x@ stock and gtx 1080.

10% faster than me!!

[mtyson]

interesting video from developer saying FFXV, built from ground up for PCs, is a generation ahead of the console versions

[CAT-THE-FIFTH]

interesting video from developer saying FFXV, built from ground up for PCs, is a generation ahead of the console versions

It seems to like using tessellation for a lot of effects apparently.

Also,GN has called the benchmark out:

https://www.gamersnexus.net/game-ben...ameworks-tests

Although we don't believe this to be intentional, the Final Fantasy XV benchmark is among the most misleading we’ve encountered in recent history. This is likely a result of restrictive development timelines and a resistance to delaying product launch and, ultimately, that developers see this as "just" a benchmark. That said, the benchmark is what's used for folks to get an early idea of how their graphics cards will perform in the game. From what we've seen, that's not accurate to reality. Not only does the benchmark lack technology shown in tech demonstrations (we hope these will be added later, like strand deformation), but it is still taking performance hits for graphics settings that fail to materialize as visual fidelity improvements. Much of this stems from GameWorks settings, so we've been in contact with nVidia over these findings for the past few days.

As we discovered after hours of testing the utility, the FFXV benchmark is disingenuous in its execution, rendering load-intensive objects outside the camera frustum and resulting in a lower reported performance metric. We accessed the hexadecimal graphics settings for manual GameWorks setting tuning, made easier by exposing .INI files via a DLL, then later entered noclip mode to dig into some performance anomalies. On our own, we’d discovered that HairWorks toggling (on/off) had performance impact in areas where no hair existed. The only reason this would happen, aside from anomalous bugs or improper use of HairWorks (also likely, and not mutually exclusive), would be if the single hair-endowed creature in the benchmark were drawn at all times.

The benchmark is rendering creatures that use HairWorks even when they’re miles away from the character and the camera. Again, this was made evident while running benchmarks in a zone with no hairworks whatsoever – zero, none – at which point we realized, by accessing the game’s settings files, that disabling HairWorks would still improve performance even when no hairworks objects were on screen. Validation is easy, too: Testing the custom graphics settings file by toggling each setting, we're able to (1) individually confirm when Flow is disabled (the fire effect changes), (2) when Turf is disabled (grass strands become textures or, potentially, particle meshes), (3) when Terrain is enabled (shows tessellation of the ground at the demo start' terrain is pushed down and deformed, while protrusions are pulled up), and (3) when HairWorks is disabled (buffalo hair becomes a planar alpha texture). We're also able to confirm, by testing the default "High," "Standard," and "Low" settings, that the game's default GameWorks configuration is set to the following (High settings):

VXAO: Off
Shadow libs: Off
Flow: On
HairWorks: On
TerrainTessellation: On
Turf: On

Benchmarking custom settings matching the above results in identical performance to the benchmark launcher window, validating that these are the stock settings. We must use the custom settings approach, as going between Medium and High offers no settings customization, and also changes multiple settings simultaneously. To isolate whether a performance change is from GameWorks versus view distance and other settings, we must individually test each GameWorks setting from a baseline configuration of "High."

Square Enix is responsible for the higher-pipeline functions of calling and rendering objects, like the buffalo, and it's not like GameWorks has the ability to reach outside of the world, grab assets, and force them into the game. We would put this squarely on Square Enix, for now, but nVidia does have some responsibility in getting the rest of their GameWorks features fully integrated and working (even if Square is ultimately coding it, nVidia has its name on the product).

What we do know is that the benchmark is unreliable, and we suggest not using it. Wait for our final launch benchmarks. This will take more time to research, but we will keep our eye on it and are in touch with the teams responsible for the game's graphics.

If you need more visual examples, check the accompanying video for this piece.

Its a very long article,but worth a read.

Edit!!

Here's what's happening: The cows -- buffalo, whatever -- are being rendered at all times during the benchmark. This includes being rendered before trigger events, after trigger events, and even upon conclusion of the benchmark. The best example might be the campsite, when we've had a clear time-of-day change and location change: In the blackest of nights -- and this was hard to find -- the buffalo still roam, miles away and in the pitch black dark of a completed benchmark, still requiring system resources. Square Enix demonstrates functioning occlusion for other objects (rocks, for one), has selectively removed objects until the player meets a trigger (the boathouse), and even has LOD lines drawn. All of this indicates that Square is fully capable of culling unseen objects, or minimizing the LOD, but has failed to do so with the particularly resource-intensive buffalo, one of the only (the only?) objects that use HairWorks in the entire benchmark sequence.

This isn’t an accidental cow. That doesn’t mean it’s a malicious cow, but it seems to be an oversight. That's not the only issue in the whole benchmark, though; GameWorks settings, as we display in the video, rarely show their merits in the benchmark. Turf is among the only visible effects, yet the framerate does fluctuate heavily. We anticipate a bigger change as the game ships. What this means for now, though, is that the benchmark is fully unreliable and unrealistic, failing to represent what the final game will hopefully be. And if it does represent the final game, well, that's another issue entirely.

We found an LOD line drawn along the map, where you can see the transition of detail – so Square Enix has marked this space as being clearly not visible to the player, and yet still – way beyond the boundaries of this line, miles away – we see the cows in full detail. There are also examples of culling present in the game, there are examples of selective removal of unseen objects, there are examples of LOD scaling, and instances of triggered object spawning and despawning.

LMAO.

Looks like the real game should be easier to run I suspect!

Please,pretty,please don't do this to Metro:Exodus,and make it better optimised!

[shaithis]

The full game will be easier to run or the full game will still render everything regardless of whether its in the scene or not?

This is Square Enix, the company that seems to have forgotten the gamers that have purchased millions of titles from them and have turned most of their resources to mobile gaming.....I am half expecting this to be a technical mess.

[CAT-THE-FIFTH]

The full game will be easier to run or the full game will still render everything regardless of whether its in the scene or not?

This is Square Enix, the company that seems to have forgotten the gamers that have purchased millions of titles from them and have turned most of their resources to mobile gaming.....I am half expecting this to be a technical mess.

Something weird is going on - there are lines which are in the game which denotes culling,but a lot of stuff is not culled,and the Hairworks cows are not even in the frame,yet is the biggest drop in performance for any companies cards.

In the video,Steve from GN basically says that with Hairworks on,you see no difference as you don't see the cows,yet the performance is kibbled,so he basically suggests you might as well switch off most of these settings since it only drops performance and seems to not really add much to the game.

I am actually more worried about Metro:Exodus TBH. I really hope that does not have that level of crap in it.

[outwar6010]

10% faster than me!!

I think I'm the fastest on hexus so far

[Ozaron]

In the video,Steve from GN basically says that with Hairworks on,you see no difference as you don't see the cows,yet the performance is kibbled,so he basically suggests you might as well switch off most of these settings since it only drops performance and seems to not really add much to the game.

Assuming you *can* turn them off. Customizable presets or death..

[CAT-THE-FIFTH]

I think I'm the fastest on hexus so far

A fine victory!!

Assuming you *can* turn them off. Customizable presets or death..

Well AMD has not released official drivers yet,but honestly I am getting sick and tired of poorly optimised PC games now.

Its becoming more and more apparent as the whole sub £300 area has just massively stagnated for the last 4 years and on top of this mining,and RAM cartels have screwed over the ambitions of gamers wanting a decent value rig.

[mapesdhs]

This benchmark is bogus, check the info on Gamers Nexus about Hairworks, etc.

[Iota]

Was interesting to watch the graphs on Task Manager on the second monitor, CPU looks fine but GPU maxxed right out.

No 1440p setting, will have to try the 4K slideshow

What was even more interesting when I ran the "high" setting, was my gpu was running at 100% load... but intermittently was dropping back to 300/150 instead of running at 1000/1260 (non-overclocked R9 290). Didn't see that running lite or standard. Ran perceptibly better on standard than on lite.

Also, I don't think the benchmark is 100% accurate for consistency for single runs, the fight scene isn't the same with each run.

Anyway, for an old system with zero overlock it didn't run as badly as I expected it too...

https://imgur.com/a/8Vl57

Also,GN has called the benchmark out:

https://www.gamersnexus.net/game-ben...ameworks-tests

Well, that cow.... seems like a sneaky nVidia thing to pull.... *remembers Batman....*