VO is a beautiful game practically no matter what settings you play on, but I've noticed something extremely weird every time I look at screenshots or videos of it.

Everything always comes out looking.... dull. Flat.
Lacking expression.

It's not just a matter of capturing static moments contributing to this feeling, nor does it seem to be any lighting or motion effects.

It almost seems to me as if VO is simply refusing to be defined, not allowing itself to collapse into a state that can be understood through retroactive observation.

Is this just me? Am I insane here?

Whether its videos I take or screenshots on the wiki and forums or old and new video alike across multiple platforms, VO content just seems to fall flat, visually speaking; it lacks depth and essence.

If you've noticed this, why do you think it is?
How can it be fixed?

If you think I'm insane, please show me a decent, non-edited raw capture from in game that you think holds the same vibrancy of life as the game itself does, and prove my apparent biases wrong; for the life of me, I cannot get over this weird observation.

Sounds like a photography assignment... :p

Use the free camera and take a selfie; I don't have screenshots or footage that does the game justice.

I know I'm bumping old threads here, figured its okay since its on the front page.

VO is not heavy on post processing effects. Things that you see often in movies - such as color grading, grain, vignetting, crazy gamma curves, black/white point changes.

Depending on who you ask, a lot of engineers feel that these types of post processing (effects that are essentially not based on science/math/physics/lighting calculations) "dirty" the image and isn't true to life. John Carmack (engineer of id softwares idtech 0 through idtech 6, the basis of several games and game engines including Quake, Call of Duty, Gldsrc (Half Life, CS)) was quoted something to this effect when developing doom 3, and having full screen effects sort of dirty his real time per-pixel lighting math (Doom 3's main claim to fame).

Its tough; because one can say filmmakers choose to intentionally color and dirty their clean outputs to set the mood and tone. While others will say; well, a war zone isn't just green and orange and tone all the time, and they want the actual real life representation of what it looks like.

I couldn't say if dev time or allocation has left us where we are today in terms of the art direction Guild Software was aiming for - but I will say that it was mostly built at a time where the post processing effects and art direction you see today just didn't really exist back then - and neither did the graphical horsepower to run it.

That being said, I personally think VO has a great graphics and lighting engine. Ignoring some assets that may have not aged well, anything in engine is beautifully lit with bright reflective highlights when needed, and incredible dark shadows when hidden from view. It really helps sell the whole "im in space, this is the only lighting source, and I'll be in pure darkness if Im on the other side".

Could it use some post processing to help "sell" that feeling? Maybe? Its an artistic choice, and choices like that don't always hit everyone the same. Art direction like that also "dates" the game, as artistic trends and colors tend to be products of their time. (Not that VO isn't a product of its own time, but its not hard to spot a war movie that was very clearly shot and filmed in the 2000s versus prior to that).

All I know is, devices today are powerful enough to really render anyone's artistic vision. You would just need the skill and labor to do it.

We certainly do use post-process effects, when they're the only solution (like glow), but that's different than what we're discussing here.

The problem described here is basically just an issue with contrast, saturation and gamma. It's well known by us, internally.

The bigger problem is that to solve it in an optimally performant way (compatible with, say, keeping Quest VR rendering at high framerates) requires touching every asset in the game (or at least processing at load-time), and potentially adjusting a lot of shaders as well.

A lot of effort, which may not be well-spent if we end up changing how the entire pipeline works anyway, in pursuit of next-gen goals.

The issue can also be mitigated by post-processing, to some extent. But, it's not that this is "dirtying" anything, but rather that there's a negative performance impact to taking that route. You're basically slapping more shaders on top of the existing image, to try and fix things that already wouldn't be a problem if they were adjusted properly at an earlier stage in the render pipeline. It's a bit of a trade-off.

In any event, contrary to popular belief, we actually do have contrast / brightness / saturation enhancing post-processing shaders, which are currently in testing on Android in ES3, rolled out there last year. These may eventually come to other platforms. They are one possible solution that we're experimenting with.

But, I'm not sure if that's the road we'll take, simply because there are a lot of other rendering technology things that are also being adjusted and juggled around. It would make this discussion very complicated if I were to include all of them. Like, the production game doesn't render to HDR right now, either, although we do have a branch that does that as well.

In any event, you can see that "Renderer, Graphics Updates" is an item listed on the Game Development Direction sticky. The challenging part is that we have to make choices that will remain suitable for supporting our future goals over a lengthy period of time (probably many years), so we have to approach this carefully.

This is why we have experimental tests and internal branches and test-assets and other things, to figure this out. But, it will happen in due course.

Maintaining a completely in-house engine and renderer brings both challenges and benefits.