Thanks for your suggestions ![]()
1. Normal & Texture Blending for objects on natural terrain
We have been thinking about something like that in the past, but I don't think the performance impact is really worth the effort... We barely have natural elements like on the screenshot (i.e. elements where a large portion of them is covered by terrain), and on the few remaining elements where such a technique makes sense (e.g. rocks), the effect would be barely noticeable in RW I guess. In RW there are probably only very few situations where such an effect really pays off...
2. Stochastic textures for natural terrain (and probably some materials)
While this is a great technique for natural looking materials, I'm not so sure if RW would really benefit that much from it... It has an impact on performance, especially when combined with triplanar mapping (which is used on our terrain). We're already using a distance-based tiling which reduces tiling on mountains, and while this is far from being perfect, it seems to be at least sufficient on naturally generated terrain. One still notices tiling on flat surfaces (like the current area surrounding the demo mountain for example), but after the world gen update, that area is covered by water anyway.
But this is something we may implement in the future ![]()
3. No-intractable debris from object
Looks like right now "physical debris" from all objects can completely break player interaction with world, for example if you will use pickaxe in front of debris - it will destroy debris piece and not object behind it. I think this is a bit unbalanced behaviour as right now debris is a visual option. So, my suggestion is to make it ignorable by all raycasts and interactions so it will be more like 3D particles than real objects.
We could maybe introduce an option for this
The original idea was to give the player a way to remove debris manually (considering it takes some time until it despawns automatically, depending on the settings), especially if it obstructs the view ![]()
4. Mouse scroll enhancements.
Mouse wheel can change cells on the hotbar, but the problem is that it probably doesn't always behave as you expect from it. For example it will not switch hotbar cell if you will spin it too fast. And if it will change - it will change only one cell. My suggestion is to make it similar to other games (like Minecraft, for example) where you can smoothly change all hotbar cells with one spin instead of constantly pausing between them or moving wheel very slow.
I fully agree on this. Scrolling definitely needs a rework so it behaves like in the Java version ![]()
5. Increase shadow bias (or make it configurable)
There are some cases of shadow bleeding of some surfaces. If shadow bias will be a bit larger - light will have less artifacts. It probably can be configurable inside graphic settings as different shadow resolution probably require different values.
Most of our lights have a unique bias setup... we could introduce a configurable global factor for this, but I'm not sure if there is a value that fits all situations...
Is there a specific light you where you've experienced any bleeding or artifacts, or could you maybe share a screenshot of a situation where this happens?
Small bug at the end of list - shader distance will be reset on every game re-launch to default value. I didn't test other values but this bug can probably have same effect on other graphic settings.
Oh, thanks for letting us know, this is indeed a bug! We'll fix that with the next update ![]()