[=Deathrey;583647]
Thanks for your response, . Could you potentially allocate someone to take an in-depth look at this, because not only the fix does not address the issue, linked to the corresponding tracking number, its “tessellate the landscape only for the highest landscape LOD” part could potentially be usability regression. All relevant links are present in Community References sections on the issue tracker page, including forum thread where it is discussed. Thanks in advance.
[/]
Best option at the moment is respond on the already opened ticket you have for AnswerHub and let the our support staff continue the investigation there. I see you did respond back to in your post you do have.
[=Bladerskb;583606]
Based on your example i don’t see the difference.
Its like people saying Epic should implement VXAO when DFAO is exactly the same thing.
https://.com/watch?v=FmbrnbpHT9E[/]
Aren’t the same at all one is voxel based the other no, anyway the people request mainly for GI solutions rather than AO, from what I see in the forum
[=;583870]
Aren’t the same at all one is voxel based the other no, anyway the people request mainly for GI solutions rather than AO, from what I see in the forum
[/]
i’m well aware of that. I’m referring to the results.
[=Bladerskb;583908]
i’m well aware of that. I’m referring to the results.
[/]
DFAO is messed up on scaled meshes, is not capable of casting AO from skeletal meshes, has extreme ghosting when it comes to movable stuff. VXAO does its job for all meshes, and it does it more accurately, with no ghosting.
And quality of VXAO appears to be much better for interiors
[spoiler]
dfao
[=Ichijo_Norifusa;584213]
DFAO is messed up on scaled meshes, is not capable of casting AO from skeletal meshes, has extreme ghosting when it comes to movable stuff. VXAO does its job for all meshes, and it does it more accurately, with no ghosting.
And quality of VXAO appears to be much better for interiors
[spoiler]
dfao
i’m sorry but there are alot of things that need to be shadowed in that vxao picture that are not shadowed.
If you want to do a real comparison, use the same scene. VXAO is available in UE4 through github.
Fire it up and compare scenes with DFAO.
lastly the cost difference is night and day. DFAO actually runs on consoles.
VXAO? not even close. The most powerful graphic card can barely handle it.
Just like with any nvidia technology, bloated inorder to prop demand for the latest and greatest graphic card.
Is this what you want? https://www…com/watch?v=rn_aZYUi2xs
UE4.13 Preview 2 - Error from Material if you combine the Metallic and Emissive Color the preview automatic white?
Note: Bug Report - UE4 Answerhub they are not answer, i have a lot question but still useless to post :(. Maybe they are busy…But thanks for new release
[=;584266]
UE4.13 Preview 2 - Error from Material if you combine the Metallic and Emissive Color the preview automatic white?
Note: Bug Report - UE4 Answerhub they are not answer, i have a lot question but still useless to post :(. Maybe they are busy…But thanks for new release
[/]
Hi ,
Can you link me to the post you put on the answerhub? Did you make sure to post it in the Bug Reports section?
I already done… but i hope someone reply and not ignore my Question thanks
One more, when i try to rename folder the result: the file transfer to new folder and leave the old folder or its just move the file to new folder created.
Legend:
FFF - old folder name.
Fsub - subfolder
material_file - file inside the folder
F - new folder name
Example:
FFF(Fsub + material_file) === after renaming the folder ===>>> F(Fsub + material_file) + FFF(Fsub)
Can you try right-clicking the folder and selecting “Fix-up redirectors”? When moving files, other things (such as Actors, Blueprints etc.) might still reference your asset, so a redirector (think: a link or shortcut) is kept in the old place.
[=Quantum Theory;582145]
Has anyone gotten mesh decals working and can post a quick how-to?
[/]
all you need to do is make a material with domain set to Deferred Decal and adjust accordingly
maybe I had a bit of a bigger expectation out of this feature, but for me the Mesh Decals feel just like another translucency mode. Sure it’s better to blend with the details under it and it’s probably cheaper (since decals are “free”), but I thought it would behave like a Decal in terms of projecting into the underlying geometry.
@DarkVeil: I guess this feature was not intended for that?
I actually wanted to use it onto a SplineMeshActor in an attempt to have spline decals
PS. if I apply a Decal material to a SplineMeshActor it crashes unless I manually enable ‘used with Spline Meshes’ first in the material
I’m new to Unreal and have 4.12.5 installed. I’ve been noticing a lot of my comment bubbles getting deleted for no reason while working in the Blueprint graph. For example, inside a Character BP event graph, I have the EventActorBeginOverlap node connected to a CastTo(character) node. The character pin from the Cast To node is promoted as a variable. I had a comment bubble on both the CastTo and the promoted variable. Once I renamed the variable, the contents of both comment bubbles were deleted. I pasted the comments back in to both. Then I added a Custom Event (not connected to anything). Wham! The comment bubbles again went blank. This makes comment bubbles very unreliable for taking notes during tutorials. I’m wondering if this will be (or has already been) fixed in the upcoming release? Thanks!
great to see progress being made with the much needed deferred mesh decal material! been looking forward to that feature for over a year now! Up until now I was using hovering geometry with masked blend mode and projected the underlying uvs/vcol in max to blend details in. Mesh Decals theoretically should make that workflow obsolete but currently there are limitations that prevent me from switching over:
> Bump Offset not being supported atm (really useful and efficient as it only occurs in tiny masked spaces and thus only alters ps cost in a fraction of the screen space http://imgur.com/a/oRBw1
> Blending is very limited for something like color as it just replaces the underlying color with the defined opacity value. It would be great to have the opportunity instead of replacing the underlying color to use something like additive to modulate the base color or roughness while using the opacity mask to define the overall blend weighting in general
Would be great to know if there were any changes planned to the current implementation of this feature.
[=;575690]
What does “landscape tessellation performance improvements” mean? Does it mean that areas with a tessellation multiplier of 0 won’t incur any tessellation cost?
[/]
The landscape is broken into multiple lod levels with different vertex densities. It dynamically changes as you move. This new change turns off tessellation on all outer LOD levels (distant mountains, mid range stuff) so that it is like you have a tessellation multiplier of 0 out there with no performance cost.
An actual zero multiplier in the material still has a performance cost in the area where tessellation is still turned on (the nearest LOD).
[=;585250]
An actual zero multiplier in the material still has a performance cost in the area where tessellation is still turned on (the nearest LOD).
[/]
thanks 
