Support Tickets

Since the Game manager LUA script keeps running above all scenes etc it would be great to be able to have it running in editor, this will enable the following and more.

Passing of tables and data like player stats between scenes (levels).
Verify that data written to tables etc in the Game manager script when in editor.
Testing and verifying that all data passed to the manager script is OK..

Having functions in the scene manager script would save from having to copy and paste them into each and every scene and speed up the load times of levels/scenes.

overall I think it would be a great benefit to have it running in editor

If you can not afford a VR rig many people look towards TrackIR, its been around for longer than the concept of VR and is in many many games.
You could have it switched on or off (off as standard)

Featuring a wide field of view, high frame rate, and 6DoF tracking technology, the TrackIR 5 dramatically increases situational awareness, adding depth to the PC gaming experience and pulling you further into the game.

When I first requested this (approx 2 years ago) it was free and you only had to register your game with the software to access the API I dont know about now but it still should be free..

• It’s a massive competitive edge in flight and space sims. It changes the game. There’s no going back. It’s that important
• The added immersion and freedom puts you close enough to the virtual world for you to lose yourself in it
• No doubt about it, TrackIR is a significant advantage in combat, and frankly a far more immersive experience. I can’t play my games without it anymore.

Some of the games using there system:

• Star Citizen
• Insurgancy
• DayZ
• ArmA 3
• Rust
• Dirt Rally
• Assetto Corsa
• Elite Dangerous
• Flight Simulator X
• European and American truck simulators

and many more...

Is it possible for you to create a feature within the Visual Mechanics for AI..

Something like

//forums.unrealengine.com/filedata/fetch

//forums.unrealengine.com/filedata/fetch

There are some lib’s on github for cover, suppression etc that you could look into for this feature.
Having a built in AI editor or an extension of the Visual Mechanics would help skyline appeal to gamers for making games and is already a part of the BIG engines out there and would help skyline shine above all other indie engines..

Please implement RakNet Plugins such as Lobby2Client - PC, Lobby2Client - Steam, TeamManager etc.

There is important Navmesh functions to create advanced AI.

Many are already included in Detour Api, Skyline just needs to interface them to Lua.

findRandomPointAroundCircle

raycast

getCost

closestPointOnPolyBoundary

findPath

TattieBoJangle has requested that the material sub panel show the material names on the icons so you know what material is assigned.

Navmesh needs a function telling if a destination is inside navmesh or not.
This can be used by AI to know if some destination can be reached.
It’s often used with AI and random movement, fleeing, wandering or other behaviours.

Box collision size can’t quickly be adjusted in the physics properties.
For example adding a box collision to a tree objects,the box is too big and can be adjusted in the editor.
It’s the same for other physics shapes.

Like most 3D engine, add a new object type that is directional light users can place, remove or control with script.
It makes possible having several directional lights active or activated individually.
And remove Sun light parameters under light tools or make it optional.

Lighting is scattered all around in Skyline, you discover some default parameters you must change are on Light tools for example you never got the idea it was here.

Those could be grouped on same window or same menu :
- Environment setup
- Sun light and Atmospheric hidden under light tools
- Post effects

Common post effects are missing to get better graphics.

-ACES tone mapping
-DOF
-color grading
-chromatic aberration

This is the lighting tweaking lot of time in Skyline

https://s18.postimg.org/4bf6ikd49/skylinegraphics.jpg

Navmesh API needs a function to allow us to setup the agent max turning speed.

Some characters could be heavy with lower turning speed.
It’s useful to tweak agent turning speed looking more natural and not too fast.

This is an old request.

Been able to render to texture using a camera in scene would help when making mini-maps, CCTV (say cameras in a prison and they monitor the inmates.

Been able to use a camera as a transmitter for Render to Texture would be awesome.

All features should be component based, actually Navmesh is a lua call , there is no component.

Skyline should get a nevmesh component.
There would be parameters exposed to component properties like acceleration , deceleration, distance min make navmesh stop, character radius if it could support avoidance.

Also some Lua functions would be needed :
- IsReachable for example, telling if destination position is inside navmesh
- Lua Access to the result array of navmesh calculation if users want to do something else with the calculated path.
- Lua Stop and resume functions , to be able to interrupt a path navigation and resume it if necessary.

I have been thinking about the endgame and if Skyline releases to the steam distribution platform.

Steam seems to be very easy to crack every piece of software/game is available as soon as it comes out via torrents
This way you could create your own DRM to an effect...

Pull data from a steam purchase to a database on your dedicated server/VPS.
When your game loads ask user to login and use there steam email address
Check the email address against your database. etc..

To do this you would have to have access via LUA and pass secure information to and receive it from the server..

1. Lua Web Socket Handles SSL, encryptions etc etc..
2. Use raknet but you would have to write your own Lua lib to work around passing data into lua via curl and passing teh certificate in lua code.

Option 2 would be better for later development when SGE exposes C++ or C# i guess.. and exposing all C++ libs to C# would mean that developers can do this them selves..

Please implement a Headerless Server

GLFT2.0 is a modern format for modern rendering.

Some engines have open source import/export plugins.

It has some advantages over Collada or FBX formats :

GLTF is a JSON based format. It’s very easy to parse and the data already comes typed.
glTF allows specifying an external file for the big data, in binary format.
glTF’s file structure is crystal clear. No extra or redundant data exists. There is only one way the scene definition can be understood
glTF 2.0 fully supports skeletons and morph targets, which can be parsed easily and unambiguously.
It supports PBR based materials using the Disney/Unreal format, which is what most engines and 3D modelling applications use nowadays
It also handles two-sidedness and transparent materials, including alpha to coverage.
glTF 2.0 supports multiple animations per file,it also supports many key interpolation types, such as Catmull Rom and Cubic Spline.

glTF is developed by individuals from companies such as Microsoft, Unity, Google, Adobe, NVidia, Oculus, etc. in collaboration with Khronos.

You can read more here

https://www.khronos.org/news/press/khronos-releases-gltf-2.0-specification

Videos .ovg shuld be added to the filters so you can see them..

Skyline does not have collision volumes objects without rigidbodies.

This is what is used to create hitboxes, collision zones, character or other objects composed of multiple collision parts.

Those collision volumes are used to have collision and detect collision against other objects or detect collision for raycast. Or to get ragdoll feature possible.

This is mostly used in games for characters , vehicles or any other dynamic entity.
Also used for static objects you want to create a simplified collision using some basic collision volumes.

https://answers.unrealengine.com/storage/temp/89957-h2-noscale.jpg

http://www.grenadetree.com/wp-content/uploads/2014/08/ragdoll01.jpg

https://cs.ingame.de/files/2015/09/csgo-hitbox-fix-update.png

Implementation of Game Analytics maybe a good feature for gaining analytics on the fly while users are playing the end game..

Commands could contain:
Disable
Enable
Send generic Event
Send Progress event
Send Error Event
etc etc

You can check for error that users may have, Player progression, Find your top performing audiences for future advertising, see how gameplay influences player retention,

the Ocean system in gen 1 was powerful, easy to use and gave you all the options you could ever want (or need) at a couple of clicks of the mouse, this system would be wasted if not brought into the Second generation of Skyline Game Engine

The System in gen 1 for volumetric clouds was a great system it would be great to see it in Gen 2.

Material sliders have sliders with big range values bigger than 1000.

It would be more accurate and better with slider range values be the same as shaders range values 0 to 1.

https://s13.postimg.org/gq2otpchj/materials.jpg

A good addition to skybox is to have an optionnal atmosphear scattering.

https://d2ujflorbtfzji.cloudfront.net/package-screenshot/59442e6e-a9fc-436a-aedc-59fe5b6ad089_scaled.jpg

https://blogs.unity3d.com/wp-content/uploads/2015/06/scatter_OccludedScattering.png

Reflection probes volumes baking for materials rendering.

Be able to add or delete reflection probes volumes per scene.

https://docs.unity3d.com/uploads/Main/RefProbeHandles.svg

Skyline will blend reflection probes volumes overlapping when rendering to avoid seems.

If no reflection probes are setup in the scene Skyline will fallback using a default scene skybox.

And have some parameters to tweaks, like resolution.

Like most engines be able to setup Physics constraints for movement and rotations on X,Y,Z axis.

https://s13.postimg.org/3upn9xe2v/constraints3.jpg

Support for multiple collision volumes for characters, for hit boxes or weapons damage volumes.

https://s1.postimg.org/5koeezqgrj/collisions.jpg

Acutal Physix engine used by Skyline is v2.8.6.4 , it’s a ten years old version.
Skyline should update to latest version.

For many games, hurling a rocket or grenade and then watch the 3D asset explode into fragments is common. Likewise, fracturing a wall into chunks is all important- like hiding a cave wall that is able to be fractured.

Here is an example: https://assetstore.unity.com/packages/tools/fracturing-destruction-9411

Mike

A box volume for when the player enters it can set/adjust various post effects.

I.e. get navmesh concave so you can get the npc to lean in and fire, the NPC would know that there is an object at a location because of the navmesh data and will know where corners are so it can lean while firing and stay in cover..

There are many many uses for Navmesh data in AI related programming and it gives you more precise control and positioning while seeking out cover than other methods (and the least time consuming) so much better then manually placing cover objects in a scene and getting the npc to position correctly on the corner, manually placing cover objects can be very very resource heavy especially if your doing a full 20km x 20km map.

You can use the Navmesh data with other flags/data on objects in a scene to let the NPC know its a small wall or a object that it can not fire over, overall its about getting the NPC positioning correctly.

This is the preferred method of most AAA and serious game makers.. :)

I.e. get navmesh concave so you can get the npc to lean in and fire, the NPC would know that there is an object at a location because of the navmesh data and will know where corners are so it can lean while firing and stay in cover..

There are many many uses for Navmesh data in AI related programming and it gives you more precise control and positioning while seeking out cover than other methods (and the least time consuming) so much better then manually placing cover objects in a scene and getting the npc to position correctly on the corner.

You can use the Navmesh data with other flags/data on objects in a scene to let the NPC know its a small wall or a object that it can not fire over, overall its about getting the NPC positioning correctly.

This is the preferred method of most AAA and serious game makers.. :)

We need the ability to multiselect assets and group them as one mesh while keeping them separate. I know you can put assets in folders and CTRL-CLICK them to move them as a group, but it’s not the same as true grouping.

Adding sphere or cube rigid body to some character, we can’t change the shape sizes in the editor properties.
Capsule dimensions are also not available in properties.

Some common shaders would be cool :)

toon shading with optionnal outlines
sub surface scattering
furr low gpu cost shader

Anti aliasing, perhaps MSAA and a faster one like FXAA.
Soft particles for great effects.

Different decals systems

• static optimized decals
• dynamic with polling decals

I mean for example you have a pool of 30 bullet decals created at game start, when you spawn a decal somewhere it uses a decal from the decal pool list.
If you have spanwed the last 30 decals, next time it will take the first decal and position it to the new position request.

Game objects for click placement is great for fast decals.
Advanced decals bending on geometry would be a great addition later.

CanReachPosition (Vec3) :
Returns if a position we ask is inside navmesh or not.
Can have many uses like checking goal position can be reached or random positions are inside navmesh.

NavmeshRaycast(Vec3 start, Vec3 goal):
Like physics raycast, but returns if there is some intersection in the navmesh from a start position to a goal position. It’s should be less expensive than raycast, but allowing to detect ground static obstacles.

Skyline prefab system is more like a grouping system, when you drop one in the scene you got all group items in the hierarchy view.

Instead make a real prefab system, a prefab is a new file type, dropping a prefab in the scene displays only one object prefab without child objects in hierarchy tree.
It’s more clear and efficient.

The object will continue to react to physic forces, but it will not use gravity.
This is used a lot for new and fun gameplay styles.

Nvidia functions description
http://docs.nvidia.com/gameworks/content/gameworkslibrary/physx/guide/Manual/RigidBodyDynamics.html

Be able to add mutliple collision volumes per object.
This could be usefull some times to get low poly approximate collisions.
Or a tool able to create some low poly collision volumes automatically from a mesh.

Those additions could be useful.

Have a new icon on hierarchy list to show/hide objects or folders containing a list of objects.

This is usefull for level design, for example you could hide different group of objects similar to show hide layers.
For example you could hide roof top objects or large objects, to be able to view some interiors and start placing small props.