In Unity3D the logic dictates that objects are not to be rendered unless in the field of view. This is obviously the way to go for optimization purposes. However, it still renders meshes that can not be seen by a player due to being occluded. I would like to solve this and was wondering if there was already a method to do so or if I had to do it myself.
Here's a picture to help illustrate my point.
So far my only real ideas are using the culling, but that still would be in a range not necessarily visible.
https://docs.unity3d.com/ScriptReference/Camera-layerCullDistances.html
I guess essentially what I need to know is how to do occlusion culling after a scene starts because the scene is generated, it's not premade.
For anyone who's interested, I asked the unity community
Answer by Bunny83 ยท 4 hours ago
No, that's not possible. At least not with the occlusion culling
system that Unity uses. The calculation which parts are visible from
which points is quite complicated and has to be precomputed in the
editor. So it won't work for procedurally generated levels.
You have to roll your own solution if you need something like that.
Just a few weeks (or month?) ago i implemented a VisPortals solution
similar to those used by Doom3 (basically how most ID Tech engines
work). It's more a prove of concept than a ready-to-use solution.
Usually i upload a webplayer demo onto my dropbox, however i just
realised that Dropbox finally prevented to directly view HTML pages
off my public folder. They now force a download of the page which
breaks everything. So if you want to try it, you have to download the
project.
Of course vis portals doesn't work in all situations. They are perfect
for closed environments which can be split nicely into seperate areas.
Of course this splitting into areas and the creation of the visportals
currently is done by hand. So you would need to automate this
yourself.
Be careful with static batching, it might break the system as each
area has to be seperate so it can be enabled / disabled seperately.
Related
I'm trying to make a simple AR Images app, for students to use but, since I don't have much programming experience, I'm trying to just use and change the example files as the base for the app.
Until now, after many hours of trying, and reading about it, I still couldn't make it work with my own images and prefabs... Using AR Images should be the easiest thing in ARCore, right?
Could someone please help, by uploading a working base project, that is independent of the ARCore example files? Or at least the main scripts with these improvements:
using several images and prefabs, like 15 or 20 max.
an easier way, like a drag and drop prefab list, of corresponding the images number or name with the correct prefab...
...maybe like shown in ARCore + Unity + Augmented Images - Load different prefabs for different Images
an easier way of updating trackables - their state, adding and removing prefabs when visible or they get out of sight
I have seen some answers to individual problems, but it's hard to put all that in the scripts, without messing something else... at least for me it is ;-)
I think this could help a lot more people besides me, by updating this problem and putting several answers in one place. Thank you!
Using: Unity 2019.2 beta - ARCore 1.10 for Unity
Here's the deal - I'm working on an algorithm/library that is able to generate a navigation mesh in virtually any environment where I can get coordinates for the controlled agent and/or for other agents within the same static environment. The only input I have, is a collection of points where an agent has been to.
(See the image here to hopefully understand what I mean)
I already got to the point where I can create navmeshes manually and navigate on them well enough. However, in larger environments, having only coordinates of, say, the controlled agent, it's really tedious and time-consuming to manually do it.
The uses for such algorithm/library for me are obvious, but I have put a lot of thought into it already, so I'll list a couple of things I'd like to accomplish:
Robotics (scans environment, only gets distance from self to a point, hence getting coordinates - no need for complicated image/video processing)
AI that is able to navigate an unknown and unseen maze (any shape or size) by exploring it
Recording walked areas and creating AI for games that don't know certain places unless they've been there
Now you hopefully see what kind of solutions I'm looking for.
I have tried a couple of things, but couldn't figure them out. One of the most successful things I've tried is giving a range to each individual point (creating a circle), and then looking for places with overlapping circles - you can most likely move on those areas. The problems with this approach started with triangulation of the areas. The resulting mesh may be a little inaccurate, but it must be able to connect to existing ("discovered") parts of the mesh seamlessly (not everything has to be interconnected somehow, as agents can disappear and reappear, but within reasonable proximity, connect the mesh).
Some more information: I'm working in C#, though solutions in java, C++/C, objective C, pseudocode etc are equally acceptable.
P.S. I'm not interested at all in answers like "just use this library" or "use this other language/environment" etc... I want an algorithm. Thank you in advance.
I can help with 2D path finding. You need to find the red outline. Then you can use a voronoi diagram with the red outline (not the agents points). Remove all edges outside the red outline and the remaining edges can be used to navigate the shape by someone/something. Read about it:http://www.cs.columbia.edu/~pblaer/projects/path_planner/.
I'm making an XNA 4 game, and I want to create checkpoints for certain areas in my stage. I figured I could create a "ghost" object of the player object that would be created whenever the player reaches a checkpoint. And when you want to reload that checkpoint, you'll start to where that ghost player is.
As for now, I thought this would be an easy way to achieve this (although I think XML may be a better solution, but I've no idea on how to use that, yet). But the player has too many variables (health, stamina, ammo, bleeding timer, silver keys, golden keys, coins, infection, position, speed, angle, states [involve dying, infected, bleeding, burning, dodging, etc], I just thought that statemets like "continuep1 = p1;" (both are instances of the same object) would re-assign all the variables to the values that the other class contains, but when I tried to re-load "p1 = continuep1" it wouldn't work. It doesn't seem to do anything.
So I'm wondering, do I have to re-assign ALL the variables one by one? Should I start using XML? Or is there a way to assign all the variables without having to do it one by one? (I'm not asking for code, unless the last question is possible)
In my opinion I would say yes, start using xml. Don't worry about using xml, it's pretty simple once you get your head around it. A great way to learn would be to see your characters information displayed in xml format.
I think the most ideal option would be to serialize your game. So this involves storing the state of your character, the position it is at, the direction it is facing(which texture is currently loaded) and the stats(health, staming, bleeding etc).
With this you could reload your game from the last save when the player dies or when the game is next played. This would solve both cases.
This tutorial on making a top-down RPG game in Xna 4 is very good in my opinion and goes into good detail. Many of the techniques in this tutorial apply to more than just this genre and style of game.
http://xnagpa.net/xna4rpg.php
Part 11A(Game Editor) is the first part of a tutorial guiding the reader in making an editor for their game. With this Winforms project items, classes, quests etc can be added to the game and are stored as Xml files. At the bottom of page 6, and pages 7 and 8 outlines methods to serialize and deserialize your game.
In Part 11C(Game Editor), particularly at the bottom of page 9 and page 10. A save game method is introduced which serializes the game. You can also look at the new game and open game methods too.
These methods work with DataManager classes which store the items, classes etc in Dictionary's. This may be how you want to model your game information, if not the code in these guides may have to be altered to work with your solution.
I thought it best to provide my source for learning on this subject, rather than regurgitating it. I would recommend reading around this, not just the pages I've pointed to. Furthermore, in a later tutorial the serialization goes a step further and the files are stored in .xnb, which is a much more secure way to store information.
I hope this has helped and that you appreciate the guide in the right direction as opposed to just being given the answer. Besides, this isn't a task that would be solved in one method anyway.
Let me know how you get on and I'd be happy to help more.
I've been trying to create a roll-playing game for a little over a year now, and I thought a good way to test it and flesh it out a bit would be to turn it into a Dungeons and Dragons variant. However, the inherent problem with using a DnD setup is that there's a lot to keep track of, to the point where it becomes impossible for a single person to do by themselves. As such, I thought of writing a C# program to help.
I ran into this problem when I tried using a series of PictureBox objects to represent the spaces on the map where characters can move. I wanted to use PictureBoxes because that would allow me to use images to represent terrain and characters occupying the given space. However, the map I am using is roughly 46 x 75 1-inch squares, totaling 3450 PictureBoxes. This, understandably, slows the application down so much that it actually freezes for minutes at a time while it redraws the map every so often.
I tried two solutions. First, I tried using a Panel object for its free scrolling capabilities in the hopes that the application wouldn't have to redraw the whole map, but rather only the subset of PictureBoxes visible at that time. This helped only a little, and not enough to make the application usable. Second, I looked online, including this forum, and found people having similar problems. However, the problems they were having were entirely unrelated to mine (aside from the whole slow redraw thing) and thus the solutions didn't really apply to me. I did see a few recurring lines that I tried:
this.SetStyle(ControlStyles.AllPaintingInWmPaint, true);
this.SetStyle(ControlStyles.OptimizedDoubleBuffer, true);
this.SetStyle(ControlStyles.ResizeRedraw, true);
I'm not really sure what they do, but they did help a bit. I am still seeing way too much lag, though.
I'm out of ideas at the moment. Is there a better control I could use or a different way of drawing? Thanks in advance for your help!
I would try to develop a solution that only renders the images that are visible on the screen at any given time. Images should only be rendered when you scroll them into focus.
This is a common technique since rendering UI is usually the slowest operation by far. I would bind a view model object to the visible objects, which means you only have to render a subset of the images instead of the entire screen.
Grids with endless scroll sometimes use this approach.
I am currently making a simple game in XNA but am at a point where testing various aspect gets a bit tricky, especially when you have to wait till you have 1000 score to see if your animation is playing correctly etc. Of course i could just edit the starting variable in the code before I launched but I have recently been interested in trying to implement a console style window which can print out values and take input to alter public variables during run-time.
I am aware that VS has the immediate window which achieves a similar thing but i would prefer mine is an actual part of the game with the intention that the user may have limited access to it in the future.
Some of the key things i have yet to find an answer to after looking around for a while are:
how i would support free text entry
how i would access variables during runtime
how i would edit these variable
I have also read about using a property grid from windows form aps (and partially reflection) which looked like it could simplify a lot of things but i am not sure how I would get that running inside my XNA game window or how i would get it to not look out of place (as the visual aspect of is seems to be aimed just for development time viewing).
All in all I'm quite open to any suggestions on how to approach this task as currently I'm not sure where to start. Thanks in advance.
I've used this in the past, and it worked great.
http://xnacc.codeplex.com/
It will require some programming to set it up to work with your game, but may be worth the effort if this is something that interests you.