Scyler_27

Wow, this topic still lives. Guess I should say: Sorry guys, this is discontinued. A while back I had an accadent that wiped my HDD as well as my backup, removing years worth of projects. This was one of those projects, so it no longer exists. Sorry I didn't mention sooner.

This thing was fun, albeit a pain to work with. Now with Windows 8 and it's retarded tiles system, I haven't a clue how this thing would behave when switching back and forth from desktop and tiles. I might very well start this back up, but if there is any chance of this happening, I will likely target it for Ubuntu instead of Windows (and no, it wouldn't be multi-platform because I would have to re-write the C++ external files for each OS which isn't especially fun when it comes to messing with windows and such. Different systems and all). Though even if I do decide, I have to see if it is possible, how to do it if it is, and I would have to write the skinning part of it myself which I haven't a clue how to do currently.

Anyway, that is what is up, sorry for the bad news! Just noticed the topic got bumped and figured I should say something.

The issue is object depth/priority. It sounds like all of your objects are the same depth, thus their priority is dependant on the order that they are placed in the room. You want all the objects to be drawn AFTER the 3D view has been speicified, otherwise they will be drawn in 2D.

To fix your above issue, give the camera object a higher depth. By default, all objects have a depth of 0. If you give your camera object a depth of 1 or higher it should be drawn before the 3D objects fixing the issue.

I hope this helps.

1. Hum, that is a possible future feature I could add. I will add it to the suggestion list.

2. In the next version, given it comes out (see above rant), I already have support for that added.

3. Fair enough.

*Sigh*... Frustrating news. Don't expect an update, at least for a while. GameMaker is becoming a pain in the butt to work with. Here is what is going on:

1. I had a nice optimization for shadow drawing that combined shapes into a single primitive. Had to remove that because GameMaker apparently limits how many triangles can be defined in a primitive. Joy. 

2. I added culling to speed things up. Turns out there is a small issue when drawing shadow caps and shadows together. Although they are drawn based on the exact same data, from the exact same source, they are drawn in slightly different positions revealing an ugly border around certain sides of objects (which became noticeable with the culling). Figured it was an issue with me somewhere until I discovered that (depending on the platform being compiled for) they are offset and drawn differently for each system! For example, the shadow caps are drawn on the shape perfectly for ubuntu, though the shadows (based on the same points, mind you) are drawn offset in a weird fashion, mostly left and top. However for windows, the shadow caps are offset top and left while the shadows themselves now have border issues around the ENTIRE object! The ONLY difference I can find between the two 'calculations' is that one uses a trianglelist while the other uses a trianglefan. 

There are a few other things I have managed to work around, but this is making things extremely complicated. I was hoping just to add some OS-specific scaling to certain values to fix the offsets, but it is confusing in that there is an offset between two shapes which are drawn with the SAME VALUES... so I am baffled as to how that is happening and how to fix it. As such, I have to figure out how to beat GM into submission and do what I want it to, which as far as I can see, will not be simple. 

Sorry about the delays, but it turns out multi-platform is even more annoying than I first figured. 

[/annoyed slightly informative rant]

@_226403, try to work with us, mate. You are making it difficult for us to help you.

Listen, you are going to have a heck of a time programming ANYTHING at ALL in 3D if you do not understand the concepts behind what you are trying to do. Look at the FPS tutorial provided and see how things work. YES it is first person, not 3rd person, but the concepts are still there and you can learn from them. It has 2D characters represented in a 3D space like you want, and shows how to set them up. If you do not have the knowledge to at least conceptualize converting a 1st person camera to a 3rd person camera, then you do not understand how the camera system works in 3D and should learn that first. 

Again, look at that tutorial. If you learn how to work with animated billboards, then it is just a matter of timing the animation properly and plopping the camera in the correct spot. 

So now, instead of "I don't want this, I don't wan't that," and "I don't want to copy other people's engines," how about you try learning from these things anyway and figure out how to actually do it? Programming of any kind will take a lot of RESEARCH and effort on your part whenever you want to learn program to something new. This can include tearing apart other people's examples or engines.