Rick

Even if that's what he really meant, the first video released of 2.4 is a video showing pureLIGHT working inside LE, so I think it's natural that a good number of people assume pureLIGHT is the big bang thing in 2.4. This is a confusing release, but honestly most LE releases are like that. The coordination in an LE release could stand some improvement, although I'm sure it's not easy to update everything needs to be with only 2 people.

It is a shame that a better partnership couldn't have been made with pureLIGHT. The funny thing is that after the month is up it would seem it's back to $500 for us LE users. Who is going to buy the LE engine for $250 and then spend $500 to get GI? Not very many, that's who. A better deal would have had to be made I think. Seems it is baked too. Bummer, a real-time GI would be really cool.

Welcome to product advertising and semantics. In software land you'd be hard pressed to find any advertisment that is really what things are. Seems to be the nature of the business.

MouseZ() in the RenderContext doesn't seem to be working correctly. I did a small test with LE in C where I just printed out the value of MouseZ() to screen and it just keeps accumulating. If I do the same thing in the RenderContext it seems to kind of loop back around after around 450ish. Sorry for finding so many bugs, but these are all the features I need

Using the framework is there a way to make models always draw on top of everything? For example with doing a 3D gui or something like the red/yellow/blue axis in editors? I see the transparency world which I assume is like a foreground world, but it still has depth to it.

I'm confused as to how it could be pre calculated lightmaps if it was working with all those dynamic balls. That seems real-time.

No, just play around with some ideas.

nevermind, looks good! Thanks! Not a big deal since you kind of did a MouseRelease on the fly, but it looks like it keeps calling it after the first time.

Oh this is going to be a separate purchase? I thought it was bundled into 2.4. Bummer http://www.purelighttech.com/product.htm $500 stand alone. Ouch. I really hope it's cheaper than that for us.

There still seems to be a problem somewhere. After calling MoveMouse(50, 50) visually the mouse seems to be moving, but calling MouseX() & MouseY() still give the old mouse coords. Is something not updating until the next frame or something? origx = RenderContext.MouseX(); origy = RenderContext.MouseY(); RenderContext.MoveMouse(50, 50); // if I check RenderContext.MouseX() or MouseY() they aren't 50, 50, but are what they were before the move

One thing I always disliked was that LE didn't have a MouseUp method that fired just once after coming from a MouseDown. I've always had to do this myself, which gets old after doing it a hundred times. Do you think it might be something you could put in the control? Maybe it could be called MouseRelease() or something to make more sense.

I probably should have said one other minor thing, but MouseHit returns and int and MouseDown/Up returns a bool. Not a big deal now, but whenever you put another fix it, it would be cool if MouseHit could also return bool. Also, Leadwerks.Mouse.Hide() doesn't seem to hide the cursor. That might be something for the Leadwerks.dll though. [EDIT] nevermind on the mouse hide. It's working. Thanks for the fast turn around on that!

It doesn't seem like MoveMouse() of the control is working in this certain case. In your example program when you click on an area, the camera rotates right away. Ideally in editors and games that doesn't happen. The clicking down usually doesn't generate a jerking movement to that position. To solve this I tried the following, but when stepping through after calling MoveMouse() and checking the MouseX() and MouseY() of the control they were left unchanged. if(RenderContext.MouseHit(MouseButtons.Left) == 1) { //Leadwerks.Mouse.Hide(); RenderContext.MoveMouse(100, 100); } if (RenderContext.MouseDown(MouseButtons.Left)) { //Camera look mx = Maths.Curve(RenderContext.MouseX() - 100, mx, 1); my = Maths.Curve(RenderContext.MouseY() - 100, my, 1); RenderContext.MoveMouse(100, 100); camrotation.X = camrotation.X + my / 5.0f; camrotation.Y = camrotation.Y - mx / 5.0f; Leadwerks.Framework.Layers.Main.Camera.Rotation = camrotation; } I just subtract from 100 instead of using 1/2 the width and height for 2 reasons. One, is to avoid the division when it's not really required, and the other is to avoid rounding issues when the control isn't an even number. Either way, this code results in the camera going insane and spinning. Any ideas? [edit] The mouse always seems to go to the middle on left mouse button pressed down too, which doesn't really make much sense given the code above. It should be going to 100,100 (which isn't the middle of my control). It's almost like somewhere it's hardcoded to move the mouse back to the center of the control, so when I move it to 100, 100 it's always calculating the difference between the control center and 100,100 (what I move it to) and that's why it's moving like crazy. If that's the case, it doesn't really seem like that should be happening.

I think models can also be in the foreground. That's what I'm interested in. Making an editor and I planned on using models for the axis selection tool, but it needs to be in the foreground I think so it's on top of the other models so I can still see the axis selection models when they are "inside" the real model to move. Hmm, I load a model in the foreground but it's not drawn on top of everything else. How can I get a model to be drawn on top of everything?

Are we able to pick models with CameraPick that are in the foreground? I seem to recall that you couldn't but wanted to see if anyone just knows off the top of their head before I start on something.

Yeah, I wasn't trying to imply that. I was just saying that if a library changes often enough people get confused (with all the old posts out there) and it makes people not "trust" it enough to use it in their projects. The old posts and such is a real problem I think. For someone new in LE C# it can be confusing as to what you are supposed to us (currently).

I was wondering how I can tell how much light is on some model. Say I want a game where you can use dark areas to hide, but those dark areas are dynamic in that sometimes light shines there and sometimes it doesn't. Is there a way I can tell this using some kind of light information or would I have to use some tricks to get this to work. I would prefer it to be real lighting information. For in the case where there is a round hallway with little side areas and a floating light ball travels around the hallway that shines light on one side of the hallway and then the other side, forcing the player to alternate sides to stay out of the light. There could be many different scenarios so faking this with triggers and such would kind of suck.

That lifetime membership is a good deal. You have some nice assets on your site and if it continues to grow that's the deal that keeps on giving. Keep up the good work.

The LE C++ library is just more mature and has been around longer. The C# library has gone through some revisions and has been around a shorter amount of time so it just needs time to become stable. Although the lesson for anyone maintaining a different language library is to not always mess around with how it works because people get annoyed when things change all the time.

If you make a simple test of this you should be able to see what it's doing.

Or to destroy the world. Depends on how you see it I guess Pancakes, usually a networking library is kind of setup like the LE library. It gives you the commands and such, but you still have to fit the piece together for your game. Nice Mumbles. Here I just thought you were hanging around all depressed and such, but you are kicking butt in development.

You wouldn't get physics that way though would you?

Was just thinking this could produce some interesting effects. Not sure how efficient it would be, but might lead to some funky stuff. Like if a cube had enough verts in it, and you applied this to it and randomly applied forces to all the bodies but kept their movement to a certain range it could create some kind of blob looking thing. Then if you applied that to a plane it might create some interesting water effects.

Before I would do any testing I was just curious if anyone knows if we can parent a meshes vertices to bodies so that the vert would move with the body causing a distortion of the mesh?

Josh has mentioned before that he would like to merge the two (LE Editor and 3DWS) so that we can create CSG inside the LE Editor. This would be ideal and I can only hope 3.0 will have this.