Pixel Perfect

I don't want to get into arguments regarding which engines are better than which and which are financially a better bet. All I can say is you pays your money and takes your choice (either upfront or afterwards if the game proves successful). But hear are some of the observations I make when looking at whats out there. Unity and UDK will always attract the lions share of interest simply because of the number of published games they have which is substantial in both camps. Free engines do look attractive up front to developers with little to no money to invest but certainly in the case of UDK come with a massive learning curve. You wont be doing much in the first 6 months of using UDK other than 'learning the ropes'. But this is a powerful engine which has powered some huge games in the past and still does ... and as such represents to many a 'safe bet' and was an easy migration for the huge amount of modders out there that already have experience of the engine. Unity has however, in the last few years, been massively successful at delilvering games to the iOS and Android games marketplace with even professional studios now choosing to develop games using it. It also has a vibrant user plugin marketplace offering a huge diversity of plugin game functionality from end users to again recognised well known developers from existing large game houses. The professional version (fully featured) is of course not free and considerably more expensive than Leadwerks. I have no idea what the learning curve is like but I suspect it is faster than UDK. LE2 is likewise a great platform to base a game engine development on and is, more than most people seem to realise, quite capable of producing publishable games. For those who $200 dollars is not a large investment, which is generally true for anyone currently employed, then it too represents a great investment with potentially larger returns than either of the other two previously mentioned engines as no further payments are necessary (the point NA was addressing). Unfortunately not a single published game to date which rightly or wrongly in the eyes of a new developer is bound to raise questions! LE3 is still an unknown at this point and I feel its unfair to comment on this although I look forward to seeing what's eventually delivered.

Thanks for the update Mac, I wasn't aware it had any vertex weighting capability, so I stand corrected It does seem that the more you explore it the more you find, it does have an astonishing amount of functionality ... but as you say probably not the app to rig anything complicated with.

It is a recommended tool mainly for its well known ability to import and export a vast number of standard model file formats including the ability to export to Leadwerks gmf. However, although it supports a lot of tools and functions for modifying/fine tuning models it is essentially a UV texture mapping app and not a modelling app; nor does it support rigging of models! Well worth the small cost of the investment though for anyone converting models including animated ones.

To be honest anyone could learn the basics of SQL in a day. It should never be held up to be an obstacle to using SQL Database technology. Plus, its unlikely anyone will be querying more than one table at a time which further simplifies the scripts, its not like the constructs are going to resemble a huge relational database! With a good SQL front end for editing purposes this is almost a 'no brainer' for storing game related (especially configuration) data.

Thanks for releasing the Crysis based shader, much appreciated

It's is also worth mentioning for anyone else reading this that 3DWS is a game level design tool too, as well as a modeling program, supporting terrain generation and texturing along with model placement. To say it's just a modelling program might confuse people! But as Josh rightly points out it's not a game engine so you would need to export the levels for use with a game engine! Leadwerks Engine actually has better level design tools too imo!

I don't think you will be able to use a byte pointer to return the string from C to Lua as Lua will simply regard this as a userdata type (equiv of a void* in C) and can do nothing with it other than allow it to be returned as a pointer to a subsequent C routine. There is no type casting in Lua as far as I'm aware. You could sent the value as a separated item tagged onto the message parameter and process those in your callback routine (e.g. 'move|1') [EDIT] Ah, didn't see Lord Hippo's reply!

Check their AABB box extents, see that you don't have anything silly which is really extending them. Are you really creating 3800+ entities Rick, have you done a count in your code? That does seem like a hell of a lot given your veg density looks to be small and its only a 128x128 terrain!

I'd have thought frustum culling should have been taking care of anything outside of the viewing area.

Metatron is right in as much that it is generally bad practice to copy the includes from their original directories and it is better to point the compiler to those, however he is factually incorrect to state that the compiler wouldn't find them as the preprocessor does indeed check the current directory. (The only exception to this I know of is in VS 2010 where if you use the angle bracket syntax as opposed to the quoted include syntax the current directory is omitted). Like all things in life there are sometimes situations where you might want a seperate copy of library files rather than pointing to a common repository which may be subject to unverified updates during your development cycle.

You will need to ensure either engine.h resides in your project root directory or your copy of 2010 includes this directory as part of its list of directories of include files. The include file is found under the CPP directory of the Leadwerks SDK. Setting a project up using the LEBuilder tool included in the SDK should prompt you for all the information you need to get an example LE project up and running.

Mr V ... now really! I actually had some sympathy for you at the beginning but you have now proved yourself to be a person of little wit and most certainly highly dubious mathematical ability! I now await your further posts purely for their entertainment value, please keep them coming. Not laughed so much in a long while If we all take shifts maybe we could keep you awake for weeks

I don't think watchers of pirated movies need additional help with that activity, whereas users of pirated development software frequently do. So if you, just for instance, were in possession of a pirate version of Leadwerks and needed help you'd be pretty pissed off by now having found that you don't have automatic access to all the forums knowledge! I somewhat think that was the point of the statement! With regard your question, the cost of LE3 will be cheaper for current owners/purchasers of LE2 even when paying the upgrade cost than to people buying just LE3. No costs have actually been announced as yet though.

Flexman has probably done the most work on this and is best able to advise you. Hopefully he'll see this and pass some knowledge on

Mr Valentine this is a great forum full, on the whole, of great people who love this engine and work harmoniously with each other, but because of those who would pirate software and spoil it for everyone it had to be secured. Now whether you like or agree with that or not, it is the case. Maintaining this stance is doing you really no favours at all, you are not going to change that! If you purchase the engine you will be allowed access, if you don't you will not. There is nothing personal in this, its simply 'how things are'. There is a 3DWS forum on the TGC site and in the absence of anything else here if you need to raise issues/discuss 3DWS with anyone including the creator then this is seemingly your best point of access. I hope we can draw this to some sort of closure as it is really achieving nothing.

Many thanks for this and all the work you've put into it. Will give me a lot more flexibility for making tools. Nice work

Demo of the Dark Voices facial animation model running in Leadwerks. I just completed this tonight having fixed the few remaining issues in Fragmotion. It looks to be working fine (after a nightmare trying to get it converted properly) and completes the first stage to ensure the model animates correctly in Leadwerks. The next phase is to code the animation control required for the lip syncing. She is simply running through her sequence of phonemes and facial expressions repeatedly. http://www.youtube.com/watch?v=rBpt_gbZrTw

Any update on this? Did you manage to resolve the problem?

You may need to increase the PointEntity rate setting (the one that's currently set to 0.001) depending on the frame rate you are getting to ensure it curves sufficiently to reach the target, especially if the targets are close together. Try the setting initially as 1.0, this should not curve at all and send you vehicle from one to the next in a straight line. This is a good first check to see that the rest of the code is working correctly. I have always used 3 (z) in this scenario as 2 has never worked for me, not sure if this is something unique to the C dll.

You're welcome

Update on this: I tried your code tonight and I couldn't get it to work beyond the first target either. I'm pretty sure it has to do with which quadrant the angle falls into when its returned and the need to process it differently depending on that. I've come up with a simpler solution which may or may not be suitable depending on exactly what you are trying to do with this but it relies on the PointEntity rate setting to produce the gradual curved change of direction. The codes below but in C++. You would probably need to modify the rate and movement values by app speed to adjust for frame rate changes. There is a small vid of it in action here, I had to zip it up as we don't seem to have a way of loading vids unless its on you tube: Move from point to point.zip // ==================================================================== // This file was generated by LEBuilder // http://leadwerks.com/werkspace // ==================================================================== #include "engine.h" #include <iostream> #include <string> #include "Framewerk.h" using namespace leadwerks; const int ScreenWidth = 800; const int ScreenHeight = 600; const char* AppTitle = "Test"; void ErrOut( const std::string& message ) { std::cerr << message << std::endl; } int main( int argn, char* argv[] ) { if( !Initialize() ) return 1; // Set graphics mode if( !Graphics(ScreenWidth,ScreenHeight) ) { ErrOut( "Failed to set graphics mode." ); return 1; } SetAppTitle( AppTitle ); // Create framework object and set it to a global object so other scripts can access it TFramework fw = CreateFramework(); if( fw == NULL ) { ErrOut( "Failed to initialize engine." ); return 1; } // Set Lua framework object SetGlobalObject( "fw", fw ); // Set Lua framework variable BP lua = GetLuaState(); lua_pushobject( lua, fw ); lua_setglobal( lua, "fw" ); lua_pop( lua, 1 ); // Get framework main camera TCamera camera = GetLayerCamera( GetFrameworkLayer(0) ); PositionEntity( camera, Vec3(0,9,0) ); // Create cube to steer TMaterial material = LoadMaterial( "abstract::cobblestones.mat" ); TMesh mesh = CreateCube(); EntityColor(mesh,Vec4(1,0,0,1),1); PositionEntity( mesh, Vec3(0,0,0) ); // Create an TMesh array for the targets TMesh targetMesh[3]; // Create the targets targetMesh[0]=CopyEntity(mesh,0); ScaleMesh(targetMesh[0],Vec3(0.5)); EntityColor(targetMesh[0],Vec4(0,0,1,1),1); PositionEntity(targetMesh[0], Vec3(-2,0,3)); targetMesh[1]=CopyEntity(targetMesh[0],0); PositionEntity(targetMesh[1], Vec3(3,0,2)); targetMesh[2]=CopyEntity(targetMesh[0],0); PositionEntity(targetMesh[2], Vec3(3,0,-3)); // Point the camear down at the main mesh PointEntity(camera,mesh); // Create ground TMesh ground = CreateCube(); ScaleEntity( ground, Vec3(10,1,10) ); PositionEntity( ground, Vec3(0,-2,0) ); PaintEntity( ground, material ); // Add some light TLight light = CreateDirectionalLight(); RotateEntity( light, Vec3(45,45,45) ); int targetNumber = 0; float targetDistance; float angle; TEntity piv1 = CreatePivot(); bool move = true; // Loop while( !KeyHit() && !AppTerminate() ) { if( !AppSuspended() ) { if(move) { // Poisition pivot at main cube PositionEntity(piv1,EntityPosition(mesh,1),1); // Point it at the current target PointEntity(piv1,targetMesh[targetNumber],3,0.001); // Calc the angle angle = EntityRotation(piv1).Y - EntityRotation(targetMesh[targetNumber]).Y; // Calc the distance to the target targetDistance = EntityDistance(mesh,targetMesh[targetNumber]); // If within 0.25 units then set up the next target if(targetDistance < 0.25) { targetNumber++; if(targetNumber > 2) { move = false; } } RotateEntity(mesh, Vec3(EntityRotation(mesh,1).X,angle,EntityRotation(mesh,1).Z),1); MoveEntity(mesh,Vec3(0,0,0.001),0); } UpdateFramework(); RenderFramework(); Flip(0); } } return Terminate(); }

Just an observation but you quoted: but in your code you are doing the opposite, that is you are dividing your distance by the angle: steerangleai[1] = -(hypdist/Angrot) I can understand the angle divided by the distance as it will progressively approach the final angle as the distance gets less (closer) although presumably you would want to clamp the distance value at 1 once it goes lower than that. I take it you are aiming for a curved movement towards the target? I've not tried the code so I don't know if this is the issue or not, it just was an obvious discrepancy between your code and your description! Also, are you creating new pivots in every iteration of the game loop, you really only need to create them once outside of the loop and then reposition them each time!

Interesting article Dan. I wonder if game companies could turn it completely around and treat it like advertising. You could play various companies off against each other to pay to have their car featured in the game

Extending the laws of your own country to those of others is to make huge assumptions and the general deciding factor in many of these types of cases is not whether it is legal or not but whether you can afford to fight it should the employer take you to court. You will undoubtedly have lost your job by then and depending on the country may not have access to state funding to fight such a case. I doubt a company would do this over a couple of shaders but the point stands! I'm sure Lord Hippo will come to whatever conclusions and decisions are right for him, he is obviously an intelligent person and more than capable of weighing these things up for himself. At least he has shown us some possibilities with shaders and as such it could be done by others.

Yeah, Mixamo is pretty smart technology but I'm not sure that it allows you to move the joints in relation to the overall skeleton, rather just set some reference points for their software to do a best fit of one of their stock skeletons. It is really clever stuff though and yes I'm sure Josh could do it too if he put his mind to it, but I suspect he's a little busy right now with some other projects Not sure how Mixamo would cope with a three legged Garthanox from the planet Fling with its two tails and six arms though