smishra

Thanks

I ended up doing a custom post filter after rendering using built in postfilter. Textures 9 and 10 seem to work for passing textures from the Lua program to custom shader. SetBuffer(opticsBuffer) fw:Render() SetBuffer(BackBuffer()) SaveBuffer(opticsBuffer,"ob.jpg",85) filter_nvg = LoadShader("abstract::nvg.vert", "abstract::nvg.frag") SetShader(filter_nvg) BindTexture(GetColorBuffer(opticsBuffer,0), 9)

Thanks for your suggestions. Right now, approach 1, replacing a couple of unused postfilter effects with custom effects seems the path of least resistance. However it is a hack. In some scenes the designer may want to use the effects that I have overwrriten. We use a few of the built in post filter effects, so replacing the entire shaders.pak will not be effective. In approach 3, is there an easy way to access the CurrentBuffer() in the shader. I tried getting the current buffer in Lua and binding it, but got an Exception error. In the long term, making the Framewerk more flexible may be the best option. One way would be to make the list of postfilters an expandable one, that can be added to from Lua itself. If it is not too much of a performance drag, coding the renderer in Lua might work out.

I had earlier created post filter effects in BlitzMax by modifying the Framework Renderer and they work great. I have since got hooked onto the editor and the ability to run the environment directly in the editor. This is of great help to the scene designers. In order to design good scenes which employ custom post filter effects, we need to be able to run the post filter effects from the editor. Is it possible to add custom post filter effects using lua. The Framework is now compiled into the engine. If there ware 2-3 slots available for custom shaders and custom filters it would be great. We could enable custom filter effect, the same way as we enable other filter effects. By default these custom filters would do nothing.

Thanks. I will look at dropbox. Lumooja, I was also thinking of subversion. Any pros and cons of subversion versus git. In contrast to normal code sync issues the files are mostly binary and often quite large. Since my clients are already familiar with subversion, I am going to set up a subversion repository and have the clients use Tortoise SVN.

We are doing a limited internal release of a Leadwerks 2.4 based application. The first version is out and was a download of about 450MB in the form of a compressed zip file of the entire folder of assets and the application executable. Of course we have made changes and want to distribute an updated copy of the application. What is the best way to do this in a more or less fool proof way? Ideally we would like to make frequent changes in response to their requests. The users are all over the country. The most fool proof way would be to make another 450MB upload of the latest and have the users download them. However this becomes cumbersome when we have frequent updates. Another way would be to use a software like rsync. We could set up a password protected folder on a server and have them sync up their local copies to the copy we put online. In this case the software updater would be a generic client. Updates from the client would be forbidden on the server. We would also want to avoid overwriting any local custom files they would have on their local server. However we would want local application files that are no longer needed to be deleted. Any other ideas?

One way may be to perform two queries 1. fast, simple query to eliminate impossible objects where x_any_object - x_ref > range and z_any_object - z_ref > range The results of this can be stored in a temporary, in memory table and then used to run the second query 2. Of the smaller set of objects that are left perform ((x_any_object - x_ref) * (x_any_object - x_ref) + (z_any_object - z_ref) * (z_any_object - z_ref)) < range * range A custom, hand optimized implementation of data structures and implementation will always be faster than something built on a general abstraction like a database layer. The question is the amount of time available for data structure specific optimization versus the general features brought in by a DB layer like sqlite --------------------- 3. Of course if these are common functions, one would eventually expect custom hardware to support this function. Wait this is what graphics cards do. :-) Maybe a better chipset/GPU architecture might help these sort of computations!

It is designed in US but manufactured in China, as far as I know.

The link does not seem to work. Has the file been moved? Thanks Sanjay