aboutsummaryrefslogtreecommitdiff
path: root/graphics.h
Commit message (Collapse)AuthorAgeFilesLines
* render: improve line raster (not fixed yet). also tag jobs with an id.HEADmasterrodri36 hours1-1/+2
|
* add a name property to Material.rodri6 days1-0/+1
|
* add a general primitive with support for points, lines and triangles.rodri2024-05-031-8/+7
| | | | | | | | also got rid of the dependency on OBJ for the entire renderer, instead letting the user load a Model from any given OBJ. this modularity will allow for other formats to be used in the same way, relying on a single, internal representation for the entire pipeline.
* replace the Triangle with a general purpose Primitive.rodri2024-04-161-8/+19
| | | | | this is only an structural replacement and doesn't add support for the other primitive types.
* add some instrumentation to measure pipeline stage time.rodri2024-04-151-0/+15
|
* implement a fully concurrent pipeline based on tiles.rodri2024-04-151-4/+3
| | | | | - got rid of the z-buffer lock to avoid contention. - little improvements to fb.c
* simplify the job scheduler. correct two mistakes regarding the Viewport.rodri2024-04-051-7/+5
| | | | | | | | | | > They made two mistakes. they hanged the wrong man and they didn't finish the job. so Clint Eastwood came back to kick my ass. the mistakes in question were that the Viewport shouldn't know about double buffering, conceptually it has a framebuffer and that's it. the second one was passing it to the renderer, when the renderer couldn't care less about what a viewport is.
* scene: add a way to remove entities and fix the clearscene routine.rodri2024-04-031-0/+1
|
* scene: implement full cleanup procedures.rodri2024-04-031-0/+1
|
* viewport: add a way to access front and back buffers.rodri2024-04-011-0/+1
|
* pass a Viewport in the Renderjob instead of a fb.rodri2024-03-261-1/+2
| | | | | do the frame buffer clearing and swapping as part of the rendering process, not within shootcamera.
* implement a (partially) concurrent pipeline.rodri2024-03-251-4/+27
|
* use the new libgeometry berp routines. add a frame counter to the camstats.rodri2024-03-211-0/+2
|
* make the fshader return a Color instead of a Memimage. clean things up.rodri2024-03-061-5/+4
|
* add a texture sampler with nearest and bilinear routines.rodri2024-03-061-0/+28
| | | | | | pass the material reference along with the vertices. also implemented back-face culling, but it's disabled for now.
* add user-defined vertex attributes (varyings) and improve the interpolation ↵rodri2024-03-031-14/+29
| | | | code.
* have separate routines for drawing and memdrawing.rodri2024-02-271-1/+6
|
* interpolate every vertex attribute when clipping and during rasterization.rodri2024-02-251-2/+8
| | | | added parameters necessary to implement the Phong shading model.
* add initial support for OBJMaterial properties.rodri2024-02-211-1/+5
|
* pass an entire entity to the shader unit. fix a bug in cliptriangle.rodri2024-02-141-1/+1
| | | | | | | | | | an entity is passed instead of a model so we can access its frame of reference to perform the model-to-world transformation. the bug in cliptriangle was due to sd[01] being declared as static which, when rendering multiple entities, was causing d0 = d1, which led to division-by-zero.
* lay out the grounds for a scene renderer.rodri2024-02-131-7/+25
| | | | | | also fixed an issue with cliptriangle() where an entire tri would get discarded if all its vertices were outside the frustum.
* completed homogeneous clipping procedure.rodri2024-02-121-0/+41
|
* improve coordinate transformations and fix projections.rodri2024-01-311-5/+2
| | | | | also got rid of Deco. there's no point in having that, just deal with image(6) files.
* make the vertex shader process actual vertices.rodri2024-01-301-2/+1
|
* import the new renderer and clean things up.rodri2024-01-301-35/+102
| | | | | | | | | | | | i integrated the renderer i've been developing on the tinyrend repo and got rid of a bunch of stuff that's no longer necessary. also began structuring things to fit the new interface i have in mind. there are still some artifacts with the projection xforms that cause issues with clipping and division by zero.
* add a line clipping procedure based on the Liang-Barsky algorithm. (thanks ↵rodri2021-07-181-0/+1
| | | | jmi2k!)
* bring flat triangle rasterization back, with some goodies for future shading ↵rodri2020-06-291-0/+14
| | | | work. add uv coordinates to the vertex.
* transcribe macros into actual functions. remove unused memdraw code.rodri2020-04-281-7/+33
|
* standalone version release.rodri2020-04-171-0/+42