Commit by micah :: r5162 hdr/encoding/opengl/ (4 files in 2 dirs):

Poke around at the encoding some more. It still isn't encoding useful data correctly, but it's creating valid blocks. Also added a background, and a Pixel MHz meter.

Commit by micah :: r5157 hdr/encoding/opengl/ (data/test.frag data/test.vert main.c):

Add some relatively cheesy mouse-driven object spinning capabilities

Commit by micah :: r5154 hdr/encoding/opengl/ (data/test.frag data/test.vert main.c):

Yay, it's using shaders now.. very boring shaders

Commit by micah :: r5151 hdr/encoding/opengl/ (10 files in 2 dirs):

Add another new model, and start amassing the GLSlang utilities. Now there's a relatively terse way to compile and link shaders using variadic shader_link_program() and shader_compile_from_files() functions.

Commit by micah :: r5148 hdr/encoding/ (19 files in 6 dirs):

Rearrange this lots, into directories for offline, opengl, and hardware based implementations. Fix lots of stuff in the OpenGL test, and add a few more models.

Commit by micah :: r5144 hdr/encoding/shader/ (4 files):

Yay, a proper bunny exported using blender

Commit by micah :: r5143 hdr/encoding/shader/ (GNUmakefile Makefile ply2mesh.py test.c):

Add a broken PLY importer that will soon be replaced by some blender-fu

Commit by micah :: r5142 hdr/encoding/ (decode_screenshot.sh shader/demo_frag.cg):

Encode some vertical bars in the fragment shader until we have something more interesting, and add a shell script that takes a screenshot, decodes it, then shows the results.

Commit by micah :: r5141 hdr/encoding/shader/demo_frag.cg:

Fix some saturation and off-by-less-than-one errors

Commit by micah :: r5113 hdr/encoding/shader/ (5 files):

Add a fragment shader that almost implements our experimental HDR encoding scheme. I should really go to work now

Commit by micah :: r5112 hdr/pcb/ (5 files):

Moved the column drivers to the other side of the PCB- there was no way the column and row drivers would both fit on the same edge. Added MOSFET protection resistors and the 1-of-16 decoder for column drive.

Commit by micah :: r5111 hdr/pcb/ (3 files):

Placed and routed the column drive transistors

Commit by micah :: r5110 hdr/pcb/ (5 files):

Rerouted the LED matrix to leave more room after determining that the traces didn't need to be as thick as I though they did. Placed and routed all row drive transistors and power resistors.

Commit by micah :: r5107 hdr/pcb/ (9 files):

Start work on a backlight module PCB for the HDR display. Three of these boards form the active backlight for a 15" display. Each consists of a 16*18 288-LED matrix arranged in a hexagonal grid, MOSFETs for row and column drive, and demultiplexing logic.

Commit by micah :: r5105 hdr/ (5 files in 2 dirs):

Start on some random things inspired by high dynamic range display devices. This is an encoding that could allow high dynamic range content to be presented within existing GUI environments and even in a way backward compatible with LDR displays. The least significant bit of each color channel is sacrificed, and used to encode an extra 16-bit plane at 1/4 the spacial resolution. The extra bits are used for reliable detection of this encoded data, so it can be separated from the LDR content on the same screen. With some tweaking, these algorithms could be implemented in hardware in real-time.