Jump to content

csteaderman

Members
  • Posts

    6
  • Joined

  • Last visited

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Joel, Thanks for the followup. I will wait to hear what you are able to get. Thanks.
  2. Thanks for the pointers, PIXMAP is just what I was looking for. Do PIXMAPs natively support alpha blending? Right now I am using one PIXMAP to draw (blit) a background image and another PIXMAP to blit a PNG with transparent background on top of the background. Unfortunately, the background color of the top PIXMAP is rendered as black.
  3. The Image is a PNG file, loaded in ROMFS. Will gdispImageCache improve rendering time in initial draw or only subsequent drawing? I have multiple MB RAM, so no problem. Pixmap was next on my list of implementation capabilities.
  4. My target hardware is an Atmel SAMA5D2 with built in display controller and an 800x480 pixel display. I have chosen the Framebuffer model for the display driver. The hardware controller supports up to 4 layers, but I am only using one right now. When I integrate Pixmap support, I will take advantage of the multiple layers. At this time I have implemented the sample benchmark program and am getting a little more than 10M pixels per second drawing random size/color boxes. I am now trying to render a full screen bitmap as the background. The problem is that I can see the image rendered row by row. Is there a configuration setting to change the buffer size that is decoded and written to the display so that more of the bitmap is rendered per call? Is there an event/callback that I can hook in order to know that uGFX is done rendering the bitmap so that I can buffer the smaller writes and then quickly switch the DMA buffer? Any pointers would be appreciated. Thanks, Charlie
  5. Thanks so much for the pointer. The final solution was even easier because the code to handle the 3 byte per pixel buffer had already been written in some support code from Atmel. I simply needed to modify the uGFX framebuffer calls to call the Atmel code. Now on to optimizations.
  6. I'm bringing up a display connected to an Atmel SAMA5D2 Xplained evaluation board. I assumed that the display was a framebuffer as Atmel's sample code assigns a buffer for the display that gets sent to the display. It is setup for 8 bits per pixel, but the framebuffer expects each pixel to be represented by 3 bytes, rather than the 4 bytes that uGFX seems to want to allocate. Is there a way for me to configure uGFX to represent each pixel by exactly 3 bytes (24 bits)?
×
×
  • Create New...