• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 3 answers
  • Subscribers 30 subscribers
  • Views 1473 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DLPC350: 8bit pattern sequence result makes border

Uryyy
Expert 1890 points
Part Number: DLPC350

I made a test picture which is gradation from white(0xffff) to black(0x0000). And, this source file newly stored in flash.

And this picture was displayed by pattern sequence 8bit depth. The displayed result looks like something of border as shown below. Please let me know the cause root or what was happened.

  • 0
    Prodigy 140 points
    Hi Uryyy,

    In my mind 2 things come. One is that this is just a Craik O'Brien illusion. This would perceptually enhance the edges of intensity increment blocks. If you have a photodiode maybe you can try to measure if the intensity keeps on increasing and this is just an illusion. If not then looks like actual downward dips. Not sure what might be causing that but to correct it you would have to measure the intensity values from your black to white with decent increments and then use a color look up table.
  • 0 in reply to Saad Idrees
    Expert 1890 points
    Hello Saad,
    Thank you. I'll measure actual luminance with photo diode.
  • 0 in reply to Uryyy
    TI__Mastermind 28005 points
    Thanks for your input, Saad.

    Uryyy, have you tried this with other pattern options? I recommend using a simpler ramp that does not go as high, in case of clipping issues.

    Best Regards,
    Philippe Dollo
  • 0 in reply to Philippe Dollo
    Expert 1890 points

    ffff-0000_gradation.zip

    Hello Philippe,

    Are the clipping issues specification with this source bmp(Please find attached bmp, you can reproduce)?  Please tell me the mechanism.

  • 0 in reply to Uryyy
    TI__Mastermind 28005 points
    Uryyy,

    When I can find some time I will try to reproduce. However, please pardon delay.

    In the meantime please test with alternative patterns. This appears to be tall image so your input resolution is incorrect. You should rule out whether the image source itself is the issue here.

    I hope this helps.

    Best Regards,
    Philippe Dollo
  • 0 in reply to Philippe Dollo
    Expert 1890 points

    Philippe,

    test_files.zip

    My test picture's bmp is recreated as below.(8bit [R7 R6...R0] → Add to Output File, 8bit [G7 G6...G0]→ Add to Output File, 8bit [B7 B6...B0]→ Add to Output File) I also attach the bmp and bin which is embedded as #3 flash.

    Pattern sequence settings are as follows.

    Flash Index 3 / 8bit R : Add pattern sequence → 8bit G : Add pattern sequence → 8bit B : Add pattern sequence

    I can also see flicker at this test.

  • 0 in reply to Uryyy
    TI__Mastermind 30875 points

    Hi Uryyy,

    If you are seeing the flickering naked eyes, this is expected, note the content displayed in pattern sequence is different from video mode. The video sequences are optimized for reducing the flickering artifact and at the same time they are not 100% pixel accurate. In case of pattern sequence they are pixel accurate and intended for structured/machine vision kind of system application. As an example - Video sequence displaying WXGA 1280x800 24bpp @ 120Hz where as if you apply pixel accurate, to display 8-bit gray scale input it requires almost same time 8.3ms, in video mode we display 24bit RGB888 in same 8.3ms window.

    Regards,

    Sanjeev