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.

VPBE vertical expansion (6/5) of 420 format - colour problem

Hi,

I am using DM365 processor with WinCE. I am trying to expand video 736Wx576H vertically in 420 format in VPBE with ratio (6/5). VideoWindow0 is used for Y and Video1 is used for C. In memory: Y data are first, after that follows C data. When I try to set VVRSZ=1 (BIT11) in MODE (vpbe OSD) register, I can get expansion 6/5, but the colours are misplaced. It looks like window1 was scaled with a higher ratio. Please can you confirm that the expansion should work in this scenario? If so, can you advise what else I need to check in registers settings?

Thank you.

  • Hi Tibor,
    Is it working without expansion?
    From the VPBE user guide the Video windows support only YCbCr .
  • Hi dibin ms,

    Yes, it is working without vertical expansion. As I described above, I am using YCbCr format – CY420, the memory location for Y data is set for VIDWIN0 and CbCr for VIDWIN1. The MISCCTL[BIT14] is set to 1. I was trying to follow section 4.4.3 in sprufg9c.pdf. Please see attached picture of the video snapshot. You can see 2 green magnets on the white board. One is in the upper position and the other one in the lower position. There is a green round shadow that carries colour information (CbCr) of the magnet. It covers larger part of the upper magnet, but no part of the lower magnet (magnet is all black). The further down you go, the misalignment between the magnet and colour increases.

    Is there any other register settings, which are not mentioned in the manual that I need to look at to correct this? Can you please do the same setup on your PC and confirm it works so I could compare my register settings and see what the problem is?

    Thanks.

  • Hi Tibor,
    I am sorry , I dont have the setup to test.
    I am suspecting different formula is applied to Y and CbCr data for expansion .
    There is one bit MODE.EF did you try changing this bit ?

    Regards,
    Dibin
  • Hi Dibin,


    I did try to set MODE.EF register to 0 and 1 but the result is the same. I spent many days modifying VPBE OSD registers, which are logically related to the problem, but didn’t succeed.
    If you can’t simulate the same scenario on your PC, what would be your other suggestion to tackle this issue?

    Thanks,
    Tibor