YUV422 semi-planar " UV plane" start address

David said:

Does this means that the " UV plane" start address is "src+720*576".

I believe it will be 'src+736*576' as pitch should be 32 bytes aligned.

Hi ，Nagabhushana 

        Thanks very much for your reply  .But I can't get the right UV plane start address in my dm6467 codecs videc_copy,the buffer is defined as below:

XDAS_Int8 * getptr;

XDM_BufDesc *inBufs;

getptr=*(inBufs->bufs);

YArray=getptr;

CbArray=YArray+ImaSize;//ImaSize=736*576;

Then I fwrite the buffer to flash ,I watch the data with "Rawplater/yv12",what I see is a black and white picture without multicolour .Do you know the reason?

thanks again.

Regards,

-David

Hi

      I add some print information and fond that on dsp side the UV plane value is all"0xffffff80,0xffffff80,0xffffff80....." and on arm side is all",0x80 ,0x80 ,0x80 .....",I haven't changed the default UV plane value .The UV plane is provided by dm6467 vpif  .And I use the dvsdk3.1 dm6467 encode  demo,I have no idea what happened,can any one help me .

Regards,

-David 

Which player(application) you used to test yuv420sp? Can you please send me the link? Application supports yuv420sp?

Also, please attach file captured, let me check with application here.

Hi,Nagabhushana 

     The picture is paste below:

I could not get rawplayer from google search. You please provide link. Also, I never found a application which displays yuv420sp format video files. I believe the application is displaying only 'Y' buffer. That is why you are seeing black and white.

Hi

I have insert the player. 

5861.RawPlayer.2.2.222.rar

Regards,

David

Hi,all

         Below is the raw yuv420PSEMI data .If you open it with some binary watch tool ，you will find that the data from address “720*576=414720”  to the end is close to 0x80(the whole UV plane).Can anyone tell me how this happens.Dose this  has something to do with VDCE.And I can see the correct color image when I use the "encodedecode "demo with a “adv7343” D/A transfer chip .The raw yuv420PSEMI data mentioned above is produced by  the write thread of dm6467 “encode” demo.

1004.h.rar

Waiting for your help,thanks in advance.

Regards,

-David