Dear All,
We don't get any reply for the previous mail. Please reply if any one
has any idea.
If this is not modified by DirectFB developers, can any one tell me
where and what should we do to fix this.
For YUV format, the colour value is shared between pixels. But this is
not considered on rotation of surfaces. We think this is the cause of
colour change in YUV surface rotation. This is our assumption only.
Thanks a lot in advance!
Regards,
Lindo
------------------------------------------------------------------------------------
Dear Niels/All,
Thanks for your replay.
We are using DSPF_UYVY surface to load an image. Our hardware decoder
is capable of decoding YUV format faster. So we are interested in
this format.
1. Is modification to fix this will be in the up coming DirectFB
release?
2. If yes, when can we expect that?
Thanks & Regards,
Lindo Lonappan.
-------------------------------------------------------------------------------------
Niels Roest wrote:
(forwarded to users list only)
Hi Lindo,
the fact that the colour changes is a bug, I think.
In principle, DirectFB does not store "vertically sampled" surfaces.
This means that the surface is turned when it is blitted.
Storing the information vertically is in general not very sensible
since the DMA acceleration engines and back-end HW work line-based, so
horizontally. If you are worried about lost precision you can always
convert to a spatial losless format such as AYUV or RGB first. You
will have lost colour resolution anyway if you do blits with an odd
step, such as moving something from x=0 to e.g. x=191.
hth
Niels
Lindo Lonappan wrote:
Dear all,
We are working with DirectFB-1.4.2.
For rotation of images we are using IDirectFBSurface::Blit(). When
the pixelformat of the surfaces are DSPF_ARGB, the rotation is
working fine. But if the pixelformat of source and destination
surface are DSPF_UYVY, after rotation there is a colour change in image.
1. Is the DirectFB support rotaion of DSPF_UYVY surface?
2. Can we create a DirectFB surface by specifying that the sub
sampling is vertical, so that the sharing of the chroma values
takes place vertically rather than the normal horizontal? We are
referring to YUV422 surface.
Please help us if any one has any idea.
Thanks a lot in advance!
Thanks & Regards,
Lindo Lonappan.
------------------------------------------------------------------------
_______________________________________________
directfb-dev mailing list
directfb-dev@directfb.org
http://mail.directfb.org/cgi-bin/mailman/listinfo/directfb-dev
_______________________________________________
directfb-dev mailing list
directfb-dev@directfb.org
http://mail.directfb.org/cgi-bin/mailman/listinfo/directfb-dev
