On 07.02.2017 8:05, Dan Dennedy wrote: > On Mon, Feb 6, 2017 at 9:13 AM Maksym Veremeyenko <ve...@m1stereo.tv > <mailto:ve...@m1stereo.tv>> wrote: > > i just discovered, that i dont clear understand a field-based > processing, so first question *why does next properties needed*: > > consumer_tff > > > This is a frame property that indicates what field order the consumer > wants. You set "top_field_first" property to 0 or 1 on the consumer, and > mlt_consumer copies to the frame as consumer_tff. The consumer defaults > to top_field_first=0 (because it is not set and mlt_properties_get_int() > returns 0 if unset). > > consumer_deinterlace > > > This is a frame property that indicates that the consumer wants > progressive output. You set the "progressive" property to 0 or 1 on the > consumer, and mlt_consumer copies it to the frame as > "consumer_deinterlace" (this name for legacy reasons). This defaults to > the mlt_profile.progressive value. > > top_field_first > > > The current state of the field order of the image on the frame. > > meta.top_field_first > > > The field order from the producer, which can be different than current > state or what the consumer requested. This is not currently set by any > producer. The framework automatically copies all properties with prefix > "meta." on a producer to the frame. So, this lets the user add > "meta.top_field_first" to the producer to override the "top_field_first" > property a producer sets on a frame. > > meta.swap_fields > > This provides another way for a user to tell the fieldorder filter to > swap fields (exchange lines). From the yml: > If you set the property meta.swap_fields=1 on the producer, then this > filter swaps the fields of an interlaced frame in addition to any field > order correction by shifting the image.
thanks for explanation - that makes some my issues clear... page with this text at https://www.mltframework.org/docs/ will be helpful for me and others.... > It is doing field-based rendering, meaning the interpolation of the > geometry can factor in time and motion from one field to the next within > a single frame's image (writing both fields interleaved within a single > image). i understand it clear, i rather was interesting in keywords *Assuming lower field first*. i definitely though all processing was as lower-field-first for historical reason, like YUYV. -- Maksym Veremeyenko ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, SlashDot.org! http://sdm.link/slashdot _______________________________________________ Mlt-devel mailing list Mlt-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mlt-devel
