A discussion in that D-I-Y group
On 27/12/17 13:49, Max Demian wrote:
On 27/12/2017 10:45, Brian Gaff wrote:
Has probably deliberately been winding me up as several folk over there do
not like AD subtitles or in the case in point, signing.
The problem as I see it from reading some web sites is this. If bandwidth
were not an issue, it would be easy to add a signed video track to the
output of a channel and hide it when its not wanted, but because these days
at least, bandwidth is seemingly cut to the bone, this is not done so one
gets the ludicrous situation where its actually cheaper on bandwidth to
leave the signer on screen all the time!
IE its part of the transmission much as AD is on the I player etc,but of
course this would not be tolerated by the average viewer, so any viewer who
needs the signing has to wait for the repeat to see the signed version,
which also ****es of the viewer who finds the signer distracting.
There really should be enough bandwidth headroom given for signing in my
Signing is bound to use a lot of bandwidth, unless a simplified
silhouette of a pair of hands is used, which can be superimposed on the
existing video as it is with subtitles. I suspect that people who use
sign language are used to have an actual person in front of them and
rely on facial expressions.
Why should signing use any more bandwidth than a picture without? The
screen has a certain number of pixels (whether SD or HD), and those
pixels will have a fixed number of bits applied to them. In fact, the
"signed" programmes I have seen have a smaller picture to accommodate
the signer and AFAIR there is a bit of wasted space at the bottom and
RHS where the signer appears. That is uniform in colour and brightness,
so I assume should lend it self to better compression than the rest of
the picture, so a signed programme should use fewer bits overall than a
I'm sure I'm wrong - but would appreciate a clear explanation!