[M4IF Technotes] question about the value of fixed_vop_time_increment

[Christoph Lampert]

On Tue, 1 Oct 2002, Amit Klir wrote:
> If we have the value of the VOP_time_increment_resolution == n , we need n-1 bits for the vop_time_increment representation.
> In this special case the number of bits for the fixed_vop_time_increment is 0.
> 
> According the standard, what is the actual number of bits that need to specify the case of  VOP_time_increment_resolution=1 
> 	0 - since 0 is the minimum between [0 1) ?
> or 
> 	1- since the value 0 is forbidden ?

Since the value 0 is _explicitly forbidden_, the value has to be 1, 
and a 1-bit value  vop-time_increment=0 has to be written to the
bitstream for every VOP, even though this value is redundant. 
The other criterion of vop_time_increment lying in the interval 
[0,vop_time_increment_resolution) is _impossible_ to fulfill in 
this case, so I guess this special case was simply overlooked when 
writing about the interval. 
Christoph Lampert
-- 
Christoph H. Lampert chl   math,uni-bonn,de | Diese Signature wurde maschi-     
Beringstr. 6, Zi. 14, 53115 Bonn, Germany | nell erstellt und bedarf
Tel. (0228) 73-2948  Fax. +49 228 73-7916 | keiner Unterschrift. AZ 27B-6