by After using my compiled version of mpeg_encode 1.5b, using the precompiled
version of mpeg_encode 1.3, I can only encode mpgs that xanim barfs on and
claims HUFF ERR repeatedly, and the view is really messed up (big blocks)
I've made my mpegs from both tgas using the INPUT_CONVERT tgatoppm
I've converted them by hand and used NO convert int he program
here's my param file
# any line beginning with '#' is a comment
# no line should be longer than 255 characters
# general format of each line is:
# <option> <spaces and/or tabs> <value>
# lines can generally be in any order
# an exception is the option 'INPUT' which must be followed by input
# files in the order in which they must appear, followed by 'END_INPUT'
#
# Also, if you use the `command` method of generating input file names,
# the command will only be executed in the INPUT_DIR if INPUT_DIR preceeds
# the INPUT parameter.
#
# <option> MUST be in UPPER CASE
#
PATTERN IBBPBBPBBPBBPBBP
OUTPUT output.mpg
# mpeg_encode really only accepts 3 different file formats, but using a
# conversion statement it can effectively handle ANY file format
#
# You must specify the type of the input files. The choices are:
# YUV, PPM, JMOVIE, Y, JPEG, PNM
# (must be upper case)
#
BASE_FILE_FORMAT PPM
#
# if YUV format (or using parallel version), must provide width and height
# YUV_SIZE widthxheight
# this option is ignored if BASE_FILE_FORMAT is not YUV and you're running
# on just one machine
#
YUV_SIZE 352x240
# If you are using YUV, there are different supported file formats.
# EYUV or UCB are the same as previous versions of this encoder.
# (All the Y's, then U's then V's, in 4:2:0 subsampling.)
# Other formats, such as Abekas, Phillips, or a general format are
# permissible, the general format is a string of Y's, U's, and V's
# to specify the file order.
INPUT_FORMAT UCB
# the conversion statement
#
# Each occurrence of '*' will be replaced by the input file
#
# e.g., if you have a bunch of GIF files, then this might be:
# INPUT_CONVERT giftoppm *
#
# e.g., if you have a bunch of files like a.Y a.U a.V, etc., then:
# INPUT_CONVERT cat *.Y *.U *.V
#
# e.g., if you are grabbing from laser disc you might have something like
# INPUT_CONVERT goto frame *; grabppm
# 'INPUT_CONVERT *' means the files are already in the base file format
#
INPUT_CONVERT *
# number of frames in a GOP.
#
# since each GOP must have at least one I-frame, the encoder will find the
# the first I-frame after GOP_SIZE frames to start the next GOP
#
# later, will add more flexible GOP signalling
#
GOP_SIZE 16
# number of slices in a frame
#
# 1 is a good number. another possibility is the number of macroblock rows
# (which is the height divided by 16)
#
SLICES_PER_FRAME 1
# directory to get all input files from (makes this file easier to read)
INPUT_DIR input
# There are a bunch of ways to specify the input files.
# from a simple one-per-line listing, to the following
# way of numbering them. See the manual for more information.
INPUT
# '*' is replaced by the numbers 01, 02, 03, 04
# if I instead do [01-11], it would be 01, 02, ..., 09, 10, 11
# if I instead do [1-11], it would be 1, 2, 3, ..., 9, 10, 11
# if I instead do [1-11+3], it would be 1, 4, 7, 10
# the program assumes none of your input files has a name ending in ']'
# if you do, too bad!!!
#
#
nccmain_*.ppm [0-40]
# can have more files here if you want...there is no limit on the number
# of files
END_INPUT
# Many of the remaining options have to do with the motion search and qscale
# FULL or HALF -- must be upper case
PIXEL HALF
# means +/- this many pixels for both P and B frame searches
# specify two numbers if you wish to serc different ranges in the two.
RANGE 10
# this must be one of {EXHAUSTIVE, SUBSAMPLE, LOGARITHMIC}
PSEARCH_ALG LOGARITHMIC
# this must be one of {SIMPLE, CROSS2, EXHAUSTIVE}
#
# note that EXHAUSTIVE is really, really, really slow
#
BSEARCH_ALG CROSS2
#
# these specify the q-scale for I, P, and B frames
# (values must be between 1 and 31)
# These are the Qscale values for the entire frame in variable bit-rate
# mode, and starting points (but not important) for constant bit rate
#
IQSCALE 8
PQSCALE 10
BQSCALE 25
# this must be ORIGINAL or DECODED
REFERENCE_FRAME ORIGINAL
# for parallel parameters see parallel.param in the exmaples subdirectory
# if you want constant bit-rate mode, specify it as follows (number is bits/sec)
:
BIT_RATE 1000000
# To specify the buffer size (327680 is default, measused in bits, for 16bit wor
ds)
BUFFER_SIZE 327680
# The frame rate is the number of frames/second (legal values:
# 23.976, 24, 25, 29.97, 30, 50 ,59.94, 60
FRAME_RATE 30
# There are many more options, see the users manual for examples....
# ASPECT_RATIO, USER_DATA, GAMMA, IQTABLE, etc.
If you have any suggestions, please e-mail a reply. Thanks
(I've had too many people whine and complain that their web browsers couldn't
read .fli files. And I've gotten tired of explaining how to correct that.
So I thought I'd transition over to mpegs)