On 10/07/2010 01:05 AM, Edouard Chalaron wrote:
> Hi Frans
>
> Well, it worked straight out of the box for me....(to my surprise I
> must say) on 11.3 AMD 64 bits and intel 32 bits.
> It has to do with the (to my view) ridiculous amount of different x264
> available.
> Try
> make distclean
> first then
> ./configure --disable-mmx --enable-opengl
> then make
>
> However I have the same error on 11.2 64 bits and on a 11 64 bits.
> Meaning that I will soon update these boxes.
> Cheers
> E
>
>
>
> --- On *Thu, 7/10/10, Frans de Boer /<[email protected]>/* wrote:
>
>
>     From: Frans de Boer <[email protected]>
>     Subject: [CinCV] Compile error
>     To: [email protected]
>     Date: Thursday, 7, October, 2010, 11:37 AM
>
>     Using OpenSuse 11.3 and the the .configure /without/ any options
>     results in:
>
>     /bin/sh ../libtool --tag=CC --tag=CC   --mode=compile gcc
>     -DHAVE_CONFIG_H -I. -I.. -I../libmpeg3   -D_LARGEFILE_SOURCE
>     -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -DHAVE_MMX -DUSE_MMX
>     -DX86_CPU  -DHAVE_FIREWIRE      -I/usr/include/mjpegtools
>     -I/usr/include/mjpegtools/mpeg2enc
>     -I/usr/include/mjpegtools/mplex   
>     -DENCORE_INCLUDE=\"encore50/encore.h\" -I.../quicktime/ffmpeg -g
>     -O2 -MT qth264.lo -MD -MP -MF .deps/qth264.Tpo -c -o qth264.lo
>     qth264.c
>     libtool: compile:  gcc -DHAVE_CONFIG_H -I. -I.. -I../libmpeg3
>     -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
>     -DHAVE_MMX -DUSE_MMX -DX86_CPU -DHAVE_FIREWIRE
>     -I/usr/include/mjpegtools -I/usr/include/mjpegtools/mpeg2enc
>     -I/usr/include/mjpegtools/mplex
>     -DENCORE_INCLUDE=\"encore50/encore.h\" -I../quicktime/ffmpeg -g
>     -O2 -MT qth264.lo -MD -MP -MF .deps/qth264.Tpo -c qth264.c  -fPIC
>     -DPIC -o .libs/qth264.o
>     qth264.c: In function ‘encode’:
>     qth264.c:245:4: warning: passing argument 1 of ‘x264_nal_encode’
>     from incompatible pointer type
>     /usr/include/x264.h:456:6: note: expected ‘struct x264_t *’ but
>     argument is of type ‘unsigned char *’
>     qth264.c:245:4: warning: passing argument 2 of ‘x264_nal_encode’
>     from incompatible pointer type
>     /usr/include/x264.h:456:6: note: expected ‘uint8_t *’ but argument
>     is of type ‘int *’
>     qth264..c:245:4: warning: passing argument 3 of ‘x264_nal_encode’
>     makes pointer from integer without a cast
>     /usr/include/x264.h:456:6: note: expected ‘struct x264_nal_t *’
>     but argument is of type ‘int’
>     qth264.c:245:4: error: too many arguments to function
>     ‘x264_nal_encode’
>     /usr/include/x264.h:456:6: note: declared here
>     qth264.c:242:14: error: void value not ignored as it ought to be
>     make[3]: *** [qth264.lo] Error 1
>     make[3]: Leaving directory
>     
> `/media/raidarray/fdb-data/projects/linux/multimedia/cinelerra/git-cinelerra/quicktime'
>     make[2]: *** [all-recursive] Error 1
>     make[2]: Leaving directory
>     
> `/media/raidarray/fdb-data/projects/linux/multimedia/cinelerra/git-cinelerra/quicktime'
>     make[1]: *** [all-recursive] Error 1
>     make[1]: Leaving directory
>     
> `/media/raidarray/fdb-data/projects/linux/multimedia/cinelerra/git-cinelerra'
>     make: *** [all] Error 2
>
>     Same result with various options.
>
>     As I asked before: advancements in ffmpeg and x264 are not
>     incorporated in the SVN repository? If so, it makes cinelerra only
>     usable on old systems without any recent update. I use x264
>     version form git.
>     Maybe I am missing something?
>
>     Frans.
>
Nope that is not working either. Like I said: I am using the latest git
version. However, about 2-3 months ago I had the same problem and nobody
reacted. One year ago I had a solution for another (mask) problem. Did
anybody incorporated it? Nope again.
This evening I tried again, to no avail.
I guess cinelerra is dead in the water and everybody is waiting for
Lumiera or the 4.x version (which they do already for the past 2-3
years). After all, nobody seems to maintain compatibility with x264 and
ffmpeg!?

Maybe I should look for another tool despite the fact that I love
cinelerra rich features (not its (in)stability ;))

Frans.

Reply via email to