For those like myself who can not get the Berkeley files to run (and
for the folks who say it works but won't answer how) here is another
source of an encoder. 

	I have not figured it out yet but it did compile and it did run neither
of which would happen on the Berkeley. Hopefully others will try this as
the documentation gives a new meaning to obscure. 

http://www.mpeg.org/MPEG/MSSG/#source

-- 
Giwer's Rule: Aps expand to consume every available
increase leaving a net zero increase in capability. 
	-- The Iron Webmaster, 71

Post a reply to this message

Matt Giwer says...
> 	For those like myself who can not get the Berkeley files to run (and
> for the folks who say it works but won't answer how) here is another
> source of an encoder. 
> 
> 	I have not figured it out yet but it did compile and it did run neither
> of which would happen on the Berkeley. Hopefully others will try this as
> the documentation gives a new meaning to obscure. 
> 
I was curious enough to d/l the lot and look into one thing and another. 
It did compile using Borland's C++Builder and it runs although I have no 
suitable files to process, so it did nothing.
Have you seen that only YUV and PPM input formats are supported? I looked 
them up and they seem to me a bit unusual. I don't think I would have a 
use for all this. First convert TGA into YUV?
The documentation: I'll probably have a look whenever I think I'm 
depressed, because it's nice to see that it can get still worse.

-- 
Regards,  Sander

Post a reply to this message

Sander wrote:
> 
> Matt Giwer says...
> >       For those like myself who can not get the Berkeley files to run (and
> > for the folks who say it works but won't answer how) here is another
> > source of an encoder.
> >
> >       I have not figured it out yet but it did compile and it did run neither
> > of which would happen on the Berkeley. Hopefully others will try this as
> > the documentation gives a new meaning to obscure.
> >
> I was curious enough to d/l the lot and look into one thing and another.
> It did compile using Borland's C++Builder and it runs although I have no
> suitable files to process, so it did nothing.
> Have you seen that only YUV and PPM input formats are supported? I looked
> them up and they seem to me a bit unusual. I don't think I would have a
> use for all this. First convert TGA into YUV?
> The documentation: I'll probably have a look whenever I think I'm
> depressed, because it's nice to see that it can get still worse.

	Pardon but I am feeling dumb after so many posts. 

	I mean I have been over the docs in much detail (and the Berkeley docs
and such) and nothing I try works. I noted what you say and have tried
it. I have noted what everyone has said and have tried it, mainly those
who say it works for them. 

	Not to brag but to make a point, my first Fortran program ran 33 years
ago. Maybe I am getting senile and missing something very obvious. I
just have no idea what my problem is. I have not tried TGA to YUV but I
have tried PPM and failed. And I noted the multiples of 16 and I don't
know. 

	At this point I'll take a tar.gz whatever of everything that people
have to does work if anyone is willing to make the effort.

Post a reply to this message

Matt Giwer wrote:

> 
>         I mean I have been over the docs in much detail (and the Berkeley docs
> and such) and nothing I try works. I noted what you say and have tried
> it. I have noted what everyone has said and have tried it, mainly those
> who say it works for them.

But you never gave any clues.  Just "It core dumps".

Do this to your berkeley encoder.
Edit your Makefile and add -g to the CCFLAGS var.
make clean
make 2>&1 | mail -s "Matt's encoder build" dic### [at] buckosoftcom
mpeg_encode blah
[ wait for it to core ]
gdb mpeg_encode
b
mail me that stack trace. (cut and paste it, don't type it in)

dik

Post a reply to this message

in povray.general, Matt Giwer says...

> 	Not to brag but to make a point, my first Fortran program ran 33 years
> ago. Maybe I am getting senile and missing something very obvious. 

[You were just 2 years behind: my first Fortran ran in 1965...] 8-P

BTW did you use PPM RAW?

-- 
Regards,  Sander

Post a reply to this message

On Sun, 17 Sep 2000 22:50:44 +0200, Sander <san### [at] stolscom> wrote:

>in povray.general, Matt Giwer says...
>
>> 	Not to brag but to make a point, my first Fortran program ran 33 years
>> ago. Maybe I am getting senile and missing something very obvious. 
>
>[You were just 2 years behind: my first Fortran ran in 1965...] 8-P
>
>BTW did you use PPM RAW?

You beat me again Sander, my first was in 1966.
dav### [at] hamiltonitecom
http://hamiltonite.com/

Post a reply to this message

Matt Giwer wrote:
> 
>         For those like myself who can not get the Berkeley files to run (and
> for the folks who say it works but won't answer how) here is another
> source of an encoder.

	As the gods would have it, in the last few weeks MPEG-1 encoder and
MPEG-2 encode and decode compiled for linux have shown up. I have tested
both and they work fine -- and Gimp 1.x includes the code for using
them. If they are needed by anyone, let me know and I will send or make
available on the web. 

	Comment on Gimp, its utilization code is not perfect. For example it
works fine on 0-99 but from 100-199 it looks for 0100-0199. But it does
that by creating a parameter file which you can hand edit. Also the
default IBP sequence produces huge and jerky mpegs. Where it has BB I
editted to one B and reduced the GOP from 16 to 11. Much smoother and
have the size. Also, increase the buffer by times 10 seems to help. 

-- 
If gays want to get out of the gene pool there is
obviously nothing lost. 
	-- The Iron Webmaster, 229

Post a reply to this message