Howdy -

On Mon, 5 Jan 2004, Matto Marjanovic wrote:

        Thanks for chiming in!

> I believe that for downscaling, yuvscaler defaults to its "RESAMPLE" mode,
>  which is similar to y4mscaler's "box" kernel.  The other yuvscaler mode

        Yes, that's what I thought - the usage summary from yuvscaler says
        resample is what it uses by default for downscaling.

> No (and not really)... the current default is 'linear'; a better would be
>  'cubic', and even better is something like 'sinc:8'.  I haven't had the

        I'm using sinc:8 - a carryover from the previous option of 
        'sinc8lan' (I think that's what it was called in y4mscaler 0.5.x).

> You need to add "-I norm=NTSC" *before* the "-O preset=SVCD"; options are
>  evaluated in order.  (This is not clear in the manpage.  I'll fix that.
>  In fact, it might make more sense to evaluate all -I options before any
>  -O options...)

        Ah ha - that's the part I didn't realize.  It's not clear which
        programs process their args once in order and which ones make
        multiple passes.

> The 'norm' is inferred from the frame rate (not the frame size).  mplayer

        Oh ok - I thought it might be keying on the frame size since 480, 240
        are NTSC as a rule and 576 and 288 are PAL usually.

>  has provided a lame approximation to "24000/1001", and y4mscaler is looking

        Sigh yes, they still haven't managed to get that right and I'm 
        tired of fighting 'em to fix it.   That's why I came up with 
        tools to strip off the header and allow me to recreate it correctly.

> 
> After that, y4mscaler will probably complain that it doesn't know the
>  sample aspect ratio (SAR) of the input --- and, really, it can't do much

        And don't bring that topic up on mplayer or ffmpeg mailing lists
        unless you are prepared for a flame fest (and then settling on
        72:79 instead of 10:11)

> If this is just some bootleg movie warez, it's probably in pretty sad shape
>  as it is, and no one will notice anyway; just say "-I sar=1:1".

        The source as I recall was a .AVI file so it was likely created
        as a quicktime movie, etc and thus 1:1 is the SAR and "-I sar=1:1"
        is the correct thing to use.

        Cheers,
        Steven Schultz



-------------------------------------------------------
This SF.net email is sponsored by: IBM Linux Tutorials.
Become an expert in LINUX or just sharpen your skills.  Sign up for IBM's
Free Linux Tutorials.  Learn everything from the bash shell to sys admin.
Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click
_______________________________________________
Mjpeg-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/mjpeg-users

Reply via email to