[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

OpenDivX digest, Vol 1 #76 - 4 msgs



Send OpenDivX mailing list submissions to
	[email protected]

To subscribe or unsubscribe via the World Wide Web, visit
	http://lists.projectmayo.com/mailman/listinfo/opendivx
or, via email, send a message with subject or body 'help' to
	[email protected]

You can reach the person managing the list at
	[email protected]

When replying, please edit your Subject line so it is more specific
than "Re: Contents of OpenDivX digest..."


Today's Topics:

   1. RE: subscribe me from the mailinglist Please ([email protected]) (c0redumb)
   2. RE: dct (c0redumb)
   3. RE: dct (prakash)
   4. Re: Strange problems on the ARM platform (Lionel Ulmer)

--__--__--

Message: 1
From: "c0redumb" <[email protected]>
To: "Kim Orheden" <[email protected]>,
	<[email protected]>
Subject: RE: [OpenDivX] subscribe me from the mailinglist Please ([email protected])
Date: Tue, 22 May 2001 03:07:21 -0700

Sorry, but to subscribe to the mailing list, please point your browser to
http://lists.projectmayo.com/mailman/listinfo/opendivx, and put in your
email address there. Hope it helps.

c0redumb

> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of Kim Orheden
> Sent: Tuesday, May 22, 2001 2:24 AM
> To: [email protected]
> Subject: [OpenDivX] subscribe me from the mailinglist Please
> ([email protected])
>
>
> This is the 5  time i try to let you knew
> about mu subscrition.
> Please fix it know!
>
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of Arpi
> Sent: Tuesday, May 22, 2001 11:06 AM
> To: [email protected]
> Subject: Re: Re: [OpenDivX] input and output format
>
>
> Hi,
>
> > > Yes, the encore takes standard I420 (aka IYUV). The format is as you
> > > described x*y byte of Y, x/2*y/2 bytes of U and x/2*y/2 bytes of V.
> > >
> > > There is nothing wrong if you want to have different format before
> encore,
> > > and after decore. All that matters is you match the output of one step
> to
> > > the input of the next step. For example, if you display only
> takes RGB,
> > > then you will need to convert you output to RGB.
> >
> > Of course I know how to convert these formats from and to RGB. It wasn't
> > meant as a _bug report_.
> > I just meant that it's strange for a codecs produces different
> output than
> > input. For example you have to convert before you can compare how good
> > original was restored.
>
> What about DEC_USER? It produces U, Y, V planes, it can be given to the
> encoder without any modification and conversion.
>
>
> A'rpi / Astral & ESP-team
>
> --
> mailto:[email protected]
> http://esp-team.scene.hu
>
> _______________________________________________
> OpenDivX mailing list
> [email protected]
> http://lists.projectmayo.com/mailman/listinfo/opendivx
>
>
> _______________________________________________
> OpenDivX mailing list
> [email protected]
> http://lists.projectmayo.com/mailman/listinfo/opendivx


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com



--__--__--

Message: 2
From: "c0redumb" <[email protected]>
To: "prakash" <[email protected]>, <[email protected]>
Subject: RE: [OpenDivX] dct
Date: Tue, 22 May 2001 03:11:57 -0700

Thanks. It sounds strange, since the DCT is thoroughtly tested, and cross
tested with standard MPEG-4 bitstreams. I will double check later. Are you
sure that those are the coefficients for the same Macroblock?

c0redumb

> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of prakash
> Sent: Monday, May 21, 2001 6:43 PM
> To: [email protected]
> Subject: [OpenDivX] dct
>
>
> Hello all,
> I have tried to compare the fdct values which come  out of the dct routine
> under linux for mpeg2(TM-5) encoder and the ones from encore. The
> difference between the DC values is large for the same YUV file used.
>
> Eg:
> 	 200  -40  -23  -13  -16  0 (are the first few values obtained
> 					from TM-5 after dct)
> 	1227  -60    0    1    0  0 (are the values from encore)
>
> Since both the fdct implementation are the same, I am bit lost at this
> huge variation in the dc values?
> The the YUV files were generated from flower garden video sequence.
>
> with regards
> Prakash
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> _______________________________________________
> OpenDivX mailing list
> [email protected]
> http://lists.projectmayo.com/mailman/listinfo/opendivx


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com



--__--__--

Message: 3
Date: Tue, 22 May 2001 13:22:11 +0300 (EET DST)
From: prakash <[email protected]>
To: c0redumb <[email protected]>
Cc: <[email protected]>
Subject: RE: [OpenDivX] dct

Hello All,
They should of the same macro block. The co-efficients are from the first
block of a macro block in the first Y frame.


with regards
Prakash













On Tue, 22 May 2001, c0redumb wrote:

> Thanks. It sounds strange, since the DCT is thoroughtly tested, and cross
> tested with standard MPEG-4 bitstreams. I will double check later. Are you
> sure that those are the coefficients for the same Macroblock?
>
> c0redumb
>
> > -----Original Message-----
> > From: [email protected]
> > [mailto:[email protected]]On Behalf Of prakash
> > Sent: Monday, May 21, 2001 6:43 PM
> > To: [email protected]
> > Subject: [OpenDivX] dct
> >
> >
> > Hello all,
> > I have tried to compare the fdct values which come  out of the dct routine
> > under linux for mpeg2(TM-5) encoder and the ones from encore. The
> > difference between the DC values is large for the same YUV file used.
> >
> > Eg:
> > 	 200  -40  -23  -13  -16  0 (are the first few values obtained
> > 					from TM-5 after dct)
> > 	1227  -60    0    1    0  0 (are the values from encore)
> >
> > Since both the fdct implementation are the same, I am bit lost at this
> > huge variation in the dc values?
> > The the YUV files were generated from flower garden video sequence.
> >
> > with regards
> > Prakash
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > _______________________________________________
> > OpenDivX mailing list
> > [email protected]
> > http://lists.projectmayo.com/mailman/listinfo/opendivx
>
>
> _________________________________________________________
> Do You Yahoo!?
> Get your free @yahoo.com address at http://mail.yahoo.com
>
>
> _______________________________________________
> OpenDivX mailing list
> [email protected]
> http://lists.projectmayo.com/mailman/listinfo/opendivx
>



--__--__--

Message: 4
Date: Tue, 22 May 2001 15:00:52 +0200
From: Lionel Ulmer <[email protected]>
To: [email protected]
Subject: Re: [OpenDivX] Strange problems on the ARM platform

> This means to whomever has write priviledges to the CVS tree need to add a
> new define to the Linux Makefile (like NO_UNALIGNED_ACCESSES) and change
> in basic_predictions.c the '#ifdef BIG_ENDIAN' by 
> '#if defined(BIG_ENDIAN) or defined(NO_UNALIGNED_ACCESSES)' (or the correct
> preprocessor command, I am a bit rusty on this point :-) ).

By the way, I just tested on a big endian platform (Sun UltraSPARC), and
copying by doing a cast to 'long *' works also very well...

So the #ifdef need to change from BIG_ENDIAN to NO_UNALIGNED_ACCESSES. No
need to add the 'or' as told in my previous mail.

    Lionel

-- 
		 Lionel Ulmer - http://www.bbrox.org/



--__--__--

_______________________________________________
OpenDivX mailing list
[email protected]
http://lists.projectmayo.com/mailman/listinfo/opendivx


End of OpenDivX Digest


Reply To Poster

Local References / HOW-TO / FAQs