Nabble - xine-docs

tip on how to read dvd dir structure from hard disk

2006-12-21T12:49:00Z

Hi everyone,

xine is a fantastic piece of work, thank you so much
to all the developers who contributed to it!

I have found how to "play a DVD" from a hard disk copy
of the original disk in a way that menu navigation
works. I couldn't find a clear reference how to
achieve this in the documentation, so I thought you
might want to add it (or make it more prominent, in
case I simply missed it!):

To play a DVD from an exact (but decrypted) copy found
in dir /movie_backup/movie1 (created, eg, by vobcopy),
just start xine as

$ xine dvd://movie_backup/movie1

Unlike what happens if I simply run

$ xine /movie_backup/movie1

the dvd:/[path] command actually allows menu
navigation to be used.

This might look trivial, but the docs don't seem to
explicitly mention that you can append an absolute
path to the "dvd:/" option to get "dvd player-like"
behavior when reading from the hard-disk. Perhaps this
(or a similar) example could be added to the list of
"examples for valid MRLs (valid media resource
locator)" printed at the end of the output of "xine
--help".

All the best, season's greetings!,

Juergen

___________________________________________________________
Der frühe Vogel fängt den Wurm. Hier gelangen Sie zum neuen Yahoo! Mail: http://mail.yahoo.de

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: website

2005-10-13T08:08:38Z

Hi,

>> I was thinking if it would be a good idea to cut down on the
>> administration section of the website and manage the news in a
>> similar way new package releases are currently managed: A file in
>> CVS (probably XML) to which one can add the news. Benefit: News go
>> through the testing website and can be checked for correctness
>> before being synced to the real website.
>
> With this, there's a potential update race if somebody else is
> modifying other parts of the web site. As things stand, new news
> items are automatically checked for well-formedness (if not
> validity); a preview page shouldn't be too hard to add.

When I did the last release, I thought this could actually be more
convenient. One already has to go through the xine_www CVS when doing
a release, so why not handling news in the same way.

Michael

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: website

2005-10-12T18:43:12Z

I demand that Michael Roitzsch may or may not have written...

>>> Apart from voting and administration, is the login used for anything
>>> else?

>> Currently: no.
>> Well, there were some provisions to let users select a website theme,
>> which gets used as soon as they log in, and there were plans to use the
>> login for our own bug tracker, which never was deployed. But these were
>> only plans...

> I was thinking if it would be a good idea to cut down on the administration
> section of the website and manage the news in a similar way new package
> releases are currently managed: A file in CVS (probably XML) to which one
> can add the news. Benefit: News go through the testing website and can be
> checked for correctness before being synced to the real website.

With this, there's a potential update race if somebody else is modifying
other parts of the web site. As things stand, new news items are
automatically checked for well-formedness (if not validity); a preview page
shouldn't be too hard to add.

Also, the site update code needs to be altered to do partial updates - this
is needed anyway for the files in data/xml (where the above race is already
possible). Now, where did I put that "to do" list...

[snip]
--
| Darren Salt | d youmustbejoking,demon,co,uk | nr. Ashington,
| Debian, | s zap,tartarus,org | Northumberland
| RISC OS | @ | Toon Army
| Let's keep the pound sterling

You will be imprisoned for contributing time and skill to a bank robbery.

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: website

2005-10-11T02:52:28Z

On Mon, Oct 03, 2005 at 07:48:11PM +0100, Darren Salt wrote:

> It may be that the current site can handle the bandwidth
> requirements (more information needed). James - if you're happy with
> this, creation of xine.tartarus.org will help, and newsite.xinehq.de
> can then be pointed at it.

Happy in principle, but I'll need some kind of figures to be
certain. It's more to do with whether the box can take it than whether
the bandwidth is available.

Surely newsite.xinehq.de could be IN A straight to ixion's ip address,
and I can VirtualHost it in from there? The alternative is a CNAME,
which is messy.

Cheers,
James

--
/--------------------------------------------------------------------------\
James Aylett xapian.org
james@... uncertaintydivision.org

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: website

2005-10-06T16:17:56Z

I demand that Siggi Langauf may or may not have written...

> On Wed, 5 Oct 2005, Darren Salt wrote:
>>>> before import into the new one, and a dump of the current db would help
>>>> :-)
>>> Who can provide that?
>> Siggi... Cc'ing (again) in case of lack of subscription to this list.

> Actually, I am subscribed to -docs,

This is good...

> and I'll finally have some time for the web site tomorrow...

Right. A database dump *would* be handy - I could populate the news section,
for a start... :-)

[snip]
>> And we need some testing of the login stuff. BTW, should we keep the old
>> voting requirement (login required) or do something else? If something
>> else, then I'll hide the login link.

> Definitely keep the login requirement. There's hardly another way to keep
> people from voting several times.

Will do.

--
| Darren Salt | d youmustbejoking,demon,co,uk | nr. Ashington,
| Debian, | s zap,tartarus,org | Northumberland
| RISC OS | @ | Toon Army
| <URL:http://www.youmustbejoking.demon.co.uk/progs.packages.html>

The days of the digital watch are numbered.

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: website

2005-10-06T04:53:41Z

Hi,

>> Apart from voting and administration, is the login used for
>> anything else?
>
> Currently: no.
> Well, there were some provisions to let users select a website
> theme, which gets used as soon as they log in, and there were plans
> to use the login for our own bug tracker, which never was deployed.
> But these were only plans...

I was thinking if it would be a good idea to cut down on the
administration section of the website and manage the news in a
similar way new package releases are currently managed: A file in CVS
(probably XML) to which one can add the news. Benefit: News go
through the testing website and can be checked for correctness before
being synced to the real website.

Of course we still need the login to trigger the syncing.

Michael

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: website

2005-10-06T04:38:23Z

On Thu, 6 Oct 2005, Michael Roitzsch wrote:

>> Surely you noticed at least /some/ of the CVS commits ;-)
>
> Now that you mention it, yes I did.
>
>> And we need some testing of the login stuff. BTW, should we keep the old
>> voting requirement (login required) or do something else? If something
>> else, then I'll hide the login link.
>
> Apart from voting and administration, is the login used for anything else?

Currently: no.
Well, there were some provisions to let users select a website theme,
which gets used as soon as they log in, and there were plans to use the
login for our own bug tracker, which never was deployed. But these were
only plans...

Cheers,
Siggi

--
A: Because it breaks the logical sequence of discussion.
Q: Why is top posting bad?

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: pts and vpts

2005-10-06T04:38:05Z

Hi,

> is it that metronom finds vpts for every frame including one that
> has pts ...

Metronom calculated a VPTS for every video frame, yes.

> if so, which value is used by vo loop for frames having both
> orginal pts and vpts ...

Output loops only look at the VPTS values.

Michael

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: Re: [xine-user] which is master clock ??

2005-10-06T04:36:22Z

Hi,

> according to my understanding , xine_stream_t represents a program
> stream containing audio and/or video ..

No, xine_stream_t is just an API. It could play anthing from a simple
MP3 file, streaming media, AVIs, ASF up to a complete DVD with menus.

> if so, seems that the xine application creates a stream using
> xine_stream_new() , what if its a transport stream having multiple
> program streams ...

Streams are limited to one video channel, so filtering is required by
either the input plugin or the demuxer.

Michael

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

[xine-devel] Re: [xine-user] which is master clock ??

2005-10-06T04:31:26Z

Ninad P wrote:

>hi
>
>
>On 10/6/05, James Courtier-Dutton <James@...> wrote:
>
>
>>Ninad P wrote:
>>
>>
>>>hi
>>>
>>>is it that xine uses audio clock to be master at all times ??
>>>
>>>thanks
>>>
>>>Ninad
>>>
>>>
>>>
>>No.
>>
>>
>
>
>thanks
>
>according to my understanding , xine_stream_t represents a program stream
>containing audio and/or video ..
>
>if so,
>seems that the xine application creates a stream using xine_stream_new() ,
>what if its a transport stream
>having multiple program streams ...
>
>i think the demux should ve authority to create stream...
>
>whats ur opinion
>
>regards
>
>Ni
>
>
>

Have you read the hackers guide in the xine-lib doc directory?

xine has a system clock source that free runs and should be
monotonic.Normally this is from a get_time_of_day() function, but any
other reference clock source can be plugged in depending on the
application. The video stream then has PTS values. The metronom uses
interpolation and offset values to convert PTS values in the stream to
VPTS values. The VPTS value is then directly comparable to the system
clock. So, once a video frame VPTS = system clock, it is displayed. The
same with Audio and Subtitles etc. So, generally all the different
elementary streams sync to the system clock. I by product of this is
that then all the audio,video and subtitles are then in sync. For video,
one just displays each frame at the correct time. Audio is more
difficult, as audio samples will clock out at whatever speed the audio
card's own clock is running. To bypass this problem, the xine-engine
resamples the audio, so in effect, bringing the audio card's clock into
sync with the system clock. I.e. If the audio card's clock is slightly
faster than the system clock, we use interpolation/resampling to insert
extra values in the audio stream, thus keeping them in sync.

James

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-devel mailing list
xine-devel@...
https://lists.sourceforge.net/lists/listinfo/xine-devel

Re: website

2005-10-06T04:30:53Z

Hi Darren,

> Surely you noticed at least /some/ of the CVS commits ;-)

Now that you mention it, yes I did.

> And we need some testing of the login stuff. BTW, should we keep
> the old voting requirement (login required) or do something else?
> If something else, then I'll hide the login link.

Apart from voting and administration, is the login used for anything
else?

Michael

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

[xine-devel] pts and vpts

2005-10-06T00:42:06Z

hi

is it that metronom finds vpts for every frame including one that has pts ...
if so,
which value is used by vo loop for frames having both orginal pts and vpts ...

thanks

Ni

Re: [xine-user] which is master clock ??

2005-10-05T22:17:04Z

On 10/6/05, James Courtier-Dutton <James@...> wrote:

Ninad P wrote:
> hi
>
> is it that xine uses audio clock to be master at all times ??
>
> thanks
>
> Ninad
>
No.

thanks

according to my understanding , xine_stream_t represents a program stream containing audio and/or video ..

if so,
seems that the xine application creates a stream using xine_stream_new() , what if its a transport stream
having multiple program streams ...

i think the demux should ve authority to create stream...

whats ur opinion

regards

Ni

Re: website

2005-10-05T17:16:22Z

On Wed, 5 Oct 2005, Darren Salt wrote:

>>> before import into the new one, and a dump of the current db would help
>>> :-)
>
>> Who can provide that?
>
> Siggi... Cc'ing (again) in case of lack of subscription to this list.

Actually, I am subscribed to -docs, and I'll finally have some time for
the web site tomorrow...

> (Siggi:
>>>> Auto-syncing this installation from CVS would be great.
>>> If it stays on tartarus.org, we (me, James) should be able to handle that.
>>>> Maybe some users would also like to contribute graphics or design
>>>> improvements.
>>> That'd be good. (Maybe I should pull in the "vote" area of the current
>>> site...?)
>
>> I would also like to try myself on some design, if I find the time. A vote
>> would indeed be cool, if we have enough contributions.
>
> Right. I'll get that added.
>
> And we need some testing of the login stuff. BTW, should we keep the old
> voting requirement (login required) or do something else? If something else,
> then I'll hide the login link.

Definitely keep the login requirement. There's hardly another way to keep
people from voting several times.

More on this tomorrow...

-siggi

--
A: Because it breaks the logical sequence of discussion.
Q: Why is top posting bad?

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: website

2005-10-05T12:14:18Z

I demand that Michael Roitzsch may or may not have written...

>>> My suggestion for the first step would be to get it into our CVS
>> $ cvs -d... co -t experimental2 xine_www

> Oh! That's good.

Surely you noticed at least /some/ of the CVS commits ;-)

>> The content of the current database will need to be adjusted somewhat
>> before import into the new one, and a dump of the current db would help
>> :-)

> Who can provide that?

Siggi... Cc'ing (again) in case of lack of subscription to this list.

(Siggi:

>>> Auto-syncing this installation from CVS would be great.
>> If it stays on tartarus.org, we (me, James) should be able to handle that.
>>> Maybe some users would also like to contribute graphics or design
>>> improvements.
>> That'd be good. (Maybe I should pull in the "vote" area of the current
>> site...?)

> I would also like to try myself on some design, if I find the time. A vote
> would indeed be cool, if we have enough contributions.

Right. I'll get that added.

And we need some testing of the login stuff. BTW, should we keep the old
voting requirement (login required) or do something else? If something else,
then I'll hide the login link.

>>> I have recently started to learn PHP myself, so I can probably lend a
>>> hand, too.
>> The test site is currently using PHP 4.1.2.

> I learned PHP5, but PHP 4.x should be close to a subset of this, right?

I think so. The documentation (as found in the Debian phpdoc package) covers
PHP 4 and 5 and generally does mention when a function was added or modified,
so there should be no problem there.

--
| Darren Salt | d youmustbejoking,demon,co,uk | nr. Ashington,
| Debian, | s zap,tartarus,org | Northumberland
| RISC OS | @ | Toon Army
| <URL:http://www.youmustbejoking.demon.co.uk/progs.packages.html>

"Bother", said Pooh, as he scrambled his partition table.

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: [xine-user] which is master clock ??

2005-10-05T11:47:01Z

Ninad P wrote:
> hi
>
> is it that xine uses audio clock to be master at all times ??
>
> thanks
>
> Ninad
>
No.

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-user mailing list
xine-user@...
https://lists.sourceforge.net/lists/listinfo/xine-user

Re: website

2005-10-05T09:22:02Z

Hi Darren,

>> My suggestion for the first step would be to get it into our CVS
>
> $ cvs -d... co -t experimental2 xine_www

Oh! That's good.

> greed. The content of the current database will need to be adjusted
> somewhat before import into the new one, and a dump of the current
> db would help :-)

Who can provide that?

>> Auto-syncing this installation from CVS would be great.
>
> If it stays on tartarus.org, we (me, James) should be able to
> handle that.
>
>> Maybe some users would also like to contribute graphics or design
>> improvements.
>
> That'd be good. (Maybe I should pull in the "vote" area of the current
> site...?)

I would also like to try myself on some design, if I find the time. A
vote would indeed be cool, if we have enough contributions.

>> I have recently started to learn PHP myself, so I can probably
>> lend a hand, too.
>
> The test site is currently using PHP 4.1.2.

I learned PHP5, but PHP 4.x should be close to a subset of this, right?

Michael

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: which is master clock ??

2005-10-05T05:37:58Z

Hi,

> is it that xine uses audio clock to be master at all times ??

No, the system clock is the master clock. But additional clocks can
be registered, if you use special hardware like an MPEG decoder card
with its own clock.

Michael

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: Re: [xine-user] xine-engine

2005-10-05T05:36:52Z

Hi,

> metronom: general pts => virtual calculation/assoc
>
> virtual pts: unit 1/90000 sec, always increasing
> can be used for synchronization
> video/audio frame with same pts also have same vpts
> but pts is likely to differ from vpts
>
> the basic idea is:
> video_pts + video_wrap_offset = video_vpts
> audio_pts + audio_wrap_offset = audio_vpts
>
> // Q -> what are these, video_wrap_offset and audio_wrap_offset
>
> - video_wrap_offset should be equal to audio_wrap_offset as to have
> perfect audio and video sync. They will differ on brief periods due
> discontinuity correction.

The comment just gives the general idea. If you look at the code, you
will find that there is indeed just one offset value called vpts_offset.

> - metronom should also interpolate vpts values most of the time as
> video_pts and audio_vpts are not given for every frame.

It does by using reported durations. See previous mail.

> - corrections to the frame rate may be needed to cope with bad
> encoded streams.

There is a smoothing algorithm which does that.

> // Q-> is this when pts or dts values or other timing info is
> currupted ??

We do not evaluate DTS values, but yes, there are quite a number of
streams with broken PTS.

> in this explaination:
>
> /* metronom prebuffer can be adjusted with
> XINE_PARAM_METRONOM_PREBUFFER.
> * it sets how much the first video/audio frame should be delayed to
> * have some prebuffering at the output layers. reducing this value
> (about
> * 1/8 sec) may result in faster seeking (good to simulate play
> backwards,
> * for example).
> */
>
> // what is metronom prebuffer??

The first frame arriving after playback begins is not tagged with the
VPTS of the current time, but a bit into the future. This allows the
output fifos some time to fill up.

> // and why it is needed ??

If the output fifos are filled, we can tolerate decreasing decoder
performance (frames taking longer to decode) to some degree.

Michael

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: xine-engine

2005-10-05T05:28:29Z

Hi,

> can anyone explain,
> VIDEO_PREDICTION_MODE and VIDEO_PTS_MODE in a function
> # static void metronom_got_video_frame (metronom_t *this,
> vo_frame_t *img)
> function in metronom.c

Metronom's primariy function is to create VPTS (virtual presentation
timestamps, those are used to schedule the actual display, so the
term is misleading) from the PTS (presentation timestamps) the stream
reports.

In VIDEO_PTS_MODE, it uses only the PTS value, in
VIDEO_PREDICTION_MODE, it uses the PTS value and the reported frame
duration.

> what "img->duration" specifies??

The decoder can attach a display duration to any frame.

Michael

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: xine docs

2005-10-05T05:19:14Z

Hi,

> does anyone have documentation on xine engine other that xine
> hackers guide..
> i am particularly interested in xine-engine and av sync part ...
> metronom !!

Sorry, the only docs beyond the hacker's guide is the actual code and
the comments in it.

Michael

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

[xine-user] which is master clock ??

2005-10-05T04:52:40Z

hi

is it that xine uses audio clock to be master at all times ??

thanks

Ninad

Re: [xine-user] xine-engine

2005-10-05T03:20:28Z

hi

this is explaination frm metronom,h.... ve few doubts
see lines starting with //

-------------------------------------------------------------------------------------------
metronom: general pts => virtual calculation/assoc

virtual pts: unit 1/90000 sec, always increasing
             can be used for synchronization
             video/audio frame with same pts also have same vpts
             but pts is likely to differ from vpts

the basic idea is:
   video_pts + video_wrap_offset = video_vpts
   audio_pts + audio_wrap_offset = audio_vpts

// Q -> what are these, video_wrap_offset and audio_wrap_offset

- video_wrap_offset should be equal to audio_wrap_offset as to have
   perfect audio and video sync. They will differ on brief periods due
   discontinuity correction.
- metronom should also interpolate vpts values most of the time as
   video_pts and audio_vpts are not given for every frame.
- corrections to the frame rate may be needed to cope with bad
   encoded streams.
// Q-> is this when pts or dts values or other timing info is currupted ??

-------------------------------------------------------------------------------------------------------
in this explaination:

/* metronom prebuffer can be adjusted with XINE_PARAM_METRONOM_PREBUFFER.
* it sets how much the first video/audio frame should be delayed to
* have some prebuffering at the output layers. reducing this value (about
* 1/8 sec) may result in faster seeking (good to simulate play backwards,
* for example).
*/

// what is metronom prebuffer??
// and why it is needed ??

thanks

Ninad

Re: [xine-user] xine-engine

2005-10-05T02:54:14Z

On 10/5/05, Ninad P <xine.in@...> wrote:
> HI
>
> can anyone explain,
> VIDEO_PREDICTION_MODE and VIDEO_PTS_MODE in a function

/* metronom video modes */
#define VIDEO_PREDICTION_MODE 0 /* use pts + frame duration */
#define VIDEO_PTS_MODE 1 /* use only pts */

Some demuxers cannot provide the duration of each frame like the ASF
demuxer, in that case the duration is 0 and pts are trusted.
In prediction mode, the metronom tries to guess the vpts of the next
frame using the duration of the current frame, but unfortunately that
is never correct and that's why the drift stuff is here to compensate
pts errors.

> # static void metronom_got_video_frame (metronom_t *this, vo_frame_t *img)
> function in metronom.c
>
> what "img->duration" specifies??

the duration of the frame in 1/90000 unit

> Thanks
>
> Ninad

Thibaut

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-user mailing list
xine-user@...
https://lists.sourceforge.net/lists/listinfo/xine-user

[xine-user] xine-engine

2005-10-05T00:34:47Z

HI

can anyone explain,
VIDEO_PREDICTION_MODE and VIDEO_PTS_MODE in a function
# static void metronom_got_video_frame (metronom_t *this, vo_frame_t *img)
function in metronom.c

what "img->duration" specifies??

Thanks

Ninad

xine docs

2005-10-05T00:27:18Z

HI

does anyone have documentation on xine engine other that xine hackers guide..
i am particularly interested in xine-engine and av sync part ... metronom !!

Regards
Ninad

Re: website

2005-10-03T12:48:11Z

I demand that Michael Roitzsch may or may not have written...

> I don't want Darren's website-overhaul to die, so let's make some actual
> plans on how to deploy this and how to make the migration as smooth as
> possible.

> My suggestion for the first step would be to get it into our CVS

$ cvs -d... co -t experimental2 xine_www

> and have a public installation somewhere (preferably on xinehq.de, maybe
> newsite.xinehq.de) which we can point our users to for testing.

Agreed. The content of the current database will need to be adjusted somewhat
before import into the new one, and a dump of the current db would help :-)

It may be that the current site can handle the bandwidth requirements (more
information needed). James - if you're happy with this, creation of
xine.tartarus.org will help, and newsite.xinehq.de can then be pointed at it.

> Auto-syncing this installation from CVS would be great.

If it stays on tartarus.org, we (me, James) should be able to handle that.

> We could then collect user opinions on this list and start hacking on it
> collaboratively in CVS until we think it is ready to make the final switch.
> (Hopefully this won't be too far away, with all the hard work Darren has
> already put into it.) Maybe some users would also like to contribute
> graphics or design improvements.

That'd be good. (Maybe I should pull in the "vote" area of the current
site...?)

> I have recently started to learn PHP myself, so I can probably lend a hand,
> too.

The test site is currently using PHP 4.1.2.

--
| Darren Salt | d youmustbejoking,demon,co,uk | nr. Ashington,
| Debian, | s zap,tartarus,org | Northumberland
| RISC OS | @ | Toon Army
| <URL:http://www.youmustbejoking.demon.co.uk/> (PGP 2.6, GPG keys)

"... using his strength. And that is his strength, his strength."

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: website

2005-10-02T09:27:36Z

Hi,

I don't want Darren's website-overhaul to die, so let's make some
actual plans on how to deploy this and how to make the migration as
smooth as possible.

My suggestion for the first step would be to get it into our CVS and
have a public installation somewhere (preferably on xinehq.de, maybe
newsite.xinehq.de) which we can point our users to for testing. Auto-
syncing this installation from CVS would be great. We could then
collect user opinions on this list and start hacking on it
collaboratively in CVS until we think it is ready to make the final
switch. (Hopefully this won't be too far away, with all the hard work
Darren has already put into it.) Maybe some users would also like to
contribute graphics or design improvements.

I have recently started to learn PHP myself, so I can probably lend a
hand, too.

Michael

-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: website

2005-09-13T10:53:38Z

I demand that Miguel Freitas may or may not have written...

> On 8/1/05,

What? *That* long ago? I don't think so...

> Darren Salt <linux@...> wrote:
>> [xine_www, tag "experimental2"]
>>>>> Is there an installation of all the hot stuff I see going into CVS?
>>>> Yes, but it's not publicly visible (and it's impractical to make it so).
>>>> Perhaps it (and a new database for it) should be installed on
>>>> experimental.xinehq.de?
>>> I am all for it. Who has the administrative powers to do so?
>> Er, Siggi, I think... CC'ing.

> i think we should make a plan to upgrade our main site to one of the
> so-called experimental branches as soon as possible.

Probably, yes...

> i believe the most important thing is not getting everything right at the
> first try (of course most of the new site must be functional so users may
> still download, read faq,

That's possible ATM.

> post bugs...),

That's Sourceforge's problem ;-)

> but rather having somebody with willingness to maintain and fix issues.
> imho Darren, being an active and competent developer would be a great
> choice.

That I can certainly do. (The database will need to be recreated; who does
that doesn't really matter too much.)

> with this lack of feedback he might soon lose his interest in working at
> the web site, just like Stephen's efforts...

I did let it slide a bit, yes ;-)

> Darren, your test site works fine here (so does experimental.xinehq.de).
> but i prefer that old xinehq.de top banner than the new graphics in your
> navigation bar.

You'll still get the old bar if you're using a browser without CSS support.
The reason why I'm not using it (with CSS) is because it just doesn't fit
into my current design. (Neither does IE, but it seems that a few fairly
simple changes should mostly cure that.)

If anybody wants to design something to replace the fairly quick-and-simple
menu images - ideally some which don't require any significant stylesheet
changes! - that'd be useful. (The image's file size will be taken into
account.)

And don't forget the source files for the images! :-)

> as a programmer, i think we better leave the graphics part to the graphics
> designer ;-)

Probably ;-)

--
| Darren Salt | d youmustbejoking,demon,co,uk | nr. Ashington,
| Debian, | s zap,tartarus,org | Northumberland
| RISC OS | @ | Toon Army
| <URL:http://www.youmustbejoking.demon.co.uk/progs.packages.html>

You don't know what you want, and are willing to go through hell to get it.

-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs

Re: website

2005-09-10T19:45:35Z

On 8/1/05, Darren Salt <linux@...> wrote:
> [xine_www, tag "experimental2"]
> >>> Is there an installation of all the hot stuff I see going into CVS?
> >> Yes, but it's not publicly visible (and it's impractical to make it so).
> >> Perhaps it (and a new database for it) should be installed on
> >> experimental.xinehq.de?
>
> > I am all for it. Who has the administrative powers to do so?
>
> Er, Siggi, I think... CC'ing.

i think we should make a plan to upgrade our main site to one of the
so called experimental branches as soon as possible. i believe the
most important thing is not getting everything right at the first try
(of course most of the new site must be functional so users may still
download, read faq, post bugs...), but rather having somebody with
willingness to maintain and fix issues. imho Darren, being an active
and competent developer would be a great choice.

with this lack of feedback he might soon lose his interest in working
at the web site, just like Stephen's efforts...

Darren, your test site works fine here (so does
experimental.xinehq.de). but i prefer that old xinehq.de top banner
than the new graphics in your navigation bar. as a programmer, i think
we better leave the graphics part to the graphics designer ;-)

Miguel

-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
xine-docs mailing list
xine-docs@...
https://lists.sourceforge.net/lists/listinfo/xine-docs