Re: D3DPRESENTFLAG_VIDEO

From: Phil Taylor (phil_at_private-citizen.com)
Date: 02/04/05


Date: Fri, 4 Feb 2005 11:10:05 -0800

video data is temporal in nature, and depends on information from previous
frames. mpeg has pand i-frames, if I remember the terminology correctly.
basically key frames and intermediate frames. the intermediate frames depend
on previous frames up to the last iframe.

3d has nothing like that.

so perhaps, if the driver knew it was going to handle video, it might keep
more state. or it might perform some other action to make the current
environment more friendly for mixed video and 3d. its really hw dependant.

given the DDK is now under wraps, there is no way to verify what it does
under the covers in specific. but typically the driver source in the DDK is
very vanilla with all the interesting bits removed. in practice drivers
could do quite a bit of work for you.

but really, you dont know, you cant affect it, and it doesnt matter except
for the setting of this flag since it may get you better performance. again,
if you need it, set it. if you dont, dont. what else is there to worry
about?

"joreg" <no@spam.com> wrote in message
news:OCjnc8lCFHA.868@TK2MSFTNGP10.phx.gbl...
> Phil Taylor wrote:
>> driver and hw dependent.
> ok. thats what i already understood from the line in the docs saying that
> its a hint to the driver. still i don't have a glue what a driver would
> choose to do in case it gets that hint. and if i would therefore like my
> application to hint the driver this flag or not.
>
> > it doesnt give you any opportunities to do anything special in the
> > app, so why are you worrying about it?
> perdon? i am worrying because i want to understand what the code that i am
> writing does. when my code relies on someone elses code i try to
> understand that code too. in this particular case i saw no other
> opportunity to find out about the meaning of this flag than asking here.
> if you don't know an answer, maybe someone else still does.
>
> thanks anyway for your comments.
> joreg.
>
>
>>
>> all this does is allow the driver/hw to understand that the current
>> rendering may contain video. if that means the driver/hw needs to do
>> something special, it has a hint and doesnt need to guess. what it does,
>> certainly varies from each generation and likely within a generation
>> depending on level of chip ( x300,x600,x800 for ati, 6300,6600,6800 for
>> nvidia ).
>>
>> it doesnt give you any opportunities to do anything special in the app,
>> so why are you worrying about it?
>>
>> "joreg" <no@spam.com> wrote in message
>> news:unElb%23YCFHA.960@TK2MSFTNGP09.phx.gbl...
>>
>>>yeah. thanks i saw that.
>>>but do you have any idea what a driver does with that hint?
>>>
>>>Phil Taylor wrote:
>>>
>>>>in the docs, under D3DPRESENTFLAG, I see
>>>>
>>>> D3DPRESENTFLAG_VIDEO 0x00000010 This is a hint to the driver that
>>>> the back buffers will contain video data.
>>>>
>>>>
>>>>do you not see that?
>>>>
>>>>"joreg" <no@spam.com> wrote in message
>>>>news:%23ii8AHXCFHA.3936@TK2MSFTNGP09.phx.gbl...
>>>>
>>>>
>>>>>halo.
>>>>>
>>>>>in the Flag item of the D3DPRESENT_PARAMETERS i can set a flag named:
>>>>>D3DPRESENTFLAG_VIDEO
>>>>>
>>>>>but i cannot find any further documentation on this. any ideas?
>>>>>
>>>>>thnaks.
>>>>>joreg.
>>>>
>>>>
>>



Relevant Pages

  • Re: VRM9 and D3DPRESENT_INTERVAL_ONE
    ... The reason I guess that the driver buffers all Direct X drawing calls upto ... several frames in advance, ... Only some direct 3d / hardware driver guys could help. ... > video thread or from my 3D loop thread. ...
    (microsoft.public.win32.programmer.directx.video)
  • Re: [RFC 0/1] lro: Generic Large Receive Offload for TCP traffic
    ... have the eHEA driver for the SKB interface, ... set in the skb header and passed up the stack. ... either strip padding from frames or reject padded frames. ... I'd suggest to reject padded frames for aggregation as we do now ...
    (Linux-Kernel)
  • Re: mpeg 2 editing
    ... >> exact match to the first MPEG, that it will fit in perfectly with the ... >> rest of the MPEG's frames, that are simply copied to the new ... > the degraded result of retranscoding mpeg-compressed video is not only ... > cheapie capture cards and DVD and hard-disk consumer camcorders ALREADY ...
    (rec.video.desktop)
  • Re: CPU requirements for HDTV MythTV
    ... So, in other words, you have no idea if you were watching 30fps or 25fps. ... Chances are you are dropping a lot of frames and just don't notice it. ... "I got fed up with jerky video playback on my Linux HTPC and Plus UP-1100P ... 23.976Hz NTSC telecine - if displays could do it, ...
    (alt.tv.tech.hdtv)
  • RE: Excessive Frame Drops in pimg application during video capture on
    ... I suspect it's the buffering filter that's dropping rather than the vcap ... What has happened in the past is the CPU is unable to encode video fast ... logic to evenly distribute the dropped frames throughout the video file. ...
    (microsoft.public.windowsce.platbuilder)