more info below...

Stephen Berry wrote:
> Vladimir Pantelic wrote:
>   
>> Stephen Berry wrote:
>>     
>>> I'm having some trouble trying to increase the quality of the video
>>> on my dm355 design. In particular I am trying to manipulate the initQ
>>> parameter in IMP4ENC_Params structure, along with some others.
>>>       
>>> /******************************************************************************
>>>
>>>  * videoEncodeAlgCreate
>>>  
>>> ******************************************************************************/
>>>
>>> static int  videoEncodeAlgCreate(Engine_Handle hEngine,
>>>                                  VIDENC1_Handle *hEncodePtr,
>>>                                  VideoEncoder videoEncoder,
>>>                                  int width, int height, int bitrate)
>>> {
>>>     VIDENC1_DynamicParams   dynamicParams;
>>>     VIDENC1_Status          encStatus;
>>>     VIDENC1_Params          params;
>>>     IMP4VENC_Params         ext_params;
>>>     XDAS_Int32              status;
>>> //    IVIDENC1                status;
>>>     char                    *algName;
>>>     VIDENC1_Handle          hEncode;
>>>     IALG_Fxns *ividEncfxns;
>>>
>>>     algName = videoEncodeAlgNames[videoEncoder];
>>>
>>>     params.size                  = sizeof(VIDENC1_Params);
>>>     params.encodingPreset        = XDM_DEFAULT;
>>>     ...
>>>       
>> ...
>>     
>>> // Set the create time extended parameters
>>>
>>> // These parameters are for the highest quality picture...
>>>     ext_params.videncParams = params;
>>>       
>> you copy the standard params here, but you do not update the size, no?
>>
>> you need to do:
>>
>> ext_params.videncParams.size = sizeof(IMP4VENC_Params);
>>
>>
>>     
> why would I need to change the size to a different structure type?
>
> I double checked the example and this was in there, I just missed it.
> I'll give it a try and see what happens... thanks!
I made this change, and the code still SEGFAULTS in the control call. If
I return after the creation of the encoder (using the theory that it is
already setup correctly) then the code faults in the first call to process.

    Steve

_______________________________________________
Davinci-linux-open-source mailing list
[email protected]
http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source

Reply via email to