Hi, all:

I know EVR can mix two streams into a single frame.
But when I create such topology and call start, I get a error code: MF_E_TOPO_UNSUPPORTED.

So I have two questions:
1) How to create such topology about mixing multi-streams into a single evr
2) Due to async programming style, it is hard to find out which step cause error.
Is there any way to know the details when I get a error code from the Invoke function

Thanks.

Although you are correct that the EVR does support multiple streams of input, the Media Foundation pipeline does not support this EVR feature in Vista. Therefore, I'd guess that your error stems from the fact that Media Foundation is trying to connect both of your inputs to the same EVR stream and is getting confused.

As an answer to your q2, if you get a topology load error (like the one you got), there will be a couple clues to help you figure out what happened. For some failures, the value associated with the MESessionTopologySet event that you get back will have a topology that was resolved as far as the MF topology loader could get before it hit trouble. If, in particular, your topology load failed due to a missing codec (which sounds like it isn't the case for you), it will have set the MF_TOPONODE_ERROR_MAJORTYPE and MF_TOPONODE_ERROR_SUBTYPE attributes on the appropriate node; this is the most common reason for topology load failures.

Becky