Licode vs. Intel Collaboration Suite for WebRTC

Intel currently provides a custom implementation of Licode called Intel Collaboration Suite for WebRTC. I’ve installed and used both, but as of right now I’m having a difficult time discerning what the actual differences are between the two. Intel doesn’t really provide any information on their framework, other than installation procedures, and API documentation. As far as I can tell, they’ve moved some things around, and changed some namespaces, but the overall interface to their implementation remains basically the same. Does anyone with experience in both know what the value-add is for using Intel’s implementation over regular Licode?

Basically what intel collaboration suite does is a transcoding of the mediastream (using their own API of the processor) and can deliver a single mosaic video of the conference. this will result in a minor bandwith consumption for the subscribers but an higher cpu utilization for your servers.
also the code is not open source I think.
Were you able to look at c++ code?

Unfortunately, I was not able to look at the code. It seems to be closed source. Due to this, I ended up having to use a combination of their documentation, and Licode source to figure out how certain things work. I wouldn’t have a problem with them being closed source, however their only support seems to consist of a low traffic forum.