Tesla C vs. M series 2050

Hello,
does anybody know where the documentation is that describes the differences between the C and M series Tesla 2000-class devices?
I found the sales PDF for both, but it is not clear what is the meaning of C and M.

M series documents are here :http://www.nvidia.com/object/preconfigured-clusters.html
C series documents are here :http://www.nvidia.com/object/personal-supercomputing.html

The primary difference is that the M series is designed for OEMs creating a fully integrated system. If you look closely at the photos of the M2050, you’ll notice the lack of a cooling fan. The system enclosure needs to provide sufficient airflow to keep the GPU from overheating. Due to the more strict case design requirements, it doesn’t look like the M2050 is sold directly to end users. A vendor designing a compact rackmount 4 GPU server would want to use the M series.

The C2050 is packaged like a standard GPU (with active cooling), and can be installed in pretty much any standard computer that meets the space and power requirements.