Mark Sullivan
Member
I love the fact that I will be able to get raw sensor data from the RED. But let's take it a step further:
In keeping with the RED's modular, non-obsolecence approach, it should provide the ability to download a user or third-party codec. This could be in the form of a bit of position-independent code for the main CPU and a documented interface so this code knows where to find a buffered raw frame and a call-back function to deliver a chunk of compressed output. I realize the camera's codec is probably not pure software but there must be a lot of special applications one could achieve this way. (Or, you could let me download a new fuse map for the encoder FPGA!) As a sophisticated user, this would give me access to the full res raw stream without having to handle the dual-FC output in an external machine. There are probably lots of industrial and security applications for specialized in-camera processing.
In keeping with the RED's modular, non-obsolecence approach, it should provide the ability to download a user or third-party codec. This could be in the form of a bit of position-independent code for the main CPU and a documented interface so this code knows where to find a buffered raw frame and a call-back function to deliver a chunk of compressed output. I realize the camera's codec is probably not pure software but there must be a lot of special applications one could achieve this way. (Or, you could let me download a new fuse map for the encoder FPGA!) As a sophisticated user, this would give me access to the full res raw stream without having to handle the dual-FC output in an external machine. There are probably lots of industrial and security applications for specialized in-camera processing.