Just a general comment to anyone reading this. Before you ask questions please credit the developers of ML with some intelligence. You may have a bright idea but the chances are they have already been there, done that and bought the t-shirt. They are already pushing the cameras way beyond the limits of what Canon themselves envisaged.
I don't know if that was indirectly addressed to me because of my question, but my doubt was to anyone that can answer me.
I don't have anything to say to the devs, but "thank you". Man, yesterday I made a video with my 50D with an old CF card (get to 30mb/s write speed) and I managed to achieve 1152 x 540 some times. Blowed my mind, I can't wait my 64gb 1000x Transcend to arrive.
But I made a question about the 2000px of width because:
1-
H.264 is line skipping just the same as non-crop mode is with raw video then upscaled to 1080p. It's not resolving full 1080p. Crop mode records 1:1 pixels from a portion of the sensor. There is no inbetween capability. It's not adjustable.
And the sensor in 50D have 4752 pixels of width.
2-
There is a comment in the code that says:
resolution should be multiple of 64x32 or 128x16
this way, we get frame size multiple of 512, so there's no write speed penalty
So, I want to understand the 2000px width, not questioning the reason. Just seams that 2048px would make more sense and I'm not finding an answer in this 100+ pages topic.
edit: I just tried the crop mode. Two seconds on some great resolutions... and there is 2048px there. My bad. So I guess there is no need for answers, although I don't understand everybody writes "2000" instead of "2048".