Re: ALEXA vs. RED?/ Demo, where?

Just a thought:
RED has a 4k bayer sensor, Alexa has a 3,5k bayer sensor, for a 1080P product I think the resolution difference is irrelevant. But, as it´s been said, both can give you the raw data and, in some point, both need a debayer process.
On the RED, the camera itself does a quick (low quality) debayer realtime for monitoring purposes (and I think Alexa should do the same) but records the raw image for a later and better quality debayer on a computer. We know, even if you need a 1080P output, you must make a full res debayer from your 4K image to get the best possible quality. After the full res debayer, you scale down to 1080P. Even with a big machine, a full res debayer is a very time time consuming process IF you have´nt a REDROCKET card (a very powerful and expensive hardware). So, the basic idea is: the camera records raw and leave the big job for a powerful computer.
The same logic goes for the Alexa if you'll work with the RAW data, but there´s also the prores output module, I think it's a great idea for streamline the workflow, to have a 1080P QT right from the camera. I think we all agree that using the RAW from the Alexa will give you the best quality image and more options for the grading process, that's clear, but I'm afraid the Prores module will give a very good looking image, good enough for a untrained eye (or someone who can't understand the benefits of RAW post) in a far more simple and cheap process than RAW. Ok, you can think of Prores material as proxies, but Prores 442 HQ and Prores 4444 are more likely online quality formats, so I understand the idea behind it is give you a high quality image to work from if you don't want to go for a RAW route (and by the way giving up the "Digital Cinema" power for a more video-style workflow) but if this image is enough for your needs, that's fine. I understand also that the Alexa has a very good sensor so it should easily give you a very good base image.
My concern is about the technical quality of the prores image, not for the video codec itself, but for the imaging process from the 3.5k sensor data to a 1080P size image. What kind of debayer process is used? As it's a real-time in camera process, I tend to think it's not a full res debayer, but if I'm wrong please let me know, but for me it's hard to believe there's so much processing power inside the camera and if may be, it should be a very energy consuming feature.
If the prores image is not finally as full-high quality as the RAW image, it's not a sin. The problem from my point of view (I'm in Santiago, Chile) is that NOBODY will use the RAW image, mainly because it will be more expensive to record and process. You may think that to have a very high quality camera (and very expensive compared to the RED) and NEVER use it in full quality is like have a Porsche and use it just to go to the corner store, but well, it's a Porsche (or ARRI in this case) and there's a lot of fashion around it.
I'll very glad if someone knows the inside workflow of the Alexa camera to get the 1080P image from the 3.5K sensor.

Thanks and congratulations for a great board.

Daniel Dávila

Post-Producer
Santiago-Chile
http://www.kine-imagenes.cl