Yes. They render slightly different. And remember that if you are scaling down, that there is an algorithm applied which in software render is user selectable. The simple answer is, don't get overly concerned about this unless you have a need to be. Pixel peeping is fine unless it starts to get in the way of creativity. If you don't have a Red Rocket and don't plan on getting one, I would suggest focusing on what you do have and making that work for you. Make sure to fully understand the Redcine-X Pro interface and all of the options. If you are not getting expected results, make sure to go down the list of variables, starting with acquisition. Make sure there are no weak links in the chain. This includes making sure that flange distance is set correctly, lens are good samples focus is nailed 100%, etc. Then make sure everything downstream is set properly and you are treating the image's tonality correctly, etc. There are a lot of potential trap doors along the way. And if you get through all of this and it comes down to the software debayer is doing less than stellar things to your image, then consider another option. But I tend to think that's not what's holding anyone back. Unless there has been a problem that has developed in one of the RCX releases. If someone really needs to compare, I suppose you could send over a trimmed R3D and I could run it through both processes.
It's one of the variables that should be eliminated yes. Although not necessarily a direct indicator of the problem here. Before sending camera back to RED, I would suggest taking it to a rental house that has some good test lenses. They might also be able to do some comparisons with another camera as well. Might have to pay for this service depending on relationship with rental house, but would probably be worth it. If you can establish that everything checks out on the camera side, then you can start to look at the R3D processing part of it.
