visceralpsyche
Well-known member
Exactly this. Well said ThomasIn theory yes you are absolutely right. Starting with any image processing with a raw signal will always be the best. In practice however that isn't always true when it comes to noise reduction. Cameras have very limited processing power and they need to de-bayer, scale and process the image 60 times per second. In the case of the GH5 that is 60 times per second at an over 4k resolution. That is a massive amount of processing power that even a top quad core computer can struggle with. The NR done in camera is nowhere near as sophisticated as that in a program like Neatvideo. Neatvideo is painfully slow for a good reason and that is how much time it can take to deal with noise reduction. NR is one of the most intensive tasks you can do in image processing. Neatvideo is also designed to compensate for some of that image compression on high noise material.
The engineers at Panasonic are some amazing people but they can only use the processing power available to them and speed and stability are an absolute must.
Just like anything else we do the answer is really it depends. It can depend on a lot of factors and having used Neatvideo a lot myself over the past few years I can tell you it does a really good job on -5 NR material. Of course if you shoot ISO12,800 and have a ton of noise use are kind of screwed either way. The in camera NR is going to create such odd blotchy artifacts I would honestly take my chances with Neatvideo making sense of the H264 macro blocking.
In the case of realistic ISO levels 200-1600 the macro blocking damage at -5NR will not be enough to harm what Neatvideo can do. If one is shooting at insane ISO levels then maybe, maybe in camera might do better but I think it is important for each person to test it out. Any NR is a massive alteration and damage to the original video. Which one is better is not the right question here and it has more to do with which one looks less weird to each person.