Jump to content

Sony RX100: Getting the best video out of it...


Bruno
 Share

Recommended Posts

Yup. That's why I always wondered why people were crazy about the HX100v. It's electronically sharp (oversharpened) and the AVCHD breaks down. Same thing with the RX100 though I haven't used it personally but every video I've seen gives that HX100v vibe.

 

AVCHD breaks down even in FS700 but the 5dmarkIII all-i never breaks down that badly. Even the old Canon 7d I still have lying around, never broke down as bad as some of the AVCHD stuff I've shot with macroblocks flying. Except for some horrible aliasing here and there.

Link to comment
Share on other sites

EOSHD Pro Color 5 for Sony cameras EOSHD Z LOG for Nikon CamerasEOSHD C-LOG and Film Profiles for All Canon DSLRs

It's odd because even with the sharpness off it looks like video, much more so than the Canon. I tried to ignore it, but it bothered me... I recently watched some videos with Iscoramas, hacked Panasonic GH no matter how high the bitrate, and they still looked like very electronic video to me. Same with most Sony camcorders. I was recently shown a film without being told what it was shot on and said "AF100 or GH2" and the fellow was surprised that I knew, he probably expected me to say "film" or something silly! But it looked really flat, and I didn't have a doubt. Poor fellow...

 

There's so much involved in CMOS, AD conversion, gain, readout methods so on and so on, then codecs too, that despite the statistics two cameras can feel completely different. Canon give you piss takingly poor stats, but as long as their final image is lovely, they can get away with it. :S

 

Yes, the codec falls apart under motion, and motion is the most useful thing due to steady shot...

 

Anyway, technology only gets better. Even though I didn't stick to the RX100 in the end, it's a good sign, and a great piece of kit for most purposes and less fussy tastes than my own.

 

I dropped my HX9v in the sea. If I hadn't I would've kept it for b-roll, but in the end I could always spot the shots too easily, and that really annoyed me after a while. I thought "just be less lazy and do it all on the Canon with a rig".

Link to comment
Share on other sites

Well I do think the RX100 looks more videoy than the GH2 - I own both.

GH2 tends to look pretty cinematic out of the BOX IMO - with the Nebula v7 444 Matrix, soft and sharp2 hack, even with the kit 14-42mm lens :)

 

I did this as a test on the RX100, and I think it looks pretty cinematic

Used portrait mode, all settings to minimum -3, in 60p mode, then graded in Vegas and conformed to 24p, some of it as 40% speed slow mo - the parts in normal speed look a bit jerky due to super fast shutter speeds and skipped frames. It was in Aperture priority mode F1.8

http://www.youtube.com/watch?v=huQx_v9NdBo

 

 

Been thinking of grabbing the PAL version.....

Link to comment
Share on other sites

Well I do thing the RX100 looks more videoy than the GH2 - I own both.

GH2 tends to look pretty cinematic out of the BOX IMO - with the Nebula v7 444 Matrix, soft and sharp2 hack, even with the kit 14-42mm lens :)

 

I did this as a test on the RX100, and I think it looks pretty cinematic

Used portrait mode, all settings to minimum -3, in 60p mode, then graded in Vegas and conformed to 24p, some it as 40% speed slow mo - the parts in normal speed look a bit jerky due to super fast shutter speeds and skipped frames. It was in Aperture priority mode F1.8

http://www.youtube.com/watch?v=huQx_v9NdBo

 

 

Been thinking of grabbing the PAL version.....

 

Try a demo of ReelSmart Motion Blur, which re-generates the missing motion blur in high-framerate footage.  Just drop the effect on the shot in a 24p timeline.  It actually works pretty well in making 60p/24p footage less jerky.

Link to comment
Share on other sites

Canon give you piss takingly poor stats, but as long as their final image is lovely, they can get away with it. :S


I'm curious if you think Canon's APS-C cameras also share that quality? I've never used any of their DSLRs that could shoot movies.
Link to comment
Share on other sites

That's why I bought the mosaic filter rather changing camera ;) I wanted the "look" but not the aliasing! And I did try out a lot of cameras. 

 

Next on the Canon SLR list will probably be an aliasing-free video focussed APS-C similar to the 5D3 in terms of quality. Probably a 7D MKii or similar. That's my guess. If so, that's be cool.

Link to comment
Share on other sites

I did a blog post about the settings I use for video as a step-by-step guide.
 
http://rungunshoot.com/how-to-set-up-your-sony-rx100-for-cinematic-video/
 
Would be interested to hear people's thoughts; I'm open to revision if there's something I'm not doing optimally.


I’m not sure of saturation at -1 for Portrait mode, which is already more saturated than standard mode.
I’ve done a few tests here and saturation at -2, even at night/high ISO looks much more natural, less blown out reds/orange etc

Try a demo of ReelSmart Motion Blur, which re-generates the missing motion blur in high-framerate footage. Just drop the effect on the shot in a 24p timeline. It actually works pretty well in making 60p/24p footage less jerky.


Thanks, but it sure is slow!
I wonder if using something like Twixtor to slow it down to 24p might be better?
Link to comment
Share on other sites

I’m not sure of saturation at -1 for Portrait mode, which is already more saturated than standard mode.
I’ve done a few tests here and saturation at -2, even at night/high ISO looks much more natural, less blown out reds/orange etc


Thanks, but it sure is slow!
I wonder if using something like Twixtor to slow it down to 24p might be better?

I'm talking about conforming 60p to 24p at 100% speed. To do slow-motion, you can just conform the footage and you should get perfect motion.  You don't need twixtor unless you're going slower than 40%.

Link to comment
Share on other sites

When dropping from 60fps to 24fps you don't need any extra frames created, you just need to discard the 36 additional frames per second.

What optical flow software like Twixtor can do is analyze the motion and recreate the 24 final frames from the original 60, creating smoother motion throughout, without any gaps, but in my experience this is not worth the trouble, as the artifacts introduced by optical flow may look quite nasty.

This obviously depends on the shots, in some situations it might work really well, but it won't work all the time, and you'll end up with inconsistent footage, and it requires loads more work and processing time.

Link to comment
Share on other sites

RX100 Questions, to get it down to a dummy level:

 

- So, 50p onto a 25p timeline is generally OK? And how comes exposure into play here? if possible never under 1/50 and over 1/120 or so? In sun light you have to go up to 1/500 sometimes, though. And what happens in low light when you have to go 1/25 (in 50p)? Footage seems just about alright, but have not dealt with placing it on various timelines etc.

 

- When I drop 50i on 25p timeline it doesn t really feel like original 25p data, FCP X seems to be doing something with it (?). Never worked with interlaced containers/wraps, looks like trouble...

 

- DRO in video creates visible noise in shadows, so why recommend it? Not so flat image profile seems preferable to ugly noise.

 

Thanks for all the tips so far.

Link to comment
Share on other sites

I use Premiere Pro.  When I import 50i clips, the properties show as 25p, so I believe they're native 25p file contained in the 50i wrapper.   The files don't need any deinterlacing like those from my 60i camcorder.

 

RX100 Questions, to get it down to a dummy level:

 

- So, 50p onto a 25p timeline is generally OK? And how comes exposure into play here? if possible never under 1/50 and over 1/120 or so? In sun light you have to go up to 1/500 sometimes, though. And what happens in low light when you have to go 1/25 (in 50p)? Footage seems just about alright, but have not dealt with placing it on various timelines etc.

 

- When I drop 50i on 25p timeline it doesn t really feel like original 25p data, FCP X seems to be doing something with it (?). Never worked with interlaced containers/wraps, looks like trouble...

 

- DRO in video creates visible noise in shadows, so why recommend it? Not so flat image profile seems preferable to ugly noise.

 

Thanks for all the tips so far.

Link to comment
Share on other sites

RX100 Questions, to get it down to a dummy level:

 

- So, 50p onto a 25p timeline is generally OK? And how comes exposure into play here? if possible never under 1/50 and over 1/120 or so? In sun light you have to go up to 1/500 sometimes, though. And what happens in low light when you have to go 1/25 (in 50p)? Footage seems just about alright, but have not dealt with placing it on various timelines etc.

 

50i is actually shot as 25p in the RX100, so expose like you would for 25fps.

The camera captures images as 25p and then encodes them as 50i, the movie file is interlaced and you need to deinterlace it in FCP if it doesn't do it automatically for you (FCP7 doesn't).

 

Going to 1/25 in low light is not very advisable if you want your motion to look "normal". Stick to 1/50 or close to that (some people defend that 1/40 looks more cinematic, maybe run some tests ans decide for yourself).

 

- When I drop 50i on 25p timeline it doesn t really feel like original 25p data, FCP X seems to be doing something with it (?). Never worked with interlaced containers/wraps, looks like trouble...

 

Deinterlacing is pretty easy, and since these images are captured progressively, there should be no drawbacks.

 

- DRO in video creates visible noise in shadows, so why recommend it? Not so flat image profile seems preferable to ugly noise.

 

I recommended using DRO at setting 2 at the most, it goes up to 5, but the images might start to suffer from it.

I'm not sure DRO is creating noise in the shadows, what is probably happening is that since it makes the shadows not as dark, the noise becomes more visible, but it was probably always there in the first place.

 

The main reason for using this is to make it easier on the encoder. If we were shooting raw we wouldn't need this flat profile, but when shooting compressed video, a flatter image will result in better and cleaner images that will give you more to work with in post.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

  • EOSHD Pro Color 5 for All Sony cameras
    EOSHD C-LOG and Film Profiles for All Canon DSLRs
    EOSHD Dynamic Range Enhancer for H.264/H.265
×
×
  • Create New...