Pete Burger
Forum Replies Created
-
[Firman Prasetia] “How to get rid of video-ish look?”
Don’t forget the lighting. If you light your video like film, it will look more like film…
Most of the time, a video-ish look is because people tend to treat video like… video and spend very few (if at all) time for lighting properly.Grading is only one single part of the so-called film-look. Lighting (or diffusing light, blocking light…) IMHO is a more than equal important aspect.
If you work under harsh light conditions, you’ll get very high-contrast pictures. So try to block or diffuse the light. Use bouncers, butterflies etc. to get rid of the harsh and contrasty shadows.
——————————————
“Tragedy is a close-up; comedy, a long shot.” – Buster KeatonMe on Twitter (english/german)
https://twitter.com/FastFoodVideo -
Depends a bit, with which software you work with.
I’m using Windows 7 64Bit with Premiere CS5 and After Effects CS5. Since CS5 it is possible to edit natively (without transcoding). It seems that my setup (older computer) is not perfect for bigger projects.
My workflow is as follows:
I copy the files from the card to my computer (.mov’s and .thm’s). Backup them to an external drive. Then I use the freeware MPEGStreamclip (https://www.squared5.com/) to transcode the .mov’s to DNxHD (https://avid.force.com/pkb/articles/en_US/Compatibility/en350875) using one of the 1080p presets.
From what I read everywhere on the web, the best codec for transcoding probably is “Neoscene” (https://cineform.com/), but it it a pay-codec.
If you don’t want to install a new codec, you can go with Quicktime Animation or Quicktime PNG (both set to 100%) – they are lossless as well.
You surely know, that transcoding into a 10Bit (or even 16Bit) codec will not give you better quality than using an 8Bit codec, since the H.264 MOVs are 8Bit anyway. The same goes for 4:2:2 or 4:4:4 codecs. No quality improvement since the native files are 4:2:0. But it won’t downgrade your footage either, just creates bigger files.
For transcoding you can also use 5D2RGB (https://rarevision.com/5dtorgb/) which – for some people – is the software of choice, because it is said to make better colour conversion. What I dislike about it, is that it has (or had) no built-in batch processing. You can transcode files only one by one… Maybe they have implemented that feature…
Hope this helps!
——————————————
“Tragedy is a close-up; comedy, a long shot.” – Buster KeatonMe on Twitter (english/german)
https://twitter.com/FastFoodVideo -
[Shaho Teyib] “i have DSLR Canon EOS 1D Mark IV and i work with After Effects CS6
so far Direct i copy all DSLR mov file footage to my Machine”You can do that. No problem, if your computer can handle the footage natively. If you encounter stability or performance issues, you might want to transcode your footage to ProRes (with Compressor, MPEGStreamclip, 5D2RGB, …) or any other lossless codec like Neoscene, DNxHD, …
When using in After Effects, I like to transcode the footage before working with it. With my hardware/software setup I get a smoother workflow.
Hope this helps
——————————————
“Tragedy is a close-up; comedy, a long shot.” – Buster KeatonMe on Twitter (english/german)
https://twitter.com/FastFoodVideo -
[Steve Crow] ” I don’t think there is anything you can do to correct this – the camera is capturing exactly what happened”
That’s what I thought as well… We discussed the problem before the shoot, but had no idea how to solve it. I hoped that – like in other projects – I would be able to “cut around” the flashes most of the time, but no chance here …
I thought about maybe stylizing the footage if everything else fails. Luckily this is not supposed to be a documentary but more like a “music-video-style mood-clip sort of thing”…
We had to capture the DJ-performance of an internationally very well-known musician, producer and film-composer and create a promo-clip for him.
[Steve Crow] “In theory, if you removed every other frame (the frames filled with a flash) and then replaced that frame with a copy of the one before or after it, then all the flash frames would be removed but I still think it would look weird (slow motion? and off-sync)”
That’s basically a great idea. Sync is not a problem. We won’t use the audio anyway, but get a completely different track. Strobing could be a problem, because of the missing frames… Hmmm… Maybe in After Effects: Remove the frame and interpolare the gap… Could work for single flashes, but we had a “thundestorm”…
But I think this could be a solution for a couple of shots. If I had to do all of them… Don’t even want to think about it 😉
Thanks a lot for the suggestion, Steve! Much appreciated! I’ll definitly play around with that!
Edit: Just want to add: The flashes per se don’t bother me at all. It’s the “half-frame” issue… Maybe (for selected shots): Mask the correctly-exposed part of the picture out and increase exposure for that part in post… Will try that as well
——————————————
“Tragedy is a close-up; comedy, a long shot.” – Buster KeatonMe on Twitter (english/german)
https://twitter.com/FastFoodVideo -
Your talent is standing quite close to the blue wall, which is – at least that’s how I see it – creating a bit of a blue spill, that sort of “neutralizes” the skintones (shifting everything into blue)…
If you look at the vectorscope, the skintone level is very close to the “11 o’clock skintone line” so I wouldn’t bother too much. Shifting the highlights slightly to orange and maybe increasing saturation a bit, might do the trick…Just my two cents
——————————————
“Tragedy is a close-up; comedy, a long shot.” – Buster KeatonMe on Twitter (english/german)
https://twitter.com/FastFoodVideo -
The overall lighting in both shots differs a lot.
So it’s hard to match them completely (if at all possible) without masking.One of the biggest problems I see is – like you’ve already written – in the day-shot the walls are sort of the highlights and the skintones the midtones. In the evening-shot both (the wall and the skintones) are more or less the midtones.
I’m not a colorist either, but I think I’d try to first match the contrast levels, the overall brightness and then I’d probably use a couple of masks to separate the walls from the other elements and treat them differently… Tough job…
——————————————
“Tragedy is a close-up; comedy, a long shot.” – Buster KeatonMe on Twitter (english/german)
https://twitter.com/FastFoodVideo -
I don’t quite understand, what you mean by
[Parisa Rezvani] “(…) in Avid, the general rule is 80% max before starting to lose substantial image quality” What confuses me is the “80%”… What do you mean by that?Zooming into footage always leads to a loss of quality. In my time of Avid training, I learned, that resizing to 112% to max. 115% is something you can get away with when adding a bit of sharpness afterwards and I most of the time stick to that “rule”… I don’t think there’s much difference between various NLEs.
There are some Plugins that do a much better job than a simple resize, like “Instant HD” or “Resizer” from Magic Bullet. With those you might be able to zoom in further for they use advanced algorithms.
——————————————
“Tragedy is a close-up; comedy, a long shot.” – Buster KeatonMe on Twitter (english/german)
https://twitter.com/FastFoodVideo -
Flourescents have a more or less greenish tint, so I’d maybe try to first neutralize the picture by adding a bit of magenta (or raising reds and blues / lowering greens – depending on the footage) and then use a second instance of colour-correction to get the look of the previous shots.
What software are you working with? Red Giant has (or had?) a free version of the incredible “Colorista” plugin for download. Even “Colorista Free” is a very powerful and more or less intuitive grading tool. It took a time until I “got” the mechanics, but I really like it.
“Colorista II” (the pay-version) even has a white balance feature, which is sort of a “one-click-solution”.Another great (commercial) plugin is Color Finesse. Very powerful and not easy to operate but you can get awesome results.
——————————————
“Tragedy is a close-up; comedy, a long shot.” – Buster KeatonMe on Twitter (english/german)
https://twitter.com/FastFoodVideo -
Agree with Steve. I more than once had the problem of clipped colour channels when shooting digital. A clipped colour channel is almost as worse when colour-correcting as highlight clipping.
Underexposure is a common thing when shooting digital, but don’t overdo (or underdo?) it. The shadow areas are where the noise appears. And the heavy H.264 compression will push noise and artifacts even more.
——————————————
“Tragedy is a close-up; comedy, a long shot.” – Buster KeatonMe on Twitter (english/german)
https://twitter.com/FastFoodVideo -
Found these two videos a couple of days ago.
Maybe interesting for someone who has the same question I had and stumbles upon this thread might them find useful:
Tascam DR-40 Vs Zoom H4n
https://youtu.be/QmwuS9QOlvsTascam DR-40 Feature Review (with a little bit of a contradiction to the above)
https://youtu.be/jBbj9GpiRbQ——————————————
“Tragedy is a close-up; comedy, a long shot.” – Buster KeatonMe on Twitter (english/german)
https://twitter.com/FastFoodVideoSome contents or functionalities here are not available due to your cookie preferences!This happens because the functionality/content marked as “Google Youtube” uses cookies that you choosed to keep disabled. In order to view this content or use this functionality, please enable cookies: click here to open your cookie preferences.