Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Help, I need to make a tough decision: 1080p finish or 4K finish?

Anson Fogel said:
On a related note - I hear you are shooting on some rivers - we took RED down the Grand Canyon last year on river, and shoot a LOT of water and rivers, happy to relate some hard learned lessons.
Click to expand...

Would you be constructively willing to share that with all of us? I am very curious about this statement.

David
 
Anson's description of Twixtor's limitations kind of scares me. David's statement that 3K mixes okay with 4K is nice to hear. One problem, though, might be that the 3K Red footage might also have to cut up against 5.6k RAW stuff from the Canon cameras, which is breathtakingly sharp and detailed. I'm not as worried about cutting Red 3K against Red 4K as I am about cutting Red 3K against Canon 5.6K RAW in a 4K RGB finish master..

Some thoughts:

1) If I have to speed up the shutter at 4K to make it easier to slow down in post, I don't actually gain a stop of light, like I might if I shot 24p with a normal shutter angle. That is a strike against shooting 4K 24p.

2) Because using the "post slow down" technique relies on software, there is a chance that between now and next summer when I post it, there will be even better Twixtor/AE Timewarp software. Isn't a new version of Kronos coming out soon, for example? Also, if I shot a scene at 4K 30fps, for example, and Twixtor could not slow it down without distortions, is it possible that I could run the same shot through AE/Kronos and it might work? I imagine that each of these interpolation softwares have different algorithms that might work in different ways? So it one fails, the other might succeed?

As Mike Most said, I will definitely test this myself, shooting 3K 50fps vs 4K 30fps X2 post slowdown. I was just curious if anyone here had actually done that already and had thoughts about it?

It does sound like I will plan to finish the film at 4K RGB now. It's just a question of whether I should be shooting at 3K fps right now on MX, or 4K 30fps with the intention of doing a post slowdown. And that might require me to shoot with a 90-degree shutter, right?
 
1. I've used Twixtor since 1998 and it really hasn't progressed much. The biggest thing you can do to improve quality with ANY retimer is to help guide it by hand. EG you draw masks to isolate the foreground from the BG. The other thing you can do in something like Flame's retimer is manually re-brush the vectors. Again, I'd go 3K!

2. Yes, you can do the trick of combining multiple slowmo algorithms. I did that last week - combined Kronos (built into AE) with Twixtor with Boris with varying levels of opacity. Made things a tiny bit smoother. It was for a crazy rush job (editor says "hey man, can you make something half-decent in half an hour?) so not enought time to do all the masks properly - it was a crowd scene.

3. Yes Anson, please share your experience with Red and point us to the thread - or mass PM us :) The hard learned lessons are especially valuable (despite the fact that you usually crazy zealots yelling at you for daring to "criticise" RED).

Bruce Allen
www.boacinema.com
 
I would much rather go for quality than 50fps. Of course effect from framerate is not direct, but a bit more motivated by the motion of subject. Sometimes 50fps would be perfect. Still, it is a low ceiling to get in exchange for x4 the detail IMO. Add to that the fact that your project screams for resolution and to me it is a no brainer. I think that with such a project you would greatly benefit from the best textures you could get.

Also, from what you have posted thus far I would not jump straight from 5.6k to 3k. You would need, not only a great motivation for slow motion, but also a stepping stone to ease the abrupt jump in resolved detail.

Unfortunately for you (and many more) RED didn't come out a year earlier than it did. Their sensor program, codec, and workflow are all up to what you want to do, today. As to the body however, you want their next. You are filming something that is, literally, visually Epic. Cudos to Jim for such a proper name.

Just trying to selfishly convince you to do what I want Tom, you obviously know all of this. I know I will be seeing this on a 4k Sony somewhere in LA.

Then if you can get a few 120k 5k shots in time for the Blu-Ray release, so be it! More fun!
 
Content is king

Content is king

Tom/Roberto -

I have to disagree with Roberto here for your purposes - sometimes a particular overcranked shot is compelling enough to make the resolution drop worth every (un)pixel. I too am obsessed with quality and thus resolution. I too worship Fricke and his methods as well as his technical choices, mastery and output. I too have lost sleep over shooting 2K (in my case, the analog for your 3K debate). But let me give you a tangible example of content overwhelming pixel count:

We spent 3 days trying to capture a specific kayaking shot, sitting on a harness in the rain, slinging the massive canon 400mm prime on red over a steep cliff, and waiting for world class kayakers to run a dodgy section in the right way. We shot the same run (targeting 1080p bluray finish at 24fps) at 4K 30fps, 3K 60fps (2:1 zoom later for 16:9, another debate) and 2K 100FPS. We got good shots at all resolutions, but the best shot content wise was 2K.

We did not like it technically. It was soft, and had minor focus and exposure issues to boot. But we did some twixtoring for a test and showed it to some folks. They gasped. The same shot at 3K and 4K does not work at all. Same shot, same framing, give or take, same run - but very different framerate.

We have that shot in our 10 minute festival preview, and invariably when we show the film preview - cutting from 4K, 3K and to this soft 2K (which is then twixtored and ramped to 220FPS in spots, making it even softer) - people ALWAYS respond to this shot most. It works. It communicates our message in a way other shots don't. Its soft compared to other shots. Its flawed technically in other ways. 900 people 2 weeks ago at a fest all made loud WHOA noises during the shot.

The audience does not care that its soft. Content is king. We often have to remind ourselves of that a we obsess over technical details - don't let perfect be the enemy of good. Like everyone else here, I have the utmost respect for your work and your mission - heck, I learned DSLR timelapse from your site! The audience responds to a feeling, a message, your vision - yes, they respond better when its technically as good as it can be. But they will respond nearly as well when its not, as long as the content is strong.

The shot I describe above is the last half of this trailer - you'll see what I mean I think:

http://www.vimeo.com/11540206

Discuss. :-)
 
Bruce Allen said:
The biggest thing you can do to improve quality with ANY retimer is to help guide it by hand. EG you draw masks to isolate the foreground from the BG. [/url]
Click to expand...

And for many situations this became quite a bit easier with the new Roto-Brush in CS5. Apart from that I absolutely agree with all the notes from Bruce.
 
Even if you factor Moore's law into the time NTSC/PAL has been around (50+ years) 1080p will be THE standard the next 20 years. Too many people invested too much money into HD technology - we are only in the beginning of 1080p as a worldwide standard - that a new higher resolving standard will be out soon, except cinema of course.

But if projects like Avatar have been done in 1080p, despite the fact that 4K raw was around already when Avatar was shot, that shifts the discussion's perspective towards content vs resolution.

A few post above Anson says "content is king" and brought a nice example how content overrides technical shortcomings. 1080p will give you much more choices in acquisition, be it a 1000 fps shot with a Phantom or a 3D rendering that would have been too expensive if rendered out above 1080p.

Post production 4K is 5 times slower than in 1080p and I for my part find that 1080p is even for the fastest boxes and the cleverest applications quite a mouthful to swallow and very time consuming.

Furthermore, I would at least think twice for which market my product is target. Does a 4K finishing justify the higher expenses in postproduction and the reduced choice in acquisition tools? Will 4K make the ROI higher? Is 4K realistic? Will the 4K mantra turn out as an obstacle, perhaps a show stopper?

My main goal is always that the project will succeed. A realistic time frame is key, everything else has a subordinate role, including resolution.

Hans
 
Hans von Sonntag said:
Even if you factor Moore's law into the time NTSC/PAL has been around (50+ years) 1080p will be THE standard the next 20 years. Too many people invested too much money into HD technology - we are only in the beginning of 1080p as a worldwide standard - that a new higher resolving standard will be out soon, except cinema of course.

But if projects like Avatar have been done in 1080p, despite the fact that 4K raw was around already when Avatar was shot, that shifts the discussion's perspective towards content vs resolution.

A few post above Anson says "content is king" and brought a nice example how content overrides technical shortcomings. 1080p will give you much more choices in acquisition, be it a 1000 fps shot with a Phantom or a 3D rendering that would have been too expensive if rendered out above 1080p.

Post production 4K is 5 times slower than in 1080p and I for my part find that 1080p is even for the fastest boxes and the cleverest applications quite a mouthful to swallow and very time consuming.

Furthermore, I would at least think twice for which market my product is target. Does a 4K finishing justify the higher expenses in postproduction and the reduced choice in acquisition tools? Will 4K make the ROI higher? Is 4K realistic? Will the 4K mantra turn out as an obstacle, perhaps a show stopper?

My main goal is always that the project will succeed. A realistic time frame is key, everything else has a subordinate role, including resolution.

Hans
Click to expand...

All of these arguments keep getting made but the fact of the matter is that we don't know what the standars are going to be in 2,5,10 years.

Tom it depends on what makes your shoots possible, shoot at your highest possible resolution while not crippling your creative possibilities, then when your finished shooting decide what to finish in. Whos to say you choose 4k and the 3k doesn't scale the way you would like it to, or you shoot 1080 and it doesn't work either. Work at the max resolution that you need to, if you need the over cranking work at the lower resolution. Whatever you do I can see that your content is going to be the king, so just do it as much justice as you can resolution wise, but don't cripple your content for resolution. With your style of project its best to keep yourself as flexible as possible while shooting, you don't want to have to think twice about lowering res and over cranking if it means you get that shot. The shot you can only ever get once.
 
Anson Fogel said:
We did not like it technically. It was soft, and had minor focus and exposure issues to boot.

Discuss. :-)
Click to expand...

Interesting post, Anson, and certainly something that works with documentary production, where my mantra has been - "There is no such thing as bad footage, only bad editors" - you can do a lot with a little, in the context of telling a good story. The individual greatness of shots means more to shooters than it does to editors who are sometimes forced to "kill the darlings" because they just don't fit the story. We all hate cutting good material - or being forced to use "less good" material, in some circumstances, but it can be a necessary evil. It is part of the process of discovery inherent to documentary producing.

For Tom's purposes - a Baraka-style film - it's tough, because the expectation is that every shot will be perfect. That is how high the bar preceding his film is set. Which I'm sure he will achieve.
 
Anson Fogel said:
Tom/Roberto -



We shot the same run (targeting 1080p bluray finish at 24fps) at 4K 30fps, 3K 60fps (2:1 zoom later for 16:9, another debate) and 2K 100FPS. We got good shots at all resolutions, but the best shot content wise was 2K.

We did not like it technically. It was soft, and had minor focus and exposure issues to boot. But we did some twixtoring for a test and showed it to some folks. They gasped. The same shot at 3K and 4K does not work at all. Same shot, same framing, give or take, same run - but very different framerate.

We have that shot in our 10 minute festival preview, and invariably when we show the film preview - cutting from 4K, 3K and to this soft 2K (which is then twixtored and ramped to 220FPS in spots, making it even softer) - people ALWAYS respond to this shot most. It works. It communicates our message in a way other shots don't. Its soft compared to other shots. Its flawed technically in other ways. 900 people 2 weeks ago at a fest all made loud WHOA noises during the shot.



http://www.vimeo.com/11540206

Discuss. :-)
Click to expand...
A loud WHOA from here too Anson! Fantastic trailer, good luck with your film.
Matching the framerate to the action is key I think.Water likes 120 to 220 at least to freeze the drops. Shutter on stills needs to be that. If you had used say 2000 fps the water AND kayak would be frozen. Boring. (Unless you want to dwell on one moment.) Fire would need a faster framerate. From night skies to walking, tossing hair, to capturing a bullet a certain clock speed best fits the action.​
 
For shots like Anson's that reach a higher level of impact by legitimate overcranking I think the answer is clear: choose your frame rate, live with the resolution limits. Keep in mind that for many viewers the dreamy quality of re-timed footage is within expectations and the visceral power of a compelling image in slo-mo provides its own punch.

Because of the type of piece you are doing a 4K finish makes sense. I have been working with some folks in an attempt to devise an affordable 4K DI pipeline. Feel free to PM me at the appropriate juncture for more info.

All the best with your project.

Cheers - #19
 
4K all the way Tom.

And when I get my first 4K display I will be hitting you up for a 4K version of your film. Of course I will pay extra and I can't think of anything else I'd rather watch on my 4K display.
 
Anson, no need to remind me that content is king. ;) But I think you didn't read my post right. I believe that having framerates available is very important, that is one of the reasons why I LOVED my HVX and had to buy one. However, in this case, I would go with 4k all the way, and hope to get an Epic later.

Nice shot!!
 
I agree that content is king! And Terrence has also changed my life...

I've got some less than perfect quality stuff that people still love...

Think about audio. The songs you love from the past (The Tom Lowe of music)Pink Floyd for example, recorded using less than perfect methods according to today's standards.
With music, sometimes it's not the expensive studio suite and fast hardware that makes the song. Sometimes it's just some wood, strings, a microphone and a bit of audacity...
 
Last edited:
You must log in or register to reply here.
Back
Top