-
Framerate Conversion: 23.98 Timeline With Some 29.97 Clips
I have a project that is being edited in a 23.98fps timeline. The majority of the footage was shot on a RED at 23.98. There are some old “archive” and stock video clips that will be cut into this edit. These clips are 29.97.
If I understand correctly, I’ll get the best results from the framerate conversion if I convert the 23.98 footage up to 29.97. Is this correct, especially considering that the majority of the sequence is 23.98? If this is the case should the conversion be the last thing I do before exporting the final master or should it be the first thing I do before I apply any effects, transitions, etc.?
-Andrew Saliga
-
12 Replies
-
In addition, some of the SD clips need to be upconverted to 720P. I’ve heard that using the Kona hardware to upconvert will give me better results than using Commpressor.
If the solution is converting the 29.97 footage to 23.98, should I upconvert it first, and then reduce the framerate? Should I perform both conversions in the same Compressor batch, or should I upconvert via Kona then run through Compressor?
-Andrew Saliga
-
[Andrew Saliga] “Is this correct, especially considering that the majority of the sequence is 23.98”
If the majority of the content is at 23.98 the down-convert the 29.97 to that, not the other way around. It’s the first thing your should do before adding anything else.
In my opinion.
__________________________________________________________________
Two years from now, spam will be solved. – Bill Gates, World Economic Forum 2004 -
I found an article on Ken Stone’s site…
https://www.kenstone.net/fcp_homepagE/converting_frame_rates_compressor.html
He says,
“In a situation where you have two different frame rates and need to convert one to match the other, there are two directions that you can go. One will produce good results, the other not so good. Let’s say that we have footage shot at 24P and 30P. If we convert the 24P footage up to 30P, the full 24 frames will be used and an additional 6 frames will be created, for a total of 30. In both FCP and Compressor, the additional 6 frames will be based on the information contained in the original 24 frames. This is the correct way to convert frame rates so they match.
The other way to go would be to take the video shot at 30 frames and reduce it down to 24 frames. This would not be the way to do frame rate conversions because, if you reduce 30 frames down to 24, you have to throw away 6 frames and those 6 frames contain an important amount of information (particularly regarding motion). Doing frame rate conversions in this manner will produce stuttery video. So, when doing frame rate conversions in either FCP or Compressor, always convert the lower frame rate up to match the higher frame rate.”
I’m thinking that the 29.97 footage should probably stay as is, since that’s the final output framerate. Throwing out those 6 frame only to add them back in later seems like a bad idea.
I’m still unsure about when do the conversion, although I’m Googling around.
-Andrew Saliga
-
Ken is totally right and I totally agree with him, however you situation is different as you said that the majority of your footage is at 23.98 (or is it 24p?)
Also you didn’t say that your final output was to 29.97 (is that i or p) or is it 30p?
__________________________________________________________________
Two years from now, spam will be solved. – Bill Gates, World Economic Forum 2004 -
The majority of the footage is 23.98fps.
Final output will be via an LCD projector or LCD screen. The client wants SD outputs, so 29.97fps was the plan, and I’d assume this should be progressive since it won’t be shown on an interlaced montior.
-Andrew Saliga
-
How much of the footage is SD?
Any particular reason why the client wants SD?
__________________________________________________________________
Two years from now, spam will be solved. – Bill Gates, World Economic Forum 2004 -
I’d say less than 15% is SD. They are just quick clips only a few seconds long.
As for the client wanting SD, haha…who knows. The main reason it was shot with a RED at 4K is because there is a lot of compositing that needed to be done in post.
There are two different spots, and one they want in SD and the other in HD. Knowing the nature of clients, it could change.
-Andrew Saliga
-
Obviously your client need ‘educating’
There’s no point in shooting HD, then down converting to SD only to then upscale to an HD or HD ready display.
I’d advise them that all of the masters should be 1080p24, and the SD versions can be taken from those. Otherwise all your RED work (and money) will be wasted.
__________________________________________________________________
Two years from now, spam will be solved. – Bill Gates, World Economic Forum 2004 -
I’m not so sure I would go that far. It does seem that today if you plan on shooting with a Red camera that you will post and deliver in HD. Still, there might be times when everything is posted HD and then final output ends up on DVD or SD Station dubs.
I do think that in this case cutting everything at 29.97 1080p would be ideal. Convert your Red footage to 29.97 in compressor. If DVD is the end result, use a compressor like DVCProHD or PreRez.
Daniel’s statement is not wrong but he forgets that we used to shoot with very expensive film and downconvert everything to Beta SP for the edit. Then cut it at AVR77 and output back to Beta SP. Ouch!
So shooting with the Red is a big plus, but I also agree that getting those clips to 29.97 in some flavor of HD and upconverting the SD using the Kona 3 will give you the least headaches in post.
Good luck,
Rich Rubasch
Tilt Media -
I agree Daniel. There is no point in downconverting, only to upconvert again. Maybe I miscommunicated or perhaps the client’s odd request is confusing.
To clarify, two different video projects will be output. Different scripts, different visuals, etc.. One video is for a conference, and another for some other event. One is being requested in SD and the other HD. Both sequences have mixed content (SD/29.97 and HD/23.98), but each is currently set up to match the settings of the desired output.
To add to the confusion, I just learned that although we are doing the editing, the RED footage is going to be sent off for color correction. We will likely be shipping of a hard drive with a ProRes master of the edit along with the RED footage, and I’m not sure if we’ll receive the color corrected files on that same hard drive or if they’ll be sending a Digi-Beta master.
I realize that this added step may complicate matters to the extent of not having an answer unless the color correction house’s settings/output are known. If this is the case, how would you recommend doing the conversions, ignoring (for now) the fact that it’s going to be sent out for color correction?
Rich,
Thanks for the advice. I’m thinking that cutting both edits in some flavor of HD at 29.97 and then downconverting the one that will be output to SD would be a good approach. How would you handle the stock footage that is SD? Like Daniel was saying, I think HD masters would be the way to go, and would give me the highest level of quality possible when making the HD>SD downconversions, but I don’t want to uprez the stock footage only to downrez it on output.-Andrew Saliga
Reply to this Discussion! Login or Sign Up