Shooting in ProRes What version is best for sit down interviews?

All those amounts are way less than non raw? Making raw far preferable on the size metric- the opposite of production house myth
 
NorBro said:
You making up these numbers? lol

As mentioned above, it's 471 for 422.

And 328 for LT and 707 for HQ.

View attachment 141046
Click to expand...

No, I'm not. Those numbers came from a data calculator. Interestingly (or not) I checked the numbers on another data calculator (AJA) and it gives the figure 530.79 Mb/s for QHD at 24p but only 238.5 GBs an hour. I checked them again on Arri's data calculator and they said 553Mb/s and 240GBs an hour. I don't know where you got your figures from but I do know I didn't make mine up.

You can apologise anytime you like.
 
Back when the Ursa Mini came out, and CFast cards were really expensive (not that they're cheap now), I remember watching a video where a guy resorted to shooting in Prores Proxy for long interviews. Hopefully it turned out okay for him, but I agree with everyone that LT is recommended minimum.
 
Liam Hall said:
No, I'm not. Those numbers came from a data calculator. Interestingly (or not) I checked the numbers on another data calculator (AJA) and it gives the figure 530.79 Mb/s for QHD at 24p but only 238.5 GBs an hour. I checked them again on Arri's data calculator and they said 553Mb/s and 240GBs an hour. I don't know where you got your figures from but I do know I didn't make mine up.

You can apologise anytime you like.
Click to expand...

No offense, but that's why there is always confusion because of doing silly things like that.

Stop using data calculators, man - and just post the numbers from the people who created it.

Page #24 - https://www.apple.com/final-cut-pro/docs/Apple_ProRes_White_Paper.pdf

Apple's ProRes Chart.jpg
 
NorBro said:
No offense, but that's why there is always confusion because of doing silly things like that.

Stop using data calculators, man - and just post the numbers from the people who created it.

Page #24 - https://www.apple.com/final-cut-pro/docs/Apple_ProRes_White_Paper.pdf

View attachment 141048
Click to expand...

To be fair, those are "target" bit rates, not absolute. They even use the word "approximate". Encode the same video file to ProRes using two different systems you'll get two different bit rates. Also, I don't think it's silly to use a data calculator made by a manufacturer of the type of recorder the OP is talking about using or one by a trusted manufacturer like ARRI. These things are guides.
 
Silly wasn't the best adjective.

But you know what, I think you're actually right.

When critically thinking about it (which I always preach on here), I've changed my mind.

In a video production world where technical attention to detail doesn't matter as much anymore, and specific standards change on a daily basis, there shouldn't be one figure from one white paper as a universal reference for information.

Everyone should share their own numbers and their own results based on their own hardware and software configuration and what the machines calculate based on their analyzation of what is in front of the lens/sensor.

(That may sound like I might be being sarcastic, but I'm not, lol.)
 
No, it's not sarcastic. I agree, though I would add, people need to use their EYES too and add that into the equation.

Actually, I searched for that White Paper and only came up with the one from 2013 which is why I used the calculator!
 
Liam Hall said:
To be fair, those are "target" bit rates, not absolute. They even use the word "approximate". Encode the same video file to ProRes using two different systems you'll get two different bit rates. Also, I don't think it's silly to use a data calculator made by a manufacturer of the type of recorder the OP is talking about using or one by a trusted manufacturer like ARRI. These things are guides.
Click to expand...

Really? I thought ProRes was constant bitrate whereas ProRes raw is variable
 
Apple calls it a variable bit rate in the white paper (which I could have sworn I didn't see that before over the years).
 
I checked some from Blackmagic earlier and they varied.

But didn't have anything else (I archive everything in H.264) to compare them too (i.e. a different scene or different hardware like Atomos' ProRes or RED's ProRes, etc).

A good test would be to set up two pieces of equipment and film the same scene and compare files.
 
I think variability anywhere proves variability

But what are those data calculators reporting? Maxima? Seems like they're not reporting the target rates
 
I just checked a few ProRes files in Quicktime:

* from my FS5/Atomos, DCI 4K 60p: ProRes 422 was 1.13 Gbit/s; ProRes LT was anywhere from about 720-800 Mbit/s

* FS5/Atomos, DCI 4K 24p: ProRes LT was 300-315 Mbit/s

* BMPCC4K, UHD 24p: ProRes HQ was about 700 Mbit/s; ProRes LT was around 320 Mbit/s

* OG BMPCC, HD 24p: ProRes HQ was about 178 Mbit/s
 
ahalpert said:
I think ProRes raw is well worth exploring. It might be more useful for concerts, though, because if your white balance and exposure are well set in camera, you may not need the power of RAW. (That being said, if you have a high dynamic range scene and very bright highlights through a window, for example, I think that ProRes raw can help you there.)

One caveat about shooting H.265 is that if you can record proxy files while you're filming, it can make the editing process painless while giving you very small file sizes. You start editing with the proxy files, which playback smoothly. Then you export the final project with the original files. the export would go faster with ProRes files, but typically your final video is much shorter than your total footage captured, so it goes pretty quickly anyway.

And you could certainly do it with the Ninja V by recording H265 in-camera and ProRes Proxy (possibly at lower resolution such as HD) in the Ninja V. Food for thought
Click to expand...

So you are saying it is best to not use RAW for interviews but for concerts? If that's the case wouldn't just ProRes LT be sufficient enough with interviews? I have an upcoming project (assuming COVID does not cause another shutdown) and I will be interviewing about 5 people for about 30-1 hour each
 
Besides saving space, the other main reason for shooting interviews in a lower quality is because most final products have interviews that are mixed with b-roll and you rarely see people speaking on screen for more than 20-30 seconds at a time.

(If your entire project consists of someone speaking for 1 hour then it's different.)

As far as the recommendation for shooting RAW in environments such as concerts, it's because they might be (and usually are) more challenging than most interviews which you normally get to set up and control. Plus the b-roll is probably more precious than a talking head (hence the desire for the better IQ).

Most would agree RAW for interviews is overkill unless the project is truly in the 1% paying you millions. (But some still shoot RAW for everything, regardless.)
 
NorBro said:
Besides saving space, the other main reason for shooting interviews in a lower quality is because most final products have interviews that are mixed with b-roll and you rarely see people speaking on screen for more than 20-30 seconds at a time.

(If your entire project consists of someone speaking for 1 hour then it's different.)

As far as the recommendation for shooting RAW in environments such as concerts, it's because they might be (and usually are) more challenging than most interviews which you normally get to set up and control. Plus the b-roll is probably more precious than a talking head (hence the desire for the better IQ).

Most would agree RAW for interviews is overkill unless the project is truly in the 1% paying you millions. (But some still shoot RAW for everything, regardless.)
Click to expand...

Makes Sense. What about the ProRes LT vs 422?
 
You must log in or register to reply here.
Back
Top