F3: F3 with 4:4:4 or Canon 1DC for theatrical distribution

[Movies by Matt]

Just curious about people's thoughts. I understand they are slightly different tools, but if the end goal is theatrical distribution. I'm wondering if the 4:4:4 10 bit 1080 to a Pix from the F3 is going to hold up better than 4:2:2 8bit 4K (Motion JPEG) on the 1DC.

Personally I'm more focused on documentary shooting, so the 1DC size is really attractive compared to the F3 and a pix, but the used market for F3's is quite tempting compared to being an early adopter of another DSLR.

All comments are appreciated.

 

[Grug]

As awkward as the F3 is ergonomically, it remains the dedicated video camera of the two. It has built-in ND filters, proper audio options, a far wider range of lenses it can use, and substantially higher dynamic range (in SLOG) than the Canon.

I've seen some lovely sample images from the 1DC, so I'm sure you could do good things with it - but I'd always choose the dedicated tool where possible.

The way I tend to look at it is this - SLOGged F3 seems to offer the closest output to the Arri Alexa of any of the current crop of large-sensor cameras, and given that the Alexa seems (image-wise) to be the pick of the bunch in digital cinema cameras at the moment, I think the F3 is the way to go if you're planning on theatrical distribution.

 

[Kolor-Pikker]

Also, the F3 is a proven production tool, I'm not sure how many people have experience with the 1DC. On the other hand, the 1DC looks like an ordinary camera and will be more... "comfortable" to use around people not used to being filmed (seeing as it's a documentary).

Both cameras will turn out fine images, pick one that suits your comfort zone better, and focus on audio instead. The only reason people go to theater anymore is for big awesome sound.

 

[Movies by Matt]

Thank you. I've weighed all the ergo & audio capabilities. I definitely don't neglect audio. Currently running a 664 with 4 channels of lectro & a 416. Am currently dreaming of a Holophone with SD-CL6 to cover my audio, but that comes after a camera & heavily leaning to the 1DC for my needs. Just thought I'd see what people thought about the codecs & if there's some huge negatives for motion Jpeg or conversely huge benefits to 4:4:4. With consistent white balance taken during shooting and no VFX, and keeping color grading to a minimum, yet still having some options to work with in post.

 

[starcentral]

There are some good threads on the topic of codec vs codec when the bit-depth is the same and color sampling is the same or even different. Conclusions seem to be there are no major discernible differences when simply viewing footage, but differences can start to become apparent when grading or other type of post work is done on the footage. Codec robustness is one thing, bit-depth is another, and color sampling yet another.

As far as 10-bit vs. 8-bit, I sound like the 8-bit advocate here in many threads on DVX but for subtle grading I've had no problems using the internal F3 codec shooting straight to SXS in 8-bit 420. Most of my work is meant for a cinematic film type of look so I control lighting, shoot consistent exposures, work with a limited contrast range in a scene, and am not shooting "wild". For myself I can make it work. Yes more bit-depth is better and will give you more flexibility in post - but do you always need it?

I'm able to go from this to this without any major compromise to the image:

Some of my work using the F3 has screened in film festivals on large theatre screens up along with other projects shot on the 5D, 7D, RED and even film. In one film festival where all material was mastered and submitted using ProRes HQ it was easy to spot the films shot on film but the footage from the F3 and DSLR's showed no major difference in terms of resolution, color and tonal information (believe me I was surprised) and all of it looked really really good. RED originating material seemed somewhere in-between film and the F3/DSLR's. How much better would 10-bit originating footage had been? Again it depends on what you shot, how you shot it, and how much you pushed or pulled in post and what you did with your footage after. I've seen 10-bit footage that does look sharper and held a little more detail in the highlights but I would not proclaim night and day differences. On the other hand I graded some RED footage on a project last year and WOW that latitude is insane I must say! But you're talking 16-bit, 65 times the amount of information over 10-bit !!!

10-bit is only 4 times the amount of info as 8-bit and most of that extra data ends up representing the bright tones, less data in the mid-tones, and even much less data in the dark tones. The only exception is if you shoot SLOG using 10-bit then at least you are getting a more uniform distribution of data per stop throughout the entire tonal range which will show its advantage in post and also contribute towards more of a "film-like" look with "film-like" characteristics.

All that being said, I think you will be fine with either camera and format. It's really not the tool so much as what you actually do with it. :happy:

 

[atarijedi]

Also look at what kind of documentaries you would be working on. The 1DC with L-series lenses will give you more latitude if you get into the dirt, grit, and wet, since it would be a weather-sealed package. But that is only beneficial if you are going to be doing that kind of work.

 

[morgan_moore]

Small HDMI.

This makes me have no interest in the canon as an A camera.

I dont know if it is still the case but the 5d2 not only do you lose signal but the shot cuts if you get an 'outage' of signal to the monitor

S

 

[Teodor Miljevic]

There are some good threads on the topic of codec vs codec when the bit-depth is the same and color sampling is the same or even different. Conclusions seem to be there are no major discernible differences when simply viewing footage, but differences can start to become apparent when grading or other type of post work is done on the footage. Codec robustness is one thing, bit-depth is another, and color sampling yet another.

As far as 10-bit vs. 8-bit, I sound like the 8-bit advocate here in many threads on DVX but for subtle grading I've had no problems using the internal F3 codec shooting straight to SXS in 8-bit 420. Most of my work is meant for a cinematic film type of look so I control lighting, shoot consistent exposures, work with a limited contrast range in a scene, and am not shooting "wild". For myself I can make it work. Yes more bit-depth is better and will give you more flexibility in post - but do you always need it?

I'm able to go from this to this without any major compromise to the image:

Some of my work using the F3 has screened in film festivals on large theatre screens up along with other projects shot on the 5D, 7D, RED and even film. In one film festival where all material was mastered and submitted using ProRes HQ it was easy to spot the films shot on film but the footage from the F3 and DSLR's showed no major difference in terms of resolution, color and tonal information (believe me I was surprised) and all of it looked really really good. RED originating material seemed somewhere in-between film and the F3/DSLR's. How much better would 10-bit originating footage had been? Again it depends on what you shot, how you shot it, and how much you pushed or pulled in post and what you did with your footage after. I've seen 10-bit footage that does look sharper and held a little more detail in the highlights but I would not proclaim night and day differences. On the other hand I graded some RED footage on a project last year and WOW that latitude is insane I must say! But you're talking 16-bit, 65 times the amount of information over 10-bit !!!

10-bit is only 4 times the amount of info as 8-bit and most of that extra data ends up representing the bright tones, less data in the mid-tones, and even much less data in the dark tones. The only exception is if you shoot SLOG using 10-bit then at least you are getting a more uniform distribution of data per stop throughout the entire tonal range which will show its advantage in post and also contribute towards more of a "film-like" look with "film-like" characteristics.

All that being said, I think you will be fine with either camera and format. It's really not the tool so much as what you actually do with it. :happy:

If you have a perfectly exposed picture it is true that the difference is small(like the example yoy showed). But if you have less then ideally exposed material, log encoded material (menaing you often waste at least two of those precious 8bits already (because of the lifted shadows and compressed highlihts)) and low contrast material the difference after you have graded it properly is quite large I think saying anything else is quite misleading.