Film transfers or Digital Cinema
Having spend a good part of a decade doing film scanning and recording, I find what happens in post plays a big part in how video looks when it's transferred to film.
Before DVX100, since most video to film transfers were done from 60i originals, it was important to minimize the motion stuttering that occurred when doing 60i to 24p conversions.
Since most people were letting the transfer house use their own software to deal with this conversion. The problem would be that the clients wouldn't know what they were getting until they saw a print.
The same with image artifacts. Its been mentioned numerous times that any image enhancing or sharpening should be turned down on the camera when shooting. The same applies for image enhancements done in post.
I have seen commercials shot on film, telecined to D1 tape, heavy post enhancement done, look slick and great on an NTSC monitor, but when taken to film, look crawly, aliased and sometimes extremely noisy. This is because the images were designed to look good on video, not film. The transfer to film is almost always an afterthought.
Even though indies have better tools, the same rules apply.
If an indie is seriously considering even a remote possibility of a film transfer, the transfer house should be involved as early into preproduction as possible.
Since most indies have limited funds, it makes more sense to use the affordable tools available to them. Sure, not all indies are techies, but forums like this are there to help out.
I would recommend that before color grading or visual effects stages, the media be converted to a 10bit or higher lossless format.
I have heard it's possible, if using a Kona or Black Magic HD card, you can just drop 8bit shots into a 10bit timeline, and still get realtime feedback.
You definitely need the extra headroom in a 10bit or higher format to avoid color clipping in the high ends or crushing in the low ends.
FCP lets you work in floating point, so if you may want to try doing some tests in that.
Even if a feature goes through a digital intermediate for a theatrical release , separate color timing is done for the video masters.
It would make sense to do the same on an indie project. Do all the color timing and enhancements for the video version, then make a "cleaner" version that would be more complementary for a filmout.
This is where it gets critical to work with the transfer house to make sure that the "cleaner" version will look good on film. The transfer house should have their system calibrated to show what your image should look like on film. This way you can head off any problems that crop up.
If you are using mixed source footage which run at different frame rates, you will have convert that footage to 24. Post houses have Teranex or Alchemist systems are designed for that sort of work, but are very expensive. You can also use plug-ins like Graeme Nattress' Standards Conversion, Red Giant's Magic Bullet or Algolith's AlgoSuite.
If you are a FCP user, Compressor allows you to do standards conversion also, using the optical flow code used in Shake, though it may be extremely slow even on a G5.
The flip side of all this is that there isn't any real need for the expense and hassles of a filmout if indies just go the digital cinema route.
Currently there is something like 170 digital screens nationwide out of 36,000 traditional cinema screens. There has been estimates that something like 6,000 digital screens are planned for 2006. It will be interesting if this turns out to be true, but that ball is rolling. There is definitely a desire at least on the studio and filmmaker side for digital to be rolled in, it's more resistance on the theater side because of the costs.
Landmark Theaters owned by Mark Cuban already announced their plans to convert 60 screens to digital, and have set up a relatively "affordable" pricing structure for filmmakers who want show their films digitally.
I have heard that even though some festivals show films digitally, it's not always a properly setup system or in a choice venue, and 35mm film features get the preferential treatment.
I feel that a theatrical release is bragging rights for the filmmaker. Sure it's cool to say that a film is playing on the big screen, but I don't find it to be critical that a film needs a theatrical release to find it's audience.
Before DVX100, since most video to film transfers were done from 60i originals, it was important to minimize the motion stuttering that occurred when doing 60i to 24p conversions.
Since most people were letting the transfer house use their own software to deal with this conversion. The problem would be that the clients wouldn't know what they were getting until they saw a print.
The same with image artifacts. Its been mentioned numerous times that any image enhancing or sharpening should be turned down on the camera when shooting. The same applies for image enhancements done in post.
I have seen commercials shot on film, telecined to D1 tape, heavy post enhancement done, look slick and great on an NTSC monitor, but when taken to film, look crawly, aliased and sometimes extremely noisy. This is because the images were designed to look good on video, not film. The transfer to film is almost always an afterthought.
Even though indies have better tools, the same rules apply.
If an indie is seriously considering even a remote possibility of a film transfer, the transfer house should be involved as early into preproduction as possible.
Since most indies have limited funds, it makes more sense to use the affordable tools available to them. Sure, not all indies are techies, but forums like this are there to help out.
I would recommend that before color grading or visual effects stages, the media be converted to a 10bit or higher lossless format.
I have heard it's possible, if using a Kona or Black Magic HD card, you can just drop 8bit shots into a 10bit timeline, and still get realtime feedback.
You definitely need the extra headroom in a 10bit or higher format to avoid color clipping in the high ends or crushing in the low ends.
FCP lets you work in floating point, so if you may want to try doing some tests in that.
Even if a feature goes through a digital intermediate for a theatrical release , separate color timing is done for the video masters.
It would make sense to do the same on an indie project. Do all the color timing and enhancements for the video version, then make a "cleaner" version that would be more complementary for a filmout.
This is where it gets critical to work with the transfer house to make sure that the "cleaner" version will look good on film. The transfer house should have their system calibrated to show what your image should look like on film. This way you can head off any problems that crop up.
If you are using mixed source footage which run at different frame rates, you will have convert that footage to 24. Post houses have Teranex or Alchemist systems are designed for that sort of work, but are very expensive. You can also use plug-ins like Graeme Nattress' Standards Conversion, Red Giant's Magic Bullet or Algolith's AlgoSuite.
If you are a FCP user, Compressor allows you to do standards conversion also, using the optical flow code used in Shake, though it may be extremely slow even on a G5.
The flip side of all this is that there isn't any real need for the expense and hassles of a filmout if indies just go the digital cinema route.
Currently there is something like 170 digital screens nationwide out of 36,000 traditional cinema screens. There has been estimates that something like 6,000 digital screens are planned for 2006. It will be interesting if this turns out to be true, but that ball is rolling. There is definitely a desire at least on the studio and filmmaker side for digital to be rolled in, it's more resistance on the theater side because of the costs.
Landmark Theaters owned by Mark Cuban already announced their plans to convert 60 screens to digital, and have set up a relatively "affordable" pricing structure for filmmakers who want show their films digitally.
I have heard that even though some festivals show films digitally, it's not always a properly setup system or in a choice venue, and 35mm film features get the preferential treatment.
I feel that a theatrical release is bragging rights for the filmmaker. Sure it's cool to say that a film is playing on the big screen, but I don't find it to be critical that a film needs a theatrical release to find it's audience.