I have done some experiments with Frame Mode. It looks gorgeous, but if you need critical focus on objects, it will work against you. The resolution loss make the image look soft at best.

Luckily there are other in-camera settings and post production work you an do to achieve the same look.

I have to add that the "softness" issue is a relative one. If you're used to looking at DigiBeta or BetaSP images shot with $50k cameras through $15k lenses, then the images will be soft, even when shot in normal TV mode. There's just no comparing the camera end of any sub $4k camera to a tip of the line broadcast workhorse. Something has to be lost in reducing the price and it's not just the housing.

On the other hand, if you're used to seeing Hi8, the softness is not even apparent. If you're used to the XL-1, the "softness" of the XL-1s is a definite improvement.

This entire thing is probably rather subjective. I think the frame mode with +1 sharpness in CP and one step in sharpness in post(unsharp mask) looks just as good as any standard mode frame. By the time I de-interlace the standard mode frame, it's lost any advantage in sharpness it had over the frame movie mode frame. I'd be interested to see the results of an optical rez test on the two.

That's very true Ozzie, "softness" is purely a subjective term. But it doesn't necessarily always mean "low resolution" either. Film has many times the resolution of NTSC, and yet it is often described as having a "softer" look. Which usually means, "less harsh"!

I have seen on many forums, this included I think, that folks have passed around the wisdom that to make video look more filmic, ya just slap on a Black ProMist and shoot away. Certainly I have done the same (and nets on the rear element, etc. etc.) in an effort to smooth out the tendency of video to emphasize detail in places you don't really want it, like on faces. I think what I dug about the Frame Movie Mode is that as well as creating a pleasing motion dynamic, the loss of resolution seemed to do the job of a short stack of filters in buffing out the hard edges.

I'm up to shoot a feature this spring which is oscillating between Digi Beta, HD and Super 16. If the digital formats win out for cost reasons (and I think they will), I'm planning on testing a full range of filters to take the edge of the video look--as that is my preference.

Call me "Mr. Softy"...

Are you sure you want to shoot it soft? If you are not going to up-res it to film I would shoot it soft. However, if the finished piece is going to film I would shoot several clips and start sending them out to various labs to judge their transfer methods and how your video holds up. The labs I've been talking to suggest turning the detail circuits off because of artifacts, not using filters and turning dynamic compression circuits on. The detail circuits cause a halo effect around objects which is very noticeable after transfer to film. The dynamic circuits tend to expand the contrast range of film. any softening can be added in post. Trying to sharpen an image that is too soft causes a hlo effect.

Jeff Donald

Charles,

I have always discouraged DPs from using Promist filters to create the "film look." All they seem to do is give that 1940s Veronica Lake look to closeups. They add nothing, in my opinion, to the video image. Sometimes a few DPs use a slight diffusing filter to soften harsh contrast like a bright window in the background. In instances like that I can see the benefit, especially with high end hi-resolution cameras. With the small chip cameras we're talking about it seems hardly necessary.

In the project I just completed (tonight we finished principal photography, but that's a story for another time) our DP originally opted to use a 1/8 promist on the XL-1s just to match it to the softer XL-1. I think that was a mistake and it compounded the softening we were already getting by going to film mode. We stopped using all difusing filters and switched to "straight" shooting and the images are very acceptable. We now have full control over the images in post. I'd rather have details I can soften than softness I can sharpen at the expense of increased noise.

My 2 cents...

I do mostly weddings and I use frame mode for the photo shoot and the post reception. All other footage is in regular mode. The 2 looks show "reality" and "party-mode" The frame mode to me is an "effect" (video look/film look). That's all. It's a "feeling" to me when I see it on TV.

What if we watched a movie like "The Terminator" in the "video look" it would probably suck. And the "live news" in the "film look" it would probably suck too. We seem to all be "adjusted" to this for some reason, yet it could of been the other way around I guess. Too late now! :)

One thing I discovered back last summer is that frame mode footage can be converted back into interlaced footage using RE:Vision's Twixtor in After Effects. What this program does is interpolate new frames in between frames for video and film, much like a 3D program like Bryce 3D does when you ask it to "tween" frames, not like the plain old frame blending in After Effects. It does an exceptional job at this, and has been used in major motion pictures for the purpose of super slo motion, say of the car crash segments in "The Fast and The Furious". I was able to take in the 30p footage, "tween" the frames and come out with 60p footage. Then with the 60p footage take and interlace those frames together into 30i, and viola, it looks like you shot the footage in normal (movie) mode in the first place. Hopefully this information will help those wafflers decide, and put and end to this discussion once and for all.

Nah, no way it will put the discussions aside -- without them, these boards would have no reason for existence! Personally, I like all the interaction and varied opinions. :)

Good info, though.......thanks for the post.

You're quite welcome, Vic. Like Kevin Bacon's character said in JFK, "The people gotta know." Later.

i think dvfilm.com has a program, either "dv film maker," or "atlantis," that can interlace and de-interlace video shot in normal or "progressive/frame movie mode," to make frame video from interlaced, or interlaced from frame video.

I own "DV filmmaker", and it is for selectively deinterlacing interlaced footage, dependent on motion. It does a decent job, although I prefer shooting in frame mode. It is not capable of reinterlacing frame mode footage. The only product I've found that is capable of doing that is Twixtor as I described in a previous post.

All the Frame/Interlaced discussions seem to revolve around whether you're planning on going to video/tv or to film...

Well, does anyone have a suggestion for those of us that don't know? For example, I'm getting ready to shoot a dramatic feature. Ideally, it would rock at Cannes and take home the Palm D'Or and get distributed by Mirimax on 1500 screens in the U.S. (and while I'm dreaming, I'd like a pony). However, the fact that the project may never see the emulsion side of a strip of polyester is definitely in my mind and if someone wanted to show it on IFC or the Sundance channel or even do a straight-to-video/dvd release, I sure as hell wouldn't scoff at them.

So, any suggestions on the best way to shoot so that it can be transferred nicely to film (in a perfect world) but can still have a nice filmic look if it stays in video?

Also, any opinions on the Cinelook plugin if one is planning on video/dvd distribution?

-Shawn McBee

Frame or Normal Mode
 

So what if a doc is shot with this in mind: a film transfer down the road for real film screenings at festivals,and maybe theatrical dist, but it will also have a life on broadcast or cable? Frame mode or Normal mode?

Consult your transfer service BEFORE you shoot. Some want interlaced footage, others may want frame-mode.