Locked-off still-life: Film v. Digital...

[Leo Ticheli]

There seems to be a very subtle, but noticeable difference between film and digital with locked-off still-life shots; film appears to me to have some "life" to it because each individual frame is not a perfect replica of the series, whereas all digital frames are exactly the same.

Some of this may be due to very slight gate weave, some due to variations in the emulsion. Are there other factors? Dancing grain?

While I'm sure there's no large body of cinematographers overly concerned about this, I wonder if it's possible to design a logarithm that would emulate this effect, either in camera or post.

My own method of dealing with this is to avoid sitting on a still-life, but keep the camera moving just a bit; it's not always possible of course.

I know it's not a burning issue; just some Friday morning musing...

Good shooting and best regards,

Leo

 

[Paul Leeming]

What about underexposing a stop or two, then pushing the ISO up in post to make the noise more prominent?

Wow, isn't it funny that we even need to contemplate how to get back some of the noise of film in order to make it feel 'alive'?!

HTH

Paul

 

[Leo Ticheli]

Thanks, Paul,

I'm not certain that it's noise or grain that's primarily responsible, nor am I an advocate for either, but you could be right.

I was mostly wondering if others notice this effect, of locked-off still-life shots seeming a bit sterile.

As I mentioned, my solution has been to keep a little subtle movement and, sometimes, add some movement to the lighting, as if it was through trees or a moving sheer curtain.

Good shooting and best regards,

Leo

 

[Jeff Coatney]

Yes, with Film, the "photosite" (emulsion grain) is animated from frame to frame. Add gate-weave and variances in the pressure plate contacting the gate and holding the negative steady and then add halation (with moving grain) and i think you could account for this look algorithmically. I think you'd need to start with a large sampling of image data, like 5K at least in order to finish at 4K. The best results would be from 4K or 5K+ sampling for a 2K finish.

 

[Rob Ruffo]

You're right - that's why I avoid 100% still shots - a tiny amount of movement, like from a Cineslider and the weird dead look goes away.