Info: How to build a film scanner (need advise & help, please) — Page 2

camroncamera said:

poita said:

You only need it for capture, so when capturing one reel at a time, and with lossless compression, you can get away with using that drive. You just capture until the drive is 80% full, and then copy it off to HDD storage, and then delete it and continue your capture.

PDB said:

poita, thanks for sharing all this info. Its great to read.

Mind if I ask some questions:

What format are you capturing in; Tiff? I assume the full bit depth of the camera, 12-bit?

With the earlier mention of lossless compression, I am wondering what capture format is ideal as well.

Why are you triggering the LED light source instead of leaving it always on? Is it to save on the life of the LED?

I think I can help on this one. If your film motion is continuous through your scanning rig with a constant light source and no shutter, your scan will be streaked as the film is pulled through the projector gate at the same instant the image sensor is making an exposure. The reason for triggering the LED light source is to freeze the film image in the gate for the imaging sensor. It works on the same principle as a flash photo that freezes action. (This flashing must be synchronized with the image capturing sensor, lest you capture half of one frame, the other half of the next frame, and the frameline between them.) If the motion of the film through the scanner is intermittent, that is, if the frames stops in the gate for a moment in the same fashion that a movie camera briefly stops each frame in the gate and then opens its shutter to make an exposure (often 1/48th second for 24FPS photography), theoretically you might be able to get away with a constant light source. http://forums.kinograph.cc/t/image-sensor-optical-components/44/37

Yes, you capture at the full bit depth available to the camera, and also can adjust the capture LUT to shift the bits to where they are needed most.
For most people’s computers, then yes you would capture to a 16bit TIFF sequence, or to an uncompressed 16bit .avi file. There are some codecs that work realtime if your system is fast enough if you want compressed images to save space, but for most people they would be capturing uncompressed. TIFFs are more demanding on disk speed and response than uncompressed AVI, but I prefer to capture to an image seqence rather than an AVI if the machine is up to it, as the TIFFs are usable without transcoding in a wider range of software packages.

Flashing the LED achieves a few things.

1. As mentioned, it freezes the film frame if using continuous motion, but even with an intermittent drive, flashing the LEDs freezes the image and stops any residual movement when the film is sitting in the gate, resulting in a sharper picture.

2. Heat. Flashing the LED means you don’t have to use heatsinks and cooling solutions, resulting in a more reliable and compact light source.

3. Colour. Using an array of RGB LEDs, you can adjust the colour mix by changing the individual flash duration of each colour. This lets you adjust the ‘white’ light to whatever mix is required for a particular film stock. For example, the faded THX 16mm print was restored to almost full colour just by changing the lighting during scanning. This results in better colour fidelity and more information captured.

4. Consistency. Changing the flash duration to control exposure gives you better consistency than driving the LEDs constantly and adjusting the power or using PWM to change the brightness.

5. Less flicker on capture than using PWM or other methods to adjust brightness as the light source is kept in phase with the camera trigger, and you are not getting brightness variance due to LED drivers.

First step for anyone is to get a projector, where you can remove the shutter and have access to the gate. That is the only thing that isn’t available ‘off the shelf’, all the parts except the projector head are purchasable new.

Currently the best approach is to match the sensor response curves to the LED emmission curves. This results in very specific LED selection criteria, to make sure you do not have light coming in that the sensor will miss.

As an aside - the comeback in vinyl is mostly due to mastering [and to hipsters 😉 ], the vinyl masters tend not to have the ‘loudness’ applied that so often afflicts music releases. (google “loudness wars” for anyone interested.) This results in the vinyl releases often having an artificially wider dynamic range than the ‘hotter’ CD release of the same album. This results in the vinyl sounding much better, but CD sounds better if taken from the same master, and the (often-digital) master tapes that the vinyl and/or CD is produced from, sounds better than both.

When both the CD and the Vinyl have the same master used, the CDs dynamic range is considerably wider, and closer to the source material than the vinyl record. People claim to be able to hear all sorts of things, but when we did some blind testing a few years back, we artificially put a ‘crackle-pop’ at the start of the digital recordings, and people waxed quite lyrically about how much warmer and true the vinyl recording sounded, except of course, that it was the CD track. We put the crackle in at the start to see if people perceived it differently if they thought they ‘knew’ it was on a record. Hearing is so easily manipulated by expectation, (this is fun for instance https://youtu.be/G-lN8vWm3m0?t=21s) and we have never had anyone consistently able to hear a difference between an analogue master and a properly digitised copy of it.

That argument could play out forever, people love their records and I don’t want to spend my small amount of time here stirring up that hornet’s nest, so to get back on track, if you choose the R, G, B LEDs correctly (sometimes this will require a mix of two different spectrum RED LEDs) then you will recover everything the film has to offer. Most prints are pretty lo-fi, and you will get more detail scanning them than could ever have been perceived projecting them.

I am lucky enough to have good friends at RMIT that did some spectral analysis for me of various light sources, projected through film test images, and the separate RGB light source conveyed the widest spectral range of any of the light sources tested, that could be captured by commercially available sensors. Unfortunately the ‘white’ LEDs with high CRI ratings fare poorly. Partially because the CRI measurement system is badly flawed, resulting in poor reproduciton in skin-tones specifically. Hopefully the industry will switch to CQS which gives a much more true measure.

Apart from Yard sales/garage sales, and film collectors, there aren’t a lot of options.
Projectors are expensive, so you are better off buying one from someone who services them and get one that is in tip top condition, that has new belts and has been properly serviced and lubricated etc. otherwise you will end up with nothing but trouble.

There are so many projectors out there it is hard to recommend particular models. You want one that has easy access to remove the gate and pressure plate, as you probably want to file those out a little to expose more of the film image area, and also ease of removing the shutter, as you will also want to remove it. Something that is known to be reliable and gentle on film is also a good idea.
The Bauer projectors are good for S8 and 16mm, the Hokushin 16mm projectors typically have an open film path and are easy to work on.
The main thing is to get one in great condition, or one that has been serviced with new belts and rubber parts replaced. It will cost hundreds, but the rest of the gear to catpure in hidef in realtime is going to set you back around $1000-$2000 or so in total, so it isn’t the expensive part of the build.

MP 30 (http://www.film-tech.com/warehouse/pics/arisia09/arisia0907.jpg) or Mignon 35 portable https://www.youtube.com/watch?v=evnc3d8uxGE take up the least space and are easy to modify