First scans

[Paul Bruening]

Hey Phil,

 

I always appreciate your help and observations.

 

How would I determine veiling? I assume it will demonstrate as a whiter or near white light adding to the image in a sort-of fogging. When I say fogging, I mean fogging-like but only impacting light values and not focus or clarity. What tests should I use to compare? Would a print out of the same image help? I'm thinking that a printout is reflected light and the negative image is backlit (light projecting within the direct path of lens light flow).

 

The thing is, how do I get around a lens? Optical printers have lenses. Even still film scanners have lenses. All these devices have backlight as well. I have seen optical printers advertised on Ebay that had fairly ordinary camera lenses as part of their image path. Do they use a trick that gets around flairing?

 

Will,

 

It's not so well covered. It's just getting close enough that it might actually work. I designed the system only to meet my own production needs. I had a heart attack eight months ago and lost my remaining production capital to the medical system. (3 days in the hospital cost $48,000.00 not counting doctors, helicopter ride and all the outpatient follow-up stuff.) My rig only does 2-perf telecine and scans. So, there's very little business available even if I had any money to promote it.

 

I'm a sad and pitiful sap who blew his money on a half-way production.

[Guest will griffith]

I'm a sad and pitiful sap who blew his money on a half-way production.

But at least you are alive to realize you did!

 

I did not mean to make fun of your situation, just

trying to lighten things up a little.

 

Good luck.

[Paul Bruening]

I didn't take it that way at all, Will. I'm enjoying your posts. I've got lotsa' half-baked ideas that might interest you. Ask away.

[Hans Engstrom]

3 days in the hospital cost $48,000.00 not counting doctors, helicopter ride and all the outpatient follow-up stuff

 

OT

 

Ouch expensive. Glad you´re ok now, thats the most important thing. My dad had a heart attack last year and had to stay for a week at the hospital but in Sweden that costs $90. Guess he paid for it with taxes for 40 years or so but I like it better that way. I still consider myself young at 33 and don´t need any expensive medical care at the moment but I don´t mind paying a little extra taxes for those who are in need.

[Mike Nichols]

My rig only does 2-perf telecine and scans. So, there's very little business available even if I had any money to promote it.

 

 

I just shot a 2 perf feature... :)

 

Guinea Pig???????

[Paul Bruening]

You know what they say, Mike, "Never approach a starving dog with food in your hand."

 

I probably won't be able to do anything on spec. But, you won't believe just how cheap I'm willing to negotiate down to. I've got all these computer work stations as well: 1 capture, 3 image, 3 sound and an administrator station. Also, there's an abundance of apartments and condos for darn cheap rent here (3BR/3BTH/LR/K for $900/M) because developers over-built and the market shrank. Flights out of NY and LA to Memphis are still pretty cheap ($400-600 RT). I'll go pick-up and drop-off your peeps. I've got a pro kitchen that can feed a dozen people 3 squares a day for cost. Plus, the swimming pool is lovely, the chics at the University are frickin' hot and the plentiful, local bars are rather cool music spots.

 

Give it a 'tink, as they say.

[Mike Nichols]

You know what they say, Mike, "Never approach a starving dog with food in your hand."

 

I probably won't be able to do anything on spec. But, you won't believe just how cheap I'm willing to negotiate down to. I've got all these computer work stations as well: 1 capture, 3 image, 3 sound and an administrator station. Also, there's an abundance of apartments and condos for darn cheap rent here (3BR/3BTH/LR/K for $900/M) because developers over-built and the market shrank. Flights out of NY and LA to Memphis are still pretty cheap ($400-600 RT). I'll go pick-up and drop-off your peeps. I've got a pro kitchen that can feed a dozen people 3 squares a day for cost. Plus, the swimming pool is lovely, the chics at the University are frickin' hot and the plentiful, local bars are rather cool music spots.

 

Give it a 'tink, as they say.

 

Still cutting the offline, but will definitely give it a 'tink!

[Phil Rhodes]

I suppose the thing to do would be to grab a piece of black... something, I don't know, aluminium foil or something and cut a slot in it, then scan it. You know the hole is absolutely transparent, you know the black is absolutely opaque, so any light pollution of the black areas will be due to your optical system.

 

Then you can experiment with exposure, aperture and shutter interval until you achieve the best performance possible. You may want a piece of unexposed, developed neg around to put behind your known-opaque mask, in order to restrict the amount of light coming through the hole to a sensible d-min.

 

If I were you, I'd build an LED backlight for it and have the camera modified to disable its shutter. You can pulse the LEDs to achieve the correct exposure and they won't wear out. Use RGB LED clusters and you can alter the colour balance, as well as avoiding the ageing problems and afterglow of white ones.

 

P

[Paul Bruening]

Hey Phil,

 

I like the pinhole aperture to check veiling. I'll give it a try.

 

I originally was going with the LED back light. Bruce had even built it. It solved a whole bunch of engineering problems in one swath. Then, I read, here, in the forum that LEDs have spiked colors. I looked them up and indeed they aren't full or balanced spectrum. Sure, they're good enough to fool the eye just like flos do. But for scanning, I couldn't let it go at that. I wanted more assurance of the best range of colors from the light source. After all this time and all the technology has passed, tungsten still has the best color curve. Since regular tungsten wasn't bright enough, we went with a halogen bulb. It is pretty close to a tungsten curve. The gas doesn't have a significant impact on the color curve. It meant bouncing the light around and, boy, did that irk Bruce. He went through three designs. The spaces are so small in there that a suitable approach took a lot of trial and error. I have to say, Bruce was a good sport about it and kept at it when most other people would have said, "F*** it!"

[Phil Rhodes]

There are certainly scanners out there which use LED backlighting. I'd be tempted to try it and see how good I could get it to be.

 

P

[Hal Smith]

The core problem with LED's is the light spectrum of each color LED doesn't match the CMY spectrum passbands of negative film nor the RGB spectrum passbands of reversal film. You need a true black body radiator like a halogen lamp to avoid those spectrum mismatch problems.

 

Everything You Ever Wanted to Know About LED's

 

Lamps like HMI's and MSR's aren't too hot (;)) either. Photos shot with film or even digital cameras of scenes using a lot of MSR or HMI lighting always look odd to me, not exactly cold as in color temperature, but somehow unnatural. Apparently the blue and green spikes in those bulbs don't match film or sensor passbands very well either. I'd hate to be one of the timers who have to fix the colors in movies shot with HMI's.

 

 

(MSR1200HR spectrum)

[Topher Ryan]

All this talk about the trouble of back lighting one of these rigs, with the limited space and all, I've had the bass-ackwards (or "round front") idea of lighting through the lens port (or modified lens). BUT this would require a mirror behind the gate BETWEEN the film and the sensor. I don't know much about the quality of mirrors available for photographic applications, but does this completely defeat the purpose of "scanning", to add a mirror between the film and sensor?

 

Does anyone with knowledge of the big-boy scanning units know if mirrors are ever allowed to "break" the path between film and sensor?

 

How much room is there in the Mitchell? The 16mm Bolex I'm working with might not even have enough room back there for a mirror to "see" full-frame anyway.

 

I can see why some people use optical printers and projectors for the film-transport, but I think it's way COOLER to use a camera!

 

(Since I would be looking in the base side, here's Mr. Tobin's take sharp side and soft side: http://www.tobincinemasystems.com/index_files/Page676.htm )

[Phil Rhodes]

Does anyone with knowledge of the big-boy scanning units know if mirrors are ever allowed to "break" the path between film and sensor?

 

I wouldn't.

 

As to backlighting, you could use tungsten, LEDs, compact fluorescents or cold cathode lamps - even an electroluminescent sheet, I guess. The point about the spectral output of red, green and blue LEDs is valid, though I have seen a custom built scanner that definitely used them. You could use white LEDs, which have an at-least-slightly more continuous spectrum, but you might, over time, have issues with decay of the phosphor emitter compound. This might also be an issue with EL sheet or cold cathode.

 

P

[Karel Bata]

I'd agree with the comments about how LEDs display poor spectral characteristics. It put me right off using them for something else. Here's a page on what 'white LEDs' are available: http://ledmuseum.home.att.net/ledwht.htm I don't understand how film lighting manufacturers marketing LED softlights get around this. Or maybe they don't...?

 

BTW you'd also need a regulated power supply and a good heat sink - LEDs change color with temperature. :(

[Alex Lindblom]

The original Spirit's uses a Perkin Elmer 300w xenon lamp, they are easy to get hold of but I guess they are to big for your rig.

 

The need for a reasonably priced film scanner is huge, so keep up the good work.

[georg lamshöft]

The ARRISCAN uses a custom-designed LED from Osram(.de) which seems to work very well, maybe there is a way to get this one (it matches the size of 35mm perfectly)?

[Karel Bata]

Paul, you might find this article interesting, particularly regarding your blue pixels,: http://www.stopmotionworks.com/articles/cbrdstrpdbare.htm

"We originally selected the Nikon D2H ... However, random noise was visible as pixilation in dark areas when the shots were played back as a movie. This pixilation effect was only visible in stop-motion photography, an application the Nikon hadn’t been designed for."

Karl mentioned that camera shutters have a life expectancy of 100,000 uses. Is that a relevant issue here?

 

Did you consider using the still capture mode of a video camera..? Or even grabbing frames from the live HD feed?

 

As I said in the other thread, recording the images might be made a lot easier by using dedicated stop-motion software like Stop Motion Pro

Edited April 4, 2009 by Karel Bata

[Paul Bruening]

Hey Karel,

 

If I can ever climb out of my debt-hole, I'll probably go to a Canon D5MII and get HD, bypassing the shutter for most projects. The weak point of the DSLR system is the shutter's wear and tear. If you go with this system, you'll probably want to keep two bodies on hand so that you can rotate them when the shutter blows out. Canon will fix, Nikon too, their shutters for around $200. So, let's figure you bought two Canon XSi's for $800 each. Maybe the first shutter will last for a 90 minute feature's worth of frames (130,000). That's $0.00615 (a little over a half penny) per scan. When you get it fixed, assuming it holds up another feature's worth of scans, the cost goes down to $0.0015 (roughly one-seventh of a penny) per scan.

 

At this kind of savings, the inconvenience of repair rotations certainly seems more tolerable.

 

While a projector already has the light management system inherent to its design, I preferred using a Mitchell's works because of the dependability, ease of conversion (it's a well known design), rock-steady, pin registration and friendliness to the delicate surfaces of film. But, without a doubt, getting a good quality of light into it with an even-field distribution has been quite a challenge.

[Chris Burke]

Hey Karel,

 

If I can ever climb out of my debt-hole, I'll probably go to a Canon D5MII and get HD, bypassing the shutter for most projects. The weak point of the DSLR system is the shutter's wear and tear. If you go with this system, you'll probably want to keep two bodies on hand so that you can rotate them when the shutter blows out. Canon will fix, Nikon too, their shutters for around $200. So, let's figure you bought two Canon XSi's for $800 each. Maybe the first shutter will last for a 90 minute feature's worth of frames (130,000). That's $0.00615 (a little over a half penny) per scan. When you get it fixed, assuming it holds up another feature's worth of scans, the cost goes down to $0.0015 (roughly one-seventh of a penny) per scan.

 

At this kind of savings, the inconvenience of repair rotations certainly seems more tolerable.

 

While a projector already has the light management system inherent to its design, I preferred using a Mitchell's works because of the dependability, ease of conversion (it's a well known design), rock-steady, pin registration and friendliness to the delicate surfaces of film. But, without a doubt, getting a good quality of light into it with an even-field distribution has been quite a challenge.

 

 

Isn't the video that comes off of the Canon 5D in the H.264 codec? I may be wrong, but I think it is and if so, isn't that a little on the compressed side for most users?

[Karel Bata]

I don't like the sound of anything mechanical breaking down! Which is why I wonder about using a flash - less wear on everything.

 

Did you explore the possibility of using a video camera? They don't have a shutter, and if they can focus on a DoF adapter GG then maybe they can manage a 35mm motion picture frame? Or even Super 16? Or do DoF adapters create substantially bigger images on their ground glasses? Though I'd bet you'd have to use a dedicated macro lens, which limits choice of camera, but certainly less chance of something breaking down. I'll do a test my daughter's HV20 soon (I can borrow it!). Funny though - I've got cans full of 35mm film knocking around here, but no individual frames to experiment with! :huh:

[Paul Bruening]

Isn't the video that comes off of the Canon 5D in the H.264 codec? I may be wrong, but I think it is and if so, isn't that a little on the compressed side for most users?

 

Sure. There's no perfect solution short of a bottomless bank account. My current system is an XL2 with EF adapter and SLR, macro lens to get workprint grade SD video. Then, I use a Kodak DCS ProC with macro for individual 4.5K wide scans (bayered, of course).

 

With a 5DMII I can get satisfactory HD (many people say that H.264 is fine for any video output) for small projects headed for video display. But, I can also get 5K scans in RAW for projects intended for big screen or prestigious resolution outputs. It's all theoretical at this point since I'm dog broke.

[Karel Bata]

Sure. There's no perfect solution short of a bottomless bank account. My current system is an XL2 with EF adapter and SLR, macro lens to get workprint grade SD video. Then, I use a Kodak DCS ProC with macro for individual 4.5K wide scans (bayered, of course).

 

With a 5DMII I can get satisfactory HD (many people say that H.264 is fine for any video output) for small projects headed for video display. But, I can also get 5K scans in RAW for projects intended for big screen or prestigious resolution outputs. It's all theoretical at this point since I'm dog broke.

 

Paul, what do you do about dust? And how many images/second (or seconds/image!) do you capture?

 

Any chance you can post a pic? I'm intrigued how you've arranged all this mechanically.

[Ben Syverson]

Some notes on this old thread...

 

The reason you're getting a lot of noise in the blue channel is not grain, but the orange mask... Are you using a ~3200°K hot light? If so that will exacerbate things. The DSLR sensor has a native white balance, so by forcing the WB on the camera to 2500°K or relying on Photoshop to remove the mask, you're basically taking a very weak blue channel and pushing it to oblivion.

 

Ideally you want to be shooting close to the native WB of the camera (daylight, more or less), using a calibrated blue backlight to neutralize the orange mask. That means your light should be substantially bluer than daylight. When you open the uncorrected camera files in Photoshop, you want to see three similar-looking channels that mainly need to be inverted. The less image processing you have to do, either in-camera or in PS, the less noise.

 

If you're looking for a good programmable RGB light source, you may only need to look as far as your cell phone. LED backlights can be incredibly even and consistent. Not to plug, but a while ago I developed an iPhone app named "Catchlight" that provides a fast interface for adjusting the color temperature of light emitted from the screen. Recently I opened it up and threw a 35mm neg on the screen. After setting the camera to 5600°K, I adjusted the light color until the orange mask was photographing as white. Once I imported that file, all I needed to do was invert and apply a minor Levels adjustment to introduce some contrast. I wound up with less noise than I get from my dedicated Nikon film scanner.

 

I'm going to try to get a low-volume 35mm scanning workflow set up at home, and I really think getting the right light color is key...

[Paul Bruening]

Some notes on this old thread...

 

The reason you're getting a lot of noise in the blue channel is not grain, but the orange mask... Are you using a ~3200°K hot light? If so that will exacerbate things. The DSLR sensor has a native white balance, so by forcing the WB on the camera to 2500°K or relying on Photoshop to remove the mask, you're basically taking a very weak blue channel and pushing it to oblivion.

 

Ideally you want to be shooting close to the native WB of the camera (daylight, more or less), using a calibrated blue backlight to neutralize the orange mask. That means your light should be substantially bluer than daylight. When you open the uncorrected camera files in Photoshop, you want to see three similar-looking channels that mainly need to be inverted. The less image processing you have to do, either in-camera or in PS, the less noise.

 

If you're looking for a good programmable RGB light source, you may only need to look as far as your cell phone. LED backlights can be incredibly even and consistent. Not to plug, but a while ago I developed an iPhone app named "Catchlight" that provides a fast interface for adjusting the color temperature of light emitted from the screen. Recently I opened it up and threw a 35mm neg on the screen. After setting the camera to 5600°K, I adjusted the light color until the orange mask was photographing as white. Once I imported that file, all I needed to do was invert and apply a minor Levels adjustment to introduce some contrast. I wound up with less noise than I get from my dedicated Nikon film scanner.

 

I'm going to try to get a low-volume 35mm scanning workflow set up at home, and I really think getting the right light color is key...

 

Thank you for your post, Ben. Your idea is interesting to me. I have looked at some iPhones on Ebay for lowest prices. I am thinking of how I could put that big screen behind my pressure plate. I can break the unit up to reduce mass. I'm not sure how it can sit behind my film area and still be accessible for controlling. I'll give it some more thought. Meanwhile, what kind of light does an iPhone screen produce. I am interested in CRI and spectrums.

 

In answer to one of your questions I use gelled, halogen light to manage color.

[Phil Rhodes]

I think I'd look at some LEDs before I went to those lengths.

 

P