DIY Dual 8 Telecine HD Digitizer under $1000?

[Thomaseo1]

As the title says, I'm looking at building a Dual 8 Telecine HD Digitizer for under $1000.

Back Story (not really relevant)
About a month ago a family friend was having a family reunion and the topic came up that they had about 210 50FT reels of Super 8 and Regular 8 film that needed to be converted. They have a projector and I have a HD camcorder and I was just going to "film" the projected image. This turned out terrible due to being out of focus, frame rate problems as well as exposure issues. I found another method that looked promising involving using a Transparent Materials Adapter enabled scanner (HP G4050) and scan each film in by hand at 4800 DPI. This resulted in the film being a bit over exposed and goes in an out of focus and frame alignments being incorrect and when corrected made frames jump about and worst off will take about a month of work to complete. This has led me to looking around at several how to sites on DIY digitizers and found that if done properly the results can look amazing.

Actual question (relevant)
With a Budget of $1000 is it possible to do the following?
Using a Dual 8 Projector replace the lamp with a programmable USB controlled RGB "white" LED array with 100-200W lamp equivalence to expose and capture each frame to 3 different Gain Levels (lets say 50% 100% and 150% of normal) in order to make a composite HDR image of each frame. LED controller would need to change levels at about 24 FPS and camera capture each level while the projector only plays at 1/3 rate = 8fps.

Here are a few of the items I believe would be required.
Projector- Variable speed and/or Ultras slow motion of 8 FPS Dual 8 Projector that historically is gentle and won't chew on film. (Ebay has tons of these)
Motor - It may be necessary to use a controlled stepping motor in order to control frame timing (I'm not sure what to look for or if there is a good projector that would already be suitable)
LED Lamp - Using combined RGB LED array to give a better white than the fake white LEDS at 100W-200W equivalence or about 1600 lumen - 3200 lumen
Controller - USB controller to control LED Gain levels and timings (can't find anything like this except cine2digits which make an obscure reference to it)
Power Regulator - Possibly additional power would be needed if the 2.5W from the USB port isn't enough to supply the controlled LED lights
Machine Vision Camera - 720P or higher (2MP preferred), ~25+ FPS, Global Shutter, controllable frame rates for timing purposes. (I think CCD is better but can a CMOS with global shutter work too or is it too grainy?)
Marco lens - Macro lens with c or cs mount than can fit inside of the original projector lens assembly
Mount - Adjustable camera mount to hold the machine vision camera in place with Projector

If you know any good suggestions on what to use and where I might find any of the more exotic items (Controllable LED array, inexpensive HD CCD Machine vision cameras) please let me know


Here are a few sites I'm getting info and Ideas from
http://www.super-8.be/s8_Eindex.htm
http://www.rickmears.us/Main/S8/s8.htm
http://super8data.com/
http://telecine.godiskind.com/
http://www.cine2digits.co.uk/

[Thomaseo1]

So far my efforts are focused on purchasing a Machine Vision Camera. Currently I'm looking at spending up 1/2 of my budget ($500) on a camera. I am specifically looking for one that has a Global Shutter (Is this a mistake or is there a cheaper solution?), however I'm not sure if a CMOS will work too or if I should only look at CCD. There are several CMOS image sensors out there that state they have a global shutter and I think these ill work too I hope.

[milesandjules]

We built many a transfer machine for cheap using a modified projectors. We used one for years …using a reed switch and magnet to trigger a mouse click for frame by frame capture into a pc using premiere 6.5 stop motion capture :lol:

But something always seem to stuff up. There would be missing frames or double frames or hair in the gate...or out of focus….or you would come back and there would be a pile of film on the floor….or the film would jam and only have transferred half the reel….or the film would start jittering as soon as you walked out of the room…. or the camera would get bumped..or the mouse would get bumped and be clicking on your desktop or other programmes.

Spent years perfecting and fiddleing buying projectors etc….it was fun, but in the end we just bought a retro 8..they just work …the software is designed for transferring film and you can adjust gamma and brightness while its transferring … and now we can spend our time building other stuff.

Good luck to you though, it can be done and results can look amazing

[granfer]

I concur with milesandjules! The straight answer to your specific basic question is ... I don't think so.
However, even if you manage to find all your basic components to your perceived specifications which, when put together, should produce your required result for under your $1000, don't be surprised when it doesn't!!

That's what happens to most experimenters (and we can never figure out why!!) and is the starting point for much frustrating EXTRA tweaking, loss of sleep and expense.

Consider what you wish to achieve. Good quality (Hollywood standard?) transfer of about 10 hours of film. ONLY 10 hours. My prediction, based on experience, is at least twice your budget, somewhere (starting from scratch) between 100 and 300 hours of effort and lots and lots and lots and lots of frustration.

Sorry if I haven't put it as politely as milesandjules! But they are right. My advice is to either send the reels for professional transfer or buy a machine that will do the job reliably (such as Retro-8) that someone has already done the donkey work on. That way you can then use it to recoup your outlay .

Best regards,
granfer (severely overspent experimenter)

[Pj]

I first built my Super 8 telecine adapting a projector, then built one with a self made simpler film transport system and now I’ve just finished building a film scanner to handle my Super 8 and 16mm films and still negatives and slides. I have spent many hours looking for options for the back light, capture cameras and have explored various ways to advance the film.

Firstly, in my experience a CCD sensor with a global shutter is better than CMOS, machine vision cameras are better but can be pricey. I have looked at all sorts of consumer digital cameras and they all present various issues, they’re too slow, it’s difficult getting decent optics with manual controls on them and unlike machine vision cameras most aren’t always able to be controlled via the computer. Personally I feel that projectors are not ideal to transport film at slower speeds, they are rough and need constant tweaking and fiddling about with which is why I created my own simple film transport.

http://www.lightbreeze.co.uk/thescanner.htm


Pav

[carllooper]

If your film frame remains stationary during capture it doesn't matter whether you are using a global shutter or a rolling shutter. The difference in shutter only matters if your subject matter is moving.

So for a stationary frame capture, a CMOS machine vision camera is a totally acceptable solution.

It's also far cheaper. And it's cheaper because CMOS is cheaper to manufacture.

Some have argued that CCD is better because the image area of CCD is larger. I'm not quite sure how to follow this logic. All else being equal (same number of sensor pixels, same lens) a larger image area, regardless of format, will tend to give a better result. The only issue will be the availability of either in the desired size.

C

Note.

Where your film is in continuous movement across the sensor there is the problem of motion blur whichever shutter you use. For a global shutter sensor you would need to ensure a short exposure time (such as a flash) in order to avoid motion blur. With a rolling shutter sensor will be required a different solution, but a little more complicated.

First, here is a good article on the difference between rolling shutter and global shutter:
http://www.ptgrey.com/support/kb/index.asp?a=4&q=115

In terms of rolling shutter the relevant paragraph is this one:

The rolling shutter in a CMOS image sensor works differently, in that the photodiodes (pixels) do not collect light at the same time. All pixels in one row of the imager collect light during exactly the same period of time, but the time light collection starts and ends is slightly different for each row. The top row of the imager is the first one to start collecting the light and is the first one to finish collecting. The start and end of the light collection for each following row is slightly delayed. The total light collection time for each row is exactly the same, and the delay between rows is constant.

The 'motion blur' that this can cause is not traditional motion blur. The movement is not rendered as softness in a certain direction. It remains sharp (indeed it should be sharper than that obtained with a global shutter) but it's distorted since each line of the scan is acquired at a different time:



So we want to keep this sharpness but remove the distortion. The solution here is to undistort the image. And the way an algorithm can do that is to work with all the capture frames acquired as the film moves across the sensor, rather than just one capture frame per film frame. The algorithm would rearrange and reintegrate each individual line from each and every frame, back into the correct spatio-temporal order. The algorithm would be no more complicated than that which a line scanner would use, other than you would be coding for a large number of them - and the benefit of which would be a far better integral result.

[Thomaseo1]

@ vmilesandjules Thanks for your feedback. I have already looked at the Retro 8 and unfortunately the biggest problems with it is it's $2000 price tag and it's standard definition specs. My limited $1000 maximum budget requires me to get very creative and get everything mostly right or passable the first time.

For the moment I'm trying to find any stats on gate timings for projectors and I'm unsure what percentage of time the film is moving through the gate vs stationary. I've made guesses ranging from 1/2=50% (over estimate) in motion time, down to 1/8=12.5% in motion time, this would tell me I need to have a camera shutter controllable with enough accuracy to avoid capturing film while its in transport. Since I want my goal to be capturing 8 actual film frames a second @ 3 lighting levels per frame, I would need a camera capturing at least (3*8=) 24 fps, but with a gate potentially being half the time taken the frame rate would need to be doubled to a frame rate at 24*2=48 fps. 1/2 to 1/3 of the captured frames per frame would not be usable and would be be dropped.
Here are a couple sequences that could be used

Projector Rate = 8 fps
Global Shutter Camera Capture Rate = 48 fps
C=capture, X=drop
Levels = 1=150% above normal Gain (overexposed), 2=100% of normal Gain (correct exposure), 3=50% below normal Gain (under exposure)
|Frame_1____|Frame_2_____|Frame_3____|Frame_4____|Frame_5____|Frame_6____|Frame_7____|Frame_8____| = 8 Frames
|X|X|C|C|X|C|X|X|C|C|X|C|X|X|C|C|X|C|X|X|C|C|X|C|X|X|C|C|X|C|X|X|C|C|X|C|X|X|C|C|X|C|X|X|C|C|X|C| = 48 Frames - (24 dropped, 24 captured)

|X|X|C|C|X|C|X|X|C|C|X|C|X|X|C|C|X|C|X|X|C|C|X|C|X|X|C|C|X|C|X|X|C|C|X|C|X|X|C|C|X|C|
|3|1|1|2|2|3|3|1|1|2|2|3|3|1|1|2|2|3|3|1|1|2|2|3|3|1|1|2|2|3|3|1|1|2|2|3|3|1|1|2|2|3|3|1|1|2|2|3 = Sharp drop off on luminescence change but is easier to program to composite images to HDR
OR
|X|X|C|C|X|C|X|X|C|C|X|C|X|X|C|C|X|C|X|X|C|C|X|C|X|X|C|C|X|C|X|X|C|C|X|C|X|X|C|C|X|C|
|1|1|2|2|3|3|3|3|2|2|1|1|1|1|2|2|3|3|3|3|2|2|1|1|1|1|2|2|3|3|3|3|2|2|1|1|1|1|2|2|3|3|3|3|2|2|1|1| = Smother luminescence change and helps prevent light retention on Camera image senor from previous pass however it may be harder or just the same to program to composite images to HDR

|Gate|______|Gate|_______|Gate|______|Gate|______|Gate|______|Gate|______|Gate|______|Gate|______|

The best priced camera I can find out there that has global shutter, at least 1280 X 720 resolution and at least 48 FPS and is less than $500, is this one.
http://www.ptgreystore.com/blackfly-12- ... ina-ar0134 $315
I am slightly wary of Gig-E (potentially dropping frames) and the fact that the sensor is a CMOS which typically have been known to give a bit more grain on low light images.

[MovieStuff]

I have already looked at the Retro 8 and unfortunately the biggest problems with it is it's $2000 price tag and it's standard definition specs.

Hi, Thomas!

The Retro-8 does not capture in standard definition.

Roger

[carllooper]

The film is held stationary in the projector for 5/6ths of a frame interval, and moved during 1/6th of that same interval.

For example, if the projector is running at 24 frames per second, then the frame interval is 1/24th second. So in this example, the film would be stationary for 5/6ths of 1/24th of a second = 0.034 secs, and would be moved for 1/6th of 1/24th sec = 0.006944 sec.

However projector speeds are not consistent. Typically you will want a sensor of some sort on the projector in order for your system to know when the film is stationary and when it is moving. For example, a Hall Effect sensor on the shutter shaft has proved popular. For film transfers you also want to remove the blades on the projector as they will needlessly interfere with acquistion.

With a 48fps camera on an 8 fps projector, the camera will capture 6 digital frames for every one film frame. If camera/projector remain syncronised, five of those images will be of the same stationary film frame and one will be moving. Unsyncronised, up to two digital frames might see the film moving, with only four frames seeing a stationary frame. Without any syncronisation strategy you will have the problem of identifying which frames are good frames and which are not. In other words you need some sort of sync strategy anyway.

If the stories I hear are correct, the first analog to digital interface for a film to video transfer system involved Roger Evans hacking a computer mouse (!) as a way to notify a given computer video capture system, what frame to extract from a digital video stream of the film. Today we're somewhat more blessed with a range of cheap programmable Ultra high definition machine vision cameras and ready-made analog/digital usb interface boards with which to work. The hard part still remains the film transport system. We either build our own transport system, hack an existing transport system (a film projector) or buy an already integrated system (such as those by Evans) but which may not meet our particular needs in terms of camera.

In terms of cost I can safely say I've spent way more money than the price of a Retro8, building various versions of a film to digital transfer system. Way more. However I've also learned way more than I ever would have otherwise. And I can claim greater ownership over the outcome - which can be very important from a certain artistic point of view. The blow out in costs is inevitably due to a lack of knowledge. One starts out by making certain assumptions (theories) about how certain things work (but what else can you do?) and then after assembling a system one discovers an error in some of the assumptions. Or one discovers the result could be improved by doing it another way (another theory). Back to the drawing board and the purchase of more components.

One really needs to know what one is aiming at. If it's to manufacture a competing product to existing integrated systems (such as the Retro8) that's a whole different ball game. But if the agenda is the creation of a one-off system of your own, which you can fine tune to your own particular needs, that's an entirely different story. Whether that can be done cheaper than a Retro8 or otherwise it is somewhat beside the point. You pay what you are willing to pay to achieve want you are wanting to achieve.

For example, the work of Mattius Norberg could not be done on a Retro8. To do what he does he needed to build his own system. Or building his own system allowed him to do what he does:

http://vimeo.com/85767429

C

[woods01]

There are a lot of great DIY telecines out there. Some are probably better than many of the commercial units as far as quality goes. But unless you are tinkerer with a nice shop and loads of time I'd say just buy one of the commercial units, transfer your film and then sell the unit on eBay after you're done. Better yet, get the family friend to either buy the unit and do all the work since its their film, or get them to go 50/50 into the purchase.

[granfer]

Thanks, woods01, for saying the same as me in slightly different words. It's a "no-brainer", isn't it?
Sadly, however, once we get an idea, we humans press on regardless ............

[carllooper]

Buildng your own film to digital transfer system is no different from making your own film. Or cooking your own cake. Or making your own chair. Or building your own house. Or your own darkroom. Or sewing your own clothes. Or doing anything else you care to do. Everything you do takes time. Watching TV takes time. Watching a movie takes time. Walking to the bus stop takes time. Sitting on the beach takes time. Drinking coffee takes time. Reading and writing on filmshooting.com takes time. Who is to say anything that takes time is any more, or any less important, than anything else one might do with one's time?

To suggest doing something yourself is somehow sad is, itself, very very sad.

C

[granfer]

Quote: "Everything you do takes time. Watching TV takes time. Watching a movie takes time. Walking to the bus stop takes time. Sitting on the beach takes time. Drinking coffee takes time. Reading and writing on filmshooting.com takes time. Who is to say anything that takes time is any more, or any less important, than anything else one might do with one's time?"

I'm afraid this contributor seems to have lost the whole point of the thread (but then he often does). It's about COST not TIME! But then cost apparently is not of any consequence to him .....

Quote: "Whether that can be done cheaper than a Retro8 or otherwise it is somewhat beside the point."

The whole point of the topic is that COST is not "beside the point". He freely admits that his own machines have cost many times the Author's Budget, but offers no opinion as to whether someone with little or no previous experience can REALLY expect to achieve his aim within his very limited budget.
To his credit, he does offer some salient facts regarding the obvious barren areas in the author's knowledge. And to those of us who are foolish enough to try and answer the actual question, he offers this ...

Quote: "To suggest (that) doing something yourself is somehow sad is, itself, very, very sad"

Now I'm not sure whether that is supposed to be patronising, needling, taking the mickey or simply another example of the pseudo philosophical/arty outpourings we are used to ..... but rest assured, carlooper, that it affects me not. I am old enough and experienced enough to have met it many times before. Just, if you can, try being CONSTRUCTIVE in your answers to those seeking help.

Before you accuse me of being DESTRUCTIVE in my advice rather than CONSTRUCTIVE, just try reading what I, and at least two others, actually said ...
that we thought the answer to his question was NO. What else we wrote was aimed at indicating how he might achieve the aim of transferring the films WITHIN his budget. Perhaps you would like to offer to do just that for him?

Even Roger, brilliant engineer and innovator that he is, would likely say that there was no way that he could have built his first fully working Workprinter, from scratch, for the equivalent then of today's $1000 even with its understandably lower specification. I'm also sure that the development cost of his Retro 8 was many times that $1000, even with the benefit of his vast experience and existing facilities. (I have NO connection or financial interest in Moviestuff, nor do I own any of their machines: I have used them only as .an example of one of many people offering Commercially engineered Transfer Machines).

My advice to Thomas remains the same: No I do not imagine for one moment that you will achieve your aim for less than $1000. Don't let anyone who is not prepared to actually give you a straight YES or NO answer to your question lead you to commit yourself to indeterminate expenditure if your limit is truly $1000.

[Pj]

I don't think it can't be done for under $1000, a decent set up will end up costing a lot more.

Pav

[MovieStuff]

Since my unit and myself have been mentioned several times as a point of reference, I thought I might weigh in on this discussion.

First off, to answer the question "Can a dual 8 digitizer be built for under $1000?"

The answer CAN be yes but is best answered with a variant on the same question,"Can YOU build a dual 8 digitizer for under $1000?"

This is not meant to be snarky but, rather, to address the larger issue of resources versus results. If we want to confine the money spent to just materials, then we can leave out time required to build it as well as the time it takes to use it. In that case, you could -IF YOU KNOW HOW- just use a digital still camera to photograph each frame of film, pulled into place by hand on a backlit light box with nothing more than cardboard strips along the edge as guides. You would then -IF YOU KNOW HOW- use a program like Photoshop or even a free graphics program to register each frame for a steady image. You would then -IF YOU KNOW HOW- import those images into an NLE program as a numbered image sequence to reconstitute your video file. And, finally, you would -IF YOU KNOW HOW- adjust gamma and color to make it look best.

The process I just described is as simple as it gets. If there is any part of that which you can not perform, right this instant, and don't have the resources to solve, then the answer to your question is less about money and more about capability. Again, this is not meant to be sarcastic. The "concept" of frame by frame scanning is easier to understand than to implement, which is often what leads people to scratch their heads and contemplate building their own machine. The fact that you have asked the question about cost on this forum would suggest that you are missing pieces of the puzzle on how to actually implement your plan, rather than how much money it would take, and are hoping that someone will fill in the blanks for you.

There's nothing wrong with that since that's the great thing about this forum. Most members are generally very helpful. But it's hard for anyone here to give you a viable yes or no answer to your question because the wild card in all of this is you and what your technical limitations are. You may totally understand the science of it but have no experience with shop tools. You may be a master craftsman but don't know the first thing about optics. You may be a first rate photographer but are mystified in front of a computer. The various disciplines required to make a scanner aren't rocket science but should not be casually disregarded as incidental.

So, perhaps the best way for us to help you is by telling us what you already know and what pieces of the puzzle are missing. Also, what are your capabilities in terms of shop skills, available tools and computer knowledge. Once we have a better profile of the person that will be building the scanner, the more likely we can collectively provide useful information. Otherwise we end up talking about theory and less about the nuts and bolts of getting it done.

Roger