Simon Lucas Posted September 19, 2015 Author Share Posted September 19, 2015 Here's an example by Carl: Link to comment Share on other sites More sharing options...
Josh Gladstone Posted September 19, 2015 Share Posted September 19, 2015 The frame lines appear soft, so I'm going to guess it's a scanning issue. What kind of scanner is it? Link to comment Share on other sites More sharing options...
Simon Lucas Posted September 20, 2015 Author Share Posted September 20, 2015 (edited) The frame lines appear soft, so I'm going to guess it's a scanning issue. What kind of scanner is it? Yes, I admit now, that they do on this scan. Odd, as the 35 mm scans I do are so crisp. It's scanned on an Epson 4490 flatbed. I place the emulsion flat on the glass and weight the strip with a small enlarger film plate. I don't know how to fix that. I tried all sorts of experiments with the scanner and found that emulsion on glass was sharper than any variations on film holder (when testing with 35mm). Would a more up to date scanner like the cost-effective v370 improve it? Of course, making final film footage I will get them scanned professionally, but I wanted to be sure that what I am shooting is the best result possible. I wonder whether that's the only issue with sharpness, though? Edited September 20, 2015 by Simon Lucas Link to comment Share on other sites More sharing options...
Simon Lucas Posted September 20, 2015 Author Share Posted September 20, 2015 Here's a crop of a 35mm neg scanned the same way. The sprockets are OK, are they not? Here's a Super 8 developer test strip with sprocket. Scanned the same way. The sprockets are indistinct. But am kidding myself that I can see grain/surface of the film? So, why would the sockets be out of focus on the Super 8 stock but Ok on the 35mm? The Super 8 stock is very flat on the glass. The 35mm does bow, but I taped the film holder down to flatten it. But - also even when I do not tape the 35mm down it seems to be pretty sharp. Link to comment Share on other sites More sharing options...
Heikki Repo Posted September 20, 2015 Share Posted September 20, 2015 Please post crops from both scans zoomed to 1:1. Epson flatbeds are usually rather blurry in small details (even V700 which I have) so I'm tempted to say that your film is okay but your scanner isn't. IMHO Epson flatbed isn't sharp enough even for scanning 35mm stills. For 4x5" and 120 it's good enough @ 3200 DPI. Link to comment Share on other sites More sharing options...
Simon Lucas Posted September 20, 2015 Author Share Posted September 20, 2015 (edited) These are both at 2400dpi. Image cropped. Here's the full uncropped file at 3.9Mb: http://www.studioshitless.com/tmp/super8/35-s8comparison2.jpg Edited September 20, 2015 by Simon Lucas Link to comment Share on other sites More sharing options...
Simon Lucas Posted September 20, 2015 Author Share Posted September 20, 2015 Please post crops from both scans zoomed to 1:1. Epson flatbeds are usually rather blurry in small details (even V700 which I have) so I'm tempted to say that your film is okay but your scanner isn't. IMHO Epson flatbed isn't sharp enough even for scanning 35mm stills. For 4x5" and 120 it's good enough @ 3200 DPI. You are saying I cannot rely on a flatbed for Super 8 tests. And I think this echos what someone else said, earlier. It seems like I have to work blind until I shoot a whole roll and have it professionally scanned. Are there any reliable ways of evaluating test footage? Link to comment Share on other sites More sharing options...
Simon Lucas Posted September 20, 2015 Author Share Posted September 20, 2015 Are there places that could do 2k scans of short (6") strips of films? Link to comment Share on other sites More sharing options...
Premium Member Tyler Purcell Posted September 20, 2015 Premium Member Share Posted September 20, 2015 I just see a scanning issue. Just because the flatbed scanner says it can scan at X resolution, doesn't mean it's imager and glass is capable of doing that. The only reason why your 35mm material looks acceptable is due to it's physical size compared to the smaller gauge. If you were to take that S8 material and drop it off at a lab for scanning, I bet you'd find it looks fine. Link to comment Share on other sites More sharing options...
Heikki Repo Posted September 21, 2015 Share Posted September 21, 2015 (edited) I suggest you shoot one cartridge of test material, process it and scan it. That's how testing is usually done. If that's not possible, you could try taking a photo of the film frame with a good DLSR and macro lens against diffused light. Edited September 21, 2015 by Heikki Repo Link to comment Share on other sites More sharing options...
Simon Lucas Posted September 21, 2015 Author Share Posted September 21, 2015 Thank-you, I'm going to have to shoot a whole roll and get it developed and scanned to see what's going on. I've been doing small strips to test developer/exposure. But it's hard to get a good idea about the results, without seeing it scanned, first. I may also try a dSLR on my lightbox. Thanks for everyone's input. Link to comment Share on other sites More sharing options...
Josh Gladstone Posted September 21, 2015 Share Posted September 21, 2015 Flatbeds are set to focus a very small distance just above the glass, which is why film holders are supposed to get you the best results. In practice, this isn't always the case because 35mm and 120 film is often not completely flat to begin with. So sometimes putting it flat on the glass, and then having it curl a little bit, puts it in the sweet spot for focusing. But theoretically, the film holders are supposed to elevate the film to the proper position. So now when you're scanning super 8 and keeping it flat to the glass, I'm guessing it's just slightly out of the focal plane. And since super 8 is such a smaller image area, when you blow it up, it becomes more noticeable. But yes, suffice it to say this is not going to be a good way to judge your picture sharpness. Do you have a projector? Link to comment Share on other sites More sharing options...
Simon Lucas Posted September 21, 2015 Author Share Posted September 21, 2015 (edited) Flatbeds are set to focus a very small distance just above the glass, which is why film holders are supposed to get you the best results. In practice, this isn't always the case because 35mm and 120 film is often not completely flat to begin with. So sometimes putting it flat on the glass, and then having it curl a little bit, puts it in the sweet spot for focusing. But theoretically, the film holders are supposed to elevate the film to the proper position. So now when you're scanning super 8 and keeping it flat to the glass, I'm guessing it's just slightly out of the focal plane. And since super 8 is such a smaller image area, when you blow it up, it becomes more noticeable. But yes, suffice it to say this is not going to be a good way to judge your picture sharpness. Do you have a projector? I've agonised over the scanner focus issue. I tried all combinations with neg holder and without, sloping the neg away from the glass to find the focus point. And I always wonder what it is that is 'focused'. Is there an actual lens? But whatever, I had to return to the very first naive scan of Tri-X 35mm I did, stuck on the glass - it seemed to clearly show the grain. And my recent Silvermax tests seemed to support that that was the sharpest, although I never feel like I can resolve the grain any more. One note is that our scanner says 4800 dpi but it is clearly interpolated and I suspect the true res. is 2400. But the Super-8 edge/sprocket does not seem to be as sharp as the 35mm sprocket. So, I have an enlarger but my lenses are for 35mm and for 120. And I would need a minox sized lens (20mm or smaller) to really blow up 8mm. Or I wonder if I could set the enlarger to project across the room and make a print? I do have a Super 8 projector on loan. I will still have to shoot a good length of film to see it for long enough. I have been shooting 7second strips. But I will be tempted to project the film when I finish it, before it is sent for scanning. One question - what kind of macro lens would I need on a dSLR (I have a d70, not full frame), to get a suitably large photo of a Super 8 frame? I guess larger than 1:1. And would it cost more than a £100 scanner? If it worked it would be a justifiable investment. Edited September 21, 2015 by Simon Lucas Link to comment Share on other sites More sharing options...
Simon Lucas Posted September 24, 2015 Author Share Posted September 24, 2015 I've just found a 28mm Componon lens for my enlarger. 30mm is recommended for printing 16mm in my enlarger manual, and I hope the 28mm will give me enough magnification for printing 8mm. Link to comment Share on other sites More sharing options...
Carl Looper Posted September 24, 2015 Share Posted September 24, 2015 You can use the enlarger to project the Super8 image directly onto the DSLR sensor. You can use any lens really. I've used a 50mm enlarger lens, and a 75mm camera lens for this. C Link to comment Share on other sites More sharing options...
Simon Lucas Posted September 24, 2015 Author Share Posted September 24, 2015 (edited) You can use the enlarger to project the Super8 image directly onto the DSLR sensor. You can use any lens really. I've used a 50mm enlarger lens, and a 75mm camera lens for this. C Carl. Nice idea! I've also just realised that this lens is well-regarded amongst numismatists. To use it as a 2x macro, I can mount it on my d70 and have bought the reverse mounting ring and a step-up. Photograph the frame on my lightbox. But, I have to say that there's something very satisfying about the idea of projecting a film-based image directly onto a dSLR sensor. Almost poetic. Care to speculate which will give best results? And what were your aims? Edited September 24, 2015 by Simon Lucas Link to comment Share on other sites More sharing options...
Josh Gladstone Posted September 24, 2015 Share Posted September 24, 2015 (edited) My super 8 and 16mm scanner uses an enlarger lens on a macro bellows rig. Enlarger lenses are good for macro photography because they give nice flat images. Edit: I'm using a Schneider Kreuznach Componon 80mm f/5.6 mounted the proper direction on the bellows, and I've also got a Nikkor 50mm. But also keep in mind I'm not using a full frame dslr, I'm using a machine vision camera, so my image area is much smaller. Edited September 24, 2015 by Josh Gladstone Link to comment Share on other sites More sharing options...
Simon Lucas Posted September 24, 2015 Author Share Posted September 24, 2015 My super 8 and 16mm scanner uses an enlarger lens on a macro bellows rig. Enlarger lenses are good for macro photography because they give nice flat images. Edit: I'm using a Schneider Kreuznach Componon 80mm f/5.6 mounted the proper direction on the bellows, and I've also got a Nikkor 50mm. But also keep in mind I'm not using a full frame dslr, I'm using a machine vision camera, so my image area is much smaller. Josh, are you digitising whole films this way? IF so, how do they compare with commercially scanned films? Are you projecting into the digital camera, the same as Carl? Another question I mean to ask Carl, is - how do you get the film projection perpendicular to the digital camera axis? I have Componon S in 50mm and 80mm, so I can experiment. Link to comment Share on other sites More sharing options...
Jose luis villar Posted September 25, 2015 Share Posted September 25, 2015 This is digitalized with a Panasonic GH2 and a reformed Bauer T502 projector to 16.66 fps. The lens of the gh2 is a macro Sigma 105 - 2.8, and I used the option to cut that has this camera to fill the frame s8. To remove the extra frames and converted to 25 fps, I have used the software P10JMC v8. Link to comment Share on other sites More sharing options...
Carl Looper Posted September 25, 2015 Share Posted September 25, 2015 (edited) Carl. Nice idea! I've also just realised that this lens is well-regarded amongst numismatists. To use it as a 2x macro, I can mount it on my d70 and have bought the reverse mounting ring and a step-up. Photograph the frame on my lightbox. But, I have to say that there's something very satisfying about the idea of projecting a film-based image directly onto a dSLR sensor. Almost poetic. Care to speculate which will give best results? And what were your aims? Once you project a Super8 image directly onto a DSLR sensor you've basically laid the foundation for how digital scanners for film work. Film on one side, sensor on the other, and a lens of your own choosing in between. You then vary the relative distances of each ( just like using an enlarger) to obtain scale and focus. The real trick (and where one ends up spending a lot of time) is in designing and building a rig for it all. My latest rig is one involving computer control of the various components (camera/lens) using stepper motors mounted in custom 3D printed brackets providing full XYZ motion of the lens and Z motion of the camera. The following video is called 16mm Optical Printer, but it's for both Super8 and 16mm as well as film and digital printing. The lens here is a 75mm camera lens. It's reversed with respect to the camera. Using a longer lens gives the whole rig a bit more space in which to move. Shorter lenses make the required distances for scale and focus shorter. Using a high quality enlarger lens is the best bet, or a macro lens for a camera, but for experimentation and getting to understand how it all works any lens will be interesting. The first lens I tried when starting out was a microscope lens held together by cardboard mounts and sticky tape on my kitchen table. Edited September 25, 2015 by Carl Looper Link to comment Share on other sites More sharing options...
Simon Lucas Posted September 25, 2015 Author Share Posted September 25, 2015 (edited) This is digitalized with a Panasonic GH2 and a reformed Bauer T502 projector to 16.66 fps. The lens of the gh2 is a macro Sigma 105 - 2.8, and I used the option to cut that has this camera to fill the frame s8. To remove the extra frames and converted to 25 fps, I have used the software P10JMC v8. Looks wonderful. The quality is so good, it makes it all the more exciting and important to me, to persevere with my own tests and experiments. It means I know it will look great, once I have my own technical issues sorted! thank-you. Btw, why 16.66 fps? Edited September 25, 2015 by Simon Lucas Link to comment Share on other sites More sharing options...
Simon Lucas Posted September 25, 2015 Author Share Posted September 25, 2015 Once you project a Super8 image directly onto a DSLR sensor you've basically laid the foundation for how digital scanners for film work. Film on one side, sensor on the other, and a lens of your own choosing in between. You then vary the relative distances of each ( just like using an enlarger) to obtain scale and focus. The real trick (and where one ends up spending a lot of time) is in designing and building a rig for it all. My latest rig is one involving computer control of the various components (camera/lens) using stepper motors mounted in custom 3D printed brackets providing full XYZ motion of the lens and Z motion of the camera. The following video is called 16mm Optical Printer, but it's for both Super8 and 16mm as well as film and digital printing. The lens here is a 75mm camera lens. It's reversed with respect to the camera. Using a longer lens gives the whole rig a bit more space in which to move. Shorter lenses make the required distances for scale and focus shorter. Using a high quality enlarger lens is the best bet, or a macro lens for a camera, but for experimentation and getting to understand how it all works any lens will be interesting. The first lens I tried when starting out was a microscope lens held together by cardboard mounts and sticky tape on my kitchen table. Carl. An amazing project. Green with envy. So, the stepper is controlling the location of the lens. Is this to gain accurate positioning of the lens at initial set-up or does it have some feedback purpose that applies continuous adjustments during scanning/printing? And, are you using a projector to print the film from, or have you built something with a stepper for frame-by-frame printing? Thanks for the encouragement. I'm looking forward to scanning still frames at the moment, but also I'm now thinking about the potential larger goal of scanning whole films. Link to comment Share on other sites More sharing options...
Carl Looper Posted September 25, 2015 Share Posted September 25, 2015 (edited) Carl. An amazing project. Green with envy. So, the stepper is controlling the location of the lens. Is this to gain accurate positioning of the lens at initial set-up or does it have some feedback purpose that applies continuous adjustments during scanning/printing? And, are you using a projector to print the film from, or have you built something with a stepper for frame-by-frame printing? Thanks for the encouragement. I'm looking forward to scanning still frames at the moment, but also I'm now thinking about the potential larger goal of scanning whole films. Yes, a modified projector is used for film transport. It too is run by a stepper motor - basically just pulled all of the original motor and circuit guts out of the projector and attached a stepper motor to the drive shaft. This allows the projector to be operated under computer control, at whatever frame rate is required (up to about 60 fps), and in step with the camera. When using a film camera the camera drive shaft is also driven by a stepper motor. When using a DSLR the camera is under computer control via a USB cable and over which captures are transferred to the hard drive. The lens and camera positioning allows for adjusting scale and focus, either as a one off prior to scanning/printing, or indeed for programmed frame by frame animation of the camera / lens during scanning/printing. Other applications include dynamic fine adjust for variations in sprocket pitch, and capture of a film frame in small subsections thereof - for subsequent stitching back together to dense pixel results (4K, 8K ...). C Edited September 25, 2015 by Carl Looper Link to comment Share on other sites More sharing options...
Simon Lucas Posted September 26, 2015 Author Share Posted September 26, 2015 (edited) Yes, a modified projector is used for film transport. It too is run by a stepper motor - basically just pulled all of the original motor and circuit guts out of the projector and attached a stepper motor to the drive shaft. This allows the projector to be operated under computer control, at whatever frame rate is required (up to about 60 fps), and in step with the camera. When using a film camera the camera drive shaft is also driven by a stepper motor. When using a DSLR the camera is under computer control via a USB cable and over which captures are transferred to the hard drive. The lens and camera positioning allows for adjusting scale and focus, either as a one off prior to scanning/printing, or indeed for programmed frame by frame animation of the camera / lens during scanning/printing. Other applications include dynamic fine adjust for variations in sprocket pitch, and capture of a film frame in small subsections thereof - for subsequent stitching back together to dense pixel results (4K, 8K ...). C Complete syncing. Sounds fantastic, and very advanced. Did you use this on that beautiful example footage I've seen of yours. I think there some shot on a Logmar, for example. How easy was it to put a stepper in a projector? Edited September 26, 2015 by Simon Lucas Link to comment Share on other sites More sharing options...
Carl Looper Posted September 26, 2015 Share Posted September 26, 2015 (edited) The original footage in that registration test wasn't shot by me. It was shot by Moises Perez (on a Logmar), and scanned on a Lasergraphics ScanStation at Pro8mm. The test I carried out was just on a copy of such - as part of an argument in relation to where the source of rocking in that clip might have originated: be it the camera, or scanner, or some post-stabilisation process. I was arguing that it couldn't be the camera. My dedicated fellow poster was arguing it couldn't be the scanner, so I recon if we're both right it must have been some post-stabilistion step. The stepper motor for the projector was relatively simple. Just had to position the motor in the same location as the original one, with some custom bracketing (3D printed) to accomodate the different shape (or "form factor") of the new motor. C Edited September 26, 2015 by Carl Looper Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now