MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Tips for photographing CRT displays?
MUFF WIGGLER Forum Index -> Video Synthesis Goto page 1, 2  Next [all]
Author Tips for photographing CRT displays?
lizlarsen
I was wondering if anyone has any advice on getting a proper photograph of a CRT display using a film camera (or otherwise)?

I'd really like to be able to get some high quality enlargements of the analog output of my system for stills and album covers, etc, without being restricted to enlarging digital captures, and to retain some of the CRT glow and scanlines in the image.

I assume there must be some technique or device for synchronizing camera exposure to open and close for a single frame of video to get proper exposure, and creating a logic signal to do exactly that would not be a problem, but I don't know enough about cameras to know where to start on the mechanical end. Any advice appreciated.
tIB
Google is your friend:

http://www.google.co.uk/search?q=photographin+crt&rls=com.microsoft:en -GB:{referrer:source?}&ie=UTF-8&oe=UTF-8&sourceid=ie7&rlz=1I7AMSA&redi r_esc=&ei=yRnYTfK2Csyv8QOSueGDBQ

http://www.digital-photography-school.com/how-to-photograph-an-image-o n-a-tv

http://www.webdesign.org/photoshop/articles/photographing-crt-televisi ons-and-monitors.6682.html
A Dingleberry Monstrosity
....<print screen>?
A Dingleberry Monstrosity
A Dingleberry Monstrosity wrote:
....<print screen>?


...sorry, did a "find all new posts" search and thought this was in off topic d'oh!
deastman
A Dingleberry Monstrosity wrote:
....<print screen>?
It's not quite the same... the pixels will glow and blend together to a certain extent. Nothing you couldn't accomplish in Photoshop, of course, but I think most people here can appreciate the joy of a purely analog approach.
lizlarsen
I have done some googling on the subject, but haven't ran across techniques other than just adjusting shutter speed to match frame rate (which is a viable technique). But that still isn't as accurate as what I'd like, which is to expose for one frame only, in sync with the video. With the shutter speed technique, you capture for the duration of one frame, but you may end up with a third of frame 1 and two thirds of frame 2. This prevents banding, but I'd like to be able to expose for all of 1 frame only, and this would require some sort of synchronization I expect. I'm curious if techniques for doing this already exist.

And yes, nothing that ends up captured on the computer is nearly the same as viewing the direct output of the video synthesizer on a nice big CRT. smile Trying to digitally upscale a 720x480 capture of a DV stream for something like a large-scale print or an album cover, etc, is just not going to be the same.
tIB
Matching shutter speed to screen rate seems a good start- assuming its ntsc standard you are looking at shooting 1/60 of a second to capture one frame... or am i missing something?

Could you look at modifying a remote shutter trigger so it triggers to your logic signal at the start of each frame? Ideas here that may be useful:

http://www.youtube.com/watch?v=rUDzM4IuF6w
http://www.instructables.com/id/Remote-shutter-trigger-for-Canon-Camer as/

Think you're going to have better results shooting onto film btw. WHat you planning to use camera wise?
lizlarsen
Quote:
Could you look at modifying a remote shutter trigger so it triggers to your logic signal at the start of each frame?


This! This is exactly what I'm asking about. smile Getting a logic pulse that is precisely one video frame out of the video synthesizer, and in sync with the video source feeding the CRT display is no problem at all... it's just that I know next to nothing about still photography and gear.

So getting some sort of shutter trigger that responds quickly enough on a rising edge pulse and pre-set to 1/30 shutter speed, OR a shutter gate, that would open and close with the incoming logic trigger, either way would work. (I imagine the latter would be more interesting...) I guess that's what I'm talking about.

Quote:
Think you're going to have better results shooting onto film btw. WHat you planning to use camera wise?


Yeah, shooting on film seems like the way to go. I can then do an extremely high res scan for enlargement. Grain and grittiness in the final image isn't a problem, i just want to eliminate any pixelization or digital artifacting from my final image as possible (other than what results after a hi-res scan of course).

I know little about still cameras. What would be a good workhorse, affordable film camera that may work here, with an external shutter trigger?
kjellb
In the old days of tape we just made a freeze frame and used long exposure. One second will get rid of any blanking artifacts. One way to go is capture to a digital device, make a freeze frame and then make an exposure.
tIB
I might be going beyond my knowledge and your needs here but I think medium format might be your friend. Id look into the older mamiya, rolleiflex and voigtlander. Ive no idea which of those have remote triggering though, or how it's set up on older cameras... I suspect its via pin rather than voltage trigger.

I suspect you might be better off concentrating on shutter speed and exposure rather than getting too bogged down in external voltage shutter triggers- if you shot a few rolls of good quality medium format film, on a decent camera, with the right settings you are going to get results.

Ive asked my mate who is/was a photographer... Ill update when he gets back to me as I could be talking complete shit! lol
lizlarsen
This all sounds like great advice. Thanks guys. smile

The main reason I'm worried about synchronization is that a lot of the images I want to capture are often the really chaotic patterns that have a lot of movement in them, which makes it difficult, creating an interlaced effect or blur if the exposure was doubled over between frames. I'm not stressing about it too much though, I just mainly am curious if it's possible.

Freeze frame and long exposure seems like a great idea too. I'll have to try that with the TBC module we're developing.
tIB
Id be tempted to do a ton of experimenting with shutter speeds and exposures in digital, and then figure out where you wanted to go in film based on the results of that. In terms of what you mention Im thinking fast shutter speed but Im not sure if youd get banding with a multiplication of the screen rate... hmmm.....
deastman
You still can't entirely eliminate pixelization... the CRT has an array of distinct RGB pixels that you're filming. As I mentioned, they will glow and blend together somewhat, but you're still dealing with a finite resolution.
lizlarsen
Well I suppose the minimum you'd want to expose for to get a whole frame of NTSC video is 1/30th of a second, which I guess in film terms is pretty slow. At 1/60th second shutter speed you capture one field, which is really only half of the image. Whether that makes a big difference or not I suppose would just depend on the source in question. Anything less than 1/60th of a second, I'd expect you would see only a portion of the picture.
ndkent
If you can take a long exposure, that's by far the easiest. You want a bunch of fields. You might guess say a 16th of a second is enough, but it has to be long enough that any partial fields are averaged. Too short of an exposure is obviously a black bar on the screen, but a longer exposure will get you part of the screen brighter than the other part. It has to be long enough that the fraction of the frame being exposed again averages with the complete frames.

On TV and Cinema sets they use a custom playback system that's synchronized
lizlarsen
Quote:
You still can't entirely eliminate pixelization... the CRT has an array of distinct RGB pixels that you're filming. As I mentioned, they will glow and blend together somewhat, but you're still dealing with a finite resolution.


Yes, but these aren't pixels in the sense of discrete values of data, what you're describing is the scanning grid of the monitor. The sweep of the electron beam from left to right is a continuous voltage without any steps or pixel grid. Of course it is limited by bandwidth, the number of scanlines, and the monitor's characteristics, but that's an altogether different thing (and visual presence) than the resolution of a digital image. Anyway, I think you get what I mean. smile It's the compression artifacts and blocky digital compression that I'd like to avoid -- I'd like for the first digitization of the image to be in extremely high resolution.
lizlarsen
Quote:
On TV and Cinema sets they use a custom playback system that's synchronized


I'd love to know more about that! Especially how they sync'ed film cameras up to do this for the right exposure speeds.
If there were an easy way to synchronize a camera shooting still images at HD resolutions to a composite feed to a CRT monitor, capturing 1 per frame, that would be pretty incredible.
daverj
There are two tried and true methods from the old days

Method 1 - Shoot a long exposure. Minimum of 1/15th second. 1/6th or even slower is better. Works best if you can freeze frame the image (a TBC can do that) You want to get several frames in the exposure to get rid of the partial fields that happen during opening and closing the shutter.

Method 2 - shoot in a dark room. Turn off the video. Open the camera shutter (time exposure). Use a vertical interval switcher to flash exactly one frame on the screen. Close the shutter.

As for synchronized shutters, that is with motion picture cameras. But the video is synchronized (genlocked) to the camera shutter. Not the other way around. I built video switchers in the 90s that were used for this in the Hollywood movie "Sliver". The switchers would handle the 24 frame video format used by movie studios for synchronized video. They had a wall of monitors in the movie playing synchronized video and my switchers were used to change the images across the bank of monitors.

The closest you could come to that with a still camera is to take the flash trigger output from the camera and use it to trigger a video switch to flash one frame of video. Then take an exposure that is long enough that the shutter has fully opened before the video flashes and doesn't close until after the video flashes. You'd need a delay after the flash trigger so the video doesn't flash to early (need to allow the shutter to open).
Cat-A-Tonic
Nikon used to make a lens (quite rare) that was specifically designed for shooting oscilloscopes.

Maybe a lens with a low F-stop, like one of the Nikon 50mm lenses, might do the trick.
great for low light situations, and long exposures with lovely color.

These would probably be good for that vector rescanning thing you have been doing Lars.

on CRT...
hmmm.....
pause the video?
use an LCD screen?
wwall
Another thing you might want to consider is the movement of the camera's shutter curtain during exposure. If you're exposing for 1/30 second and the shutter is opening horizontally while the CRT beam is sweeping vertically, you will only partially capture the image. Of course, that depends on the design of the camera. Some may be better than others for this purpose.
CursedFrogurt
I'm having trouble seeing why shooting with a digital camera is not an option here. I can see why you might not want to do direct to computer digital capture if you can't end up with the resolution you want, but with a decent DSLR and some experimentation with shutter speed and exposure settings, I'm sure you can capture a good image. The only thing film introduces into this equation is the uncertainty of having to wait a few days while your prints are developed, and making experimentation cost a shitload more. I love working with film, I think it is a wonderful process, but in terms of basic image capture in this situation, I think there's no real benefit to film. A decent consumer-grade digital slr is going to put out an image that's big enough to blow up to very large proportions without noticeable pixelation. Maybe there's something I'm not seeing, but honestly if you don't have a decent DSLR, I'd borrow one from a friend and shoot a couple hundred pictures until you get the one you want. I think you'll spend far less time finding the right shutter settings and then taking 200 pictures in a row than you will setting up some crazy diy screen sync device.

A good digital camera is probably the most direct way to get a high-resolution digitization of your image.
andrewF
I have to take shots of my 'scope a lot for my research. Always turn off the lights.
I can only get good images with my archaic Canon powershot A70, flash off/macro on. My newer nikon coolpix is rubbish for 'scope shots.
Some of my pics are HERE

In the good old days there were polaroid cameras designed to mount on 'scopes. Still around but film is pretty much all gone

lizlarsen
This has been a totally fascinating discussion! Thanks for all the posts.

I think digital is going to be how I start experimenting for sure. I have a Canon HFS100 video camera that can do stills, and came with someone's DSLR lens and some accessories for it... but I haven't messed with the DSLR stuff yet, and I'm not sure what shutter settings I have.

Thanks for teaching me stuff guys!

Andrew those scope shots are totally gorgeous.
antimatter
andrewF wrote:
I have to take shots of my 'scope a lot for my research. Always turn off the lights.
I can only get good images with my archaic Canon powershot A70, flash off/macro on. My newer nikon coolpix is rubbish for 'scope shots.
Some of my pics are HERE

In the good old days there were polaroid cameras designed to mount on 'scopes. Still around but film is pretty much all gone



Ha, yes , 'the good old days' indeed !

nice stuff , Andrew, filmic , what scope are you using?
ndkent
creatorlars wrote:
Quote:
On TV and Cinema sets they use a custom playback system that's synchronized


I'd love to know more about that! Especially how they sync'ed film cameras up to do this for the right exposure speeds.
If there were an easy way to synchronize a camera shooting still images at HD resolutions to a composite feed to a CRT monitor, capturing 1 per frame, that would be pretty incredible.


As I understood it from my film production days the cameras are already locked to crystal sync since the sound needs to be too (or else you lose sync on the sound). The video playback system is altered to run in sync with the recording gear which is not altered. You tend to hire the system and the technicians if you do a pro shoot.

I did get a chance to mess with a low tech varispeed camera once when I was fishing around to see if I could kinescope material but it proved impossible I guess I was running up against the blades in the shutter and once I got it slow I think I started seeing hot bands. I sent my material off to a professional kinescope video to film place that was still in commercial operation (circa 1987) for 16mm.

Of course there aren't actual pixels in analog video but you do eventually see the r, g and b colored physical phosphor elements on a typical color monitor with a high res camera. I remember in the 90s we had a 4x5 film recorder at work for making chromes of computer files. Inside it was a high end high res black and white monitor, I forget, maybe 4K lines. It would flip through the rgb channels one at a time and you'd hear the click of the color filter between the crt and the camera.
MUFF WIGGLER Forum Index -> Video Synthesis Goto page 1, 2  Next [all]
Page 1 of 2
Powered by phpBB © phpBB Group