Camera Testing

[IvanKane]

Hi

I have an Eclair NPR and I need to test it to see if everything is a OK.

Someone told me that I should draw up a grid then shoot it, rewind and shoot it again in the exact potision. Then project the image and see if the lines are shaking or out of line, if not then everything should be alright. Now could anyone please explain to me how to do it cause this doesn't really make sense to me?

[David Mullen ASC]

It's to check steadiness & registration. If you simply shot a grid and projected it, you could not tell the camera's steadiness apart from the projector's steadiness.

 

However, if you double-exposed the same grid in two passes on the same film, offset (shoot the second pass slightly panned over) so you could see both sets of grids on the same frame (not perfectly lined-up so that it looked like the same grid), then any camera steadiness problems would show up because you'd see the lines of one grid shaking or moving around RELATIVE to the other grid.

 

But it's important that when shooting the two exposures of the grid that the camera is rock-steady on the tripod so that you don't add shaking from that.

[Nathan Milford]

This is how I typically shoot a registration test in the shop (and I do it almost daily):

 

We have a special rig, but you can tape up a piece of paper with a grid on it. It is better if the background is black and the gridlines are while (I'll explain later), but if you can't make something like that in mspaint or photoshop any grid will do.

 

Setup the camera on the tripod and make sure it's not going to move and make sure no one around moves it. Light the grid (make sure you're underexposing by a stop for the double exposure) and line up the grid in the TV safe pumpkin in your viewfinder, focus then put some tape over the eyepiece so light doesn't get in and fog your film while you run the test.

 

I shoot one 50 foot pass at 24 frames per second on each magazine. The mag can sometimes be the cause of the registration problem as well and it's best if you shoot at least three mags if you have them as this gives you a better idea what might be going wrong (i.e. one mags fails and two pass, its the mag, if all fail it's the camera body).

 

This time tilt the camera a bit so the lines will not be overlayed by the lines from the first pass. I rewind each magazine in a dark room and re-thread it up. For each mag, on the second pass, I shoot 25 feet (of the original 50 feet) at 24 fps, then the last 25 feet at top speed (32, 50, 75fps, or whatever). This second pass you should slate the speed and the magazine serial number. I just put a piece of white gaffers tape and a sharpie on the grid out of the way.

 

I put the three or four cores of 50 feet in a can and send it off to the lab. When you send it to the lab tell them to DEVELOP NORMAL and to make NO PRINT. You will then screen the negative (the black background will then be white and the white lines will then be black, its easier to read), as you don't need the redundancy of a print, and watch if the lines move. Some, minor breathing, movement can be acceptable... you should be able to tell if the camera has a problem. If you don't have the facility to screen the negative, the lab may do it for you if you're nice >8)

 

If the camera fails, send the camera AND the test to somewhere like Du-All where they can rectify it. Most experienced technicians can look at a reg test and intuit from the type of line movement what might be causing the problem.

 

Good luck,

nathan

[IvanKane]

This is how I typically shoot a registration test in the shop (and I do it almost daily):

 

We have a special rig, but you can tape up a piece of paper with a grid on it.  It is better if the background is black and the gridlines are while (I'll explain later), but if you can't make something like that in mspaint or photoshop any grid will do.

 

Setup the camera on the tripod and make sure it's not going to move and make sure no one around moves it.  Light the grid (make sure you're underexposing by a stop for the double exposure) and line up the grid in the TV safe pumpkin in your viewfinder, focus then put some tape over the eyepiece so light doesn't get in and fog your film while you run the test.

 

I shoot one 50 foot pass at 24 frames per second on each magazine.  The mag can sometimes be the cause of the registration problem as well and it's best if you shoot at least three mags if you have them as this gives you a better idea what might be going wrong (i.e. one mags fails and two pass, its the mag, if all fail it's the camera body).

 

This time tilt the camera a bit so the lines will not be overlayed by the lines from the first pass.  I rewind each magazine in a dark room and re-thread it up.  For each mag, on the second pass, I shoot 25 feet (of the original 50 feet) at 24 fps, then the last 25 feet at top speed (32, 50, 75fps, or whatever).  This second pass you should slate the speed and the magazine serial number.  I just put a piece of white gaffers tape and a sharpie on the grid out of the way.

 

I put the three or four cores of 50 feet in a can and send it off to the lab.  When you send it to the lab tell them to DEVELOP NORMAL and to make NO PRINT.  You will then screen the negative (the black background will then be white and the white lines will then be black, its easier to read), as you don't need the redundancy of a print, and watch if the lines move.  Some, minor breathing, movement can be acceptable... you should be able to tell if the camera has a problem.  If you don't have the facility to screen the negative, the lab may do it for you if you're nice >8)

 

If the camera fails, send the camera AND the test to somewhere like Du-All where they can rectify it.  Most experienced technicians can look at a reg test and intuit from the type of line movement what might be causing the problem.

 

Good luck,

nathan

<{POST_SNAPBACK}>

 

 

Thank you so much for your time to help.

I really didn't expect anyone to actually give me such detail explanation.

Best Regards,

Ivan

[IvanKane]

It's to check steadiness & registration. If you simply shot a grid and projected it, you could not tell the camera's steadiness apart from the projector's steadiness.

 

However, if you double-exposed the same grid in two passes on the same film, offset (shoot the second pass slightly panned over) so you could see both sets of grids on the same frame (not perfectly lined-up so that it looked like the same grid), then any camera steadiness problems would show up because you'd see the lines of one grid shaking or moving around RELATIVE to the other grid.

 

But it's important that when shooting the two exposures of the grid that the camera is rock-steady on the tripod so that you don't add shaking from that.

<{POST_SNAPBACK}>

 

 

Thanks David :)

[Scot McPhie]

Thank you so much for your time to help.

I really didn't expect anyone to actually give me such detail explanation.

Best Regards,

Ivan

<{POST_SNAPBACK}>

 

 

Ditto for me too - this is just what Ineed to test out my cantankerous R16 and highly suspect magazines :-)

 

Scot

[Miguel Bunster]

Hi,

 

I was checking this thread because I am shooting in Chile with an SR3 and want to test it today. What I want to test is a registration test, lens test and magz.

This thread is great for registration test but i wanted to shoot 25ft first and after telecine just overlay the same clip but offset the time by 1/2 a second so i can se relative movement (and offset the video placement). I knwo its not ideal but moeny is an issue so cant print or shoot to much film. I think it can work pretty well (resolution is low to judge any minor movment).Now for lense test i am ussing the Carl Zeis super speed and wanted some recomendations to test them. (Once agian not much film to do so). I was planing to concentrate on testing the wider apertures for any aberrations. Any ideas would be epic. Thanks!

And well gona scratch test the magz and see any light leaks.

I know ideally a lens test should be longer but budget wis i got like 300ft for this test....

ANy input would be great.

Thanks very much!

M

[David Mullen ASC]

You need to do the overlay (double-exposure) in camera to check registration -- any other method and you're introducing outside factors so won't know if the problem is due to the camera's steadiness or the telecine, etc.

 

Also, you don't need to print the steadiness test -- you could project the processed original negative if all you are looking for is how the two grids move in relation to each other.