Super8 signal processing

Forum covering all aspects of small gauge cinematography! This is the main discussion forum.

Moderator: Andreas Wideroe

carllooper
Senior member
Posts: 1206
Joined: Wed Nov 03, 2010 1:00 am
Real name: Carl Looper

Re: Super8 signal processing

Post by carllooper » Mon Nov 15, 2010 9:34 am

Here are the initial proof of concept tests I did a few weeks ago now but posted here:

These have been scaled down (and cropped) from a 3K scan to 1920 x 1080. The source footage is Tri X Super8, shot on a Canon 1014XLS, in 1985.

A single frame:
http://members.iinet.net.au/~carllooper ... eFrame.png

And a single frame integrated from three frames (SR)
http://members.iinet.net.au/~carllooper ... edTest.png

This is interesting - the three frames, as an MPEG4 avi, without any SR:
http://members.iinet.net.au/~carllooper ... otTest.avi

The avi may require installation of the ffdshow codec (highly recommended for any compression work) from here:
http://ffdshow-tryout.sourceforge.net/download.php

Carl
Carl Looper
http://artistfilmworkshop.org/

MustafaUmutSarac
Posts: 2
Joined: Mon Dec 06, 2010 4:25 am
Real name: Mustafa Umut Sarac

Re: Super8 signal processing

Post by MustafaUmutSarac » Mon Dec 06, 2010 5:06 am

Hello ,

This is my first post to the forum and I want to present my best.

I read that LED used as light source at the machine. This is wrong technology. If you look to the spectrometry of new powerful white leds , you find that they have very very strong and sharp two color peaks in their spectrum.
First blue and Second is green.

It is like telecine with latest Mercedes BMW lamps. They are halogen but no different than the white led , dont fall in to the white word too much , they are blue and green.

You must find drum scanner real neutral halogen lamps. But they consume lots of electricity and if you continue to spend 20 seconds per frame , it costs you lots of money but these lamps are so powerful , you can decrease the scanning time too much.

But they are hot also and close range contact would be harm the film so you can use fiber optics and carry the lamp in a safe place.

the last thing about halogen lamps , longer they work , older they are , more yellow the results. You must cool them and shot down in few hours per.

If you want to use as microscope , you can buy a microscope from Leica for 500 dollars and attach Leica histology cameras to the viewfinder or special gate.

It takes enormous detailed instant pictures and disassembling the microscope is faraway easier to construct the setup.

You must use stepper motor to feed with film and electronics , software are all ready. Only you have to do isto attach sprockets to a tube.

Grain is the most exciting , sprit firing thing for me at 8 , If I want grainless , I buy a Sony camera and go.

Mustafa Umut Sarac

Istanbul

User avatar
Nicholas Kovats
Posts: 772
Joined: Sat Mar 25, 2006 7:21 pm
Real name: Nicholas Kovats
Location: Toronto, Canada

Re: Super8 signal processing

Post by Nicholas Kovats » Tue Dec 07, 2010 10:27 pm

Wecome to the list, Mustafa.

Carl Looper could be deeply immersed in the Super 8 processing laboratory at the moment. :)
Nicholas Kovats
Shoot film! facebook.com/UltraPan8WidescreenFilm

carllooper
Senior member
Posts: 1206
Joined: Wed Nov 03, 2010 1:00 am
Real name: Carl Looper

Re: Super8 signal processing

Post by carllooper » Thu Dec 09, 2010 6:58 am

MustafaUmutSarac wrote:I read that LED used as light source at the machine. This is wrong technology. If you look to the spectrometry of new powerful white leds , you find that they have very very strong and sharp two color peaks in their spectrum. First blue and Second is green ... You must find drum scanner real neutral halogen lamps ... You must cool them and shot down in few hours per.
[/quote]
Yes - LEDs are not the best idea.
If you want to use as microscope , you can buy a microscope from Leica for 500 dollars and attach Leica histology cameras to the viewfinder or special gate.
Am already using microsope objective - no problem.
You must use stepper motor to feed with film and electronics , software are all ready. Only you have to do isto attach sprockets to a tube.
Yes - am using a USB computer controlled stepper motor.
Grain is the most exciting , sprit firing thing for me at 8 , If I want grainless , I buy a Sony camera and go.
The degraining apsect of the process is just a side-effect of the process - not the purpose.

Carl
Carl Looper
http://artistfilmworkshop.org/

carllooper
Senior member
Posts: 1206
Joined: Wed Nov 03, 2010 1:00 am
Real name: Carl Looper

Re: Super8 signal processing

Post by carllooper » Thu Dec 09, 2010 11:20 pm

freedom4kids wrote:Carl Looper could be deeply immersed in the Super 8 processing laboratory at the moment. :)
Indeed I am. The forum is very addictive. I have to avoid visiting it somtimes - if I want to get any work done. If an when you sense my absence you can be assured the reason is I am working hard on my "supe-r8solution" project. :)
Carl Looper
http://artistfilmworkshop.org/

MustafaUmutSarac
Posts: 2
Joined: Mon Dec 06, 2010 4:25 am
Real name: Mustafa Umut Sarac

Re: Super8 signal processing

Post by MustafaUmutSarac » Fri Dec 10, 2010 9:06 am

Thank you Carl for your response. I wish all the best at your project.
Histology is a science which research generally human tissues.
Every nuance at the colors , details is create so much difference.
Leica is one of the oldest microscope factory and the best.
First Leitz 35mm lens was Elmax and it is Ernst Leitz , Max Berek.
Max Berek is the best microscope optics , mineralogy microscope expert at the time and he developed many Leitz lenses.
Leica have a lots of cheap microscopes and their cameras.
If you could not succeed at your project , may be you look to this concept.

If I am not wrong , halogen lamps have very higher voltage use and findinng a suitable and small source is expensive and difficult.

If I am not wrong , some of the glasses of halogen lamps are created to filter the UV light to protect the garments at the shops.

UV is the enemy of film polyester base and the emulsion , They tend to crack.

Other useful technology is to project films to a half transparent glass and take the pictures with a big digital camera from the other side.

This technology is used at film industry for to get the more film like pictures from digital cameras and able to use every kind of big , small lenses with digital cinema cameras.

I think above technology is widely used , trusted and cheapest setup.

Thank you freedomforkids to your welcoming words.

Best ,

Mustafa Umut Sarac

Istanbul

carllooper
Senior member
Posts: 1206
Joined: Wed Nov 03, 2010 1:00 am
Real name: Carl Looper

Re: Super8 signal processing

Post by carllooper » Mon Dec 13, 2010 6:42 am

MustafaUmutSarac wrote:Leica have a lots of cheap microscopes and their cameras.If you could not succeed at your project , may be you look to this concept.
Thanks Mustapha. Much appreciated. The lens I'm using is adequate but could be better. Cost me about US140. While it's a planar I've noticed there is slight softening towards the edges - so the idea of using a better lens - such as the ones you have suggested (Lecia lens) I will consider seriously.

The other option is a good quality enlarger lens suggested by peaceman. This is probably a better idea than a microscope objective because an enlarger lens is designed to project onto flat photographic paper - or in my case is the flat camera sensor.

Regarding light source I'm of the mind to keep the light level low, or bright and short (flash) so as not to risk damage to the film. I'm not too concerned about the light being the correct colour because it can be corrected through multiple exposures (one for each colour component).

For example, suppose the light source (without any film) registers the following values for a 1 second exposure:

R = 100%
G = 75%
B = 50%

Then by exposing a film three times, with the following intervals:

1 second
1.33 secs
2 secs

one can combine red from the first, green from the second, and blue form the third to get a full bandwidth copy.

Carl
Carl Looper
http://artistfilmworkshop.org/

carllooper
Senior member
Posts: 1206
Joined: Wed Nov 03, 2010 1:00 am
Real name: Carl Looper

Re: Super8 signal processing

Post by carllooper » Mon Dec 13, 2010 11:59 pm

MustafaUmutSarac wrote:Leica have a lots of cheap microscopes
Hi Mustapha,
I'm currently looking at the Leica website for microscope objectives:

http://www.leica-microsystems.com/produ ... bjectives/

The objective I'm currently using is a 4X objective (required Super8 to Sensor magnification).

The site doesn't list any 4X objectives so I'm assuming the lenses on the site are rated in terms of the magnification that would occur when combined with a generic microscope eyepiece (10X).

So the one I'd need would be labelled 40X - (actually being a 4X objective).

Would you agree?

Carl
Carl Looper
http://artistfilmworkshop.org/

Jonathan
Posts: 14
Joined: Tue Nov 12, 2002 9:21 pm

Re: Super8 signal processing

Post by Jonathan » Tue Dec 14, 2010 10:20 am

Hi Carl,

no, the eyepiece magnification is not factored in. Microscope objectives are usually rated with the magnification they achieve alone, since you can combine them with various eyepieces. There are actually quite a few 4x lenses there, search in the range 0.4-6.3x. You probably want no immersion (dry), and brightfield illumination.
Be aware that most modern microscope lenses do not directly create an image, but are corrected for an infinite tube length. A fully equipped microscope has (several) objective lenses, corrected for infinite tube length, a tube with extra optical elements which creates a real image, e.g. on the CCD of a microscope camera, or as an intermediate image that is then viewed with the eyepiece. The mag. of the image on the CCD is the one which is printed on the objective lens.

You might get away with still trying to use a small magnification lens and just focus on the CCD anyway. I don't know if the image quality is optimal then. The proper way would be to get a stand alone tube lens to form an image. Or you get one of the older finite conjugate microscope lenses which directly focus to form a real image.
You can e.g. look at what Edmund Optics have, but I have never used one of their finite conjugate lenses.

http://www.edmundoptics.com/onlinecatal ... oryid=1739

Best regards,

Jonathan.

carllooper
Senior member
Posts: 1206
Joined: Wed Nov 03, 2010 1:00 am
Real name: Carl Looper

Re: Super8 signal processing

Post by carllooper » Tue Dec 14, 2010 10:49 pm

Jonathan wrote:no, the eyepiece magnification is not factored in. Microscope objectives are usually rated with the magnification they achieve alone, since you can combine them with various eyepieces.
Hi Johnathon,

thanks for that. Re-reading the Leica website I see I was misreading the 0.4 to 6.3 reference as 0.4 to 0.63. My mistake.

Have been using a planar 4X objective (and labelled as such) for some months now. It forms a real image. I have it mounted to a bellows (via an adapter ring) and it projects the Super8 image directly onto the sensor. I chose 4X since the sensor is about 4X the size of the Super8 frame.

However I can get variations in the magnification by altering the focal distance, and projection length of the image.

Now it does give very good results.

This is TriX film, imaged at 3K wide, super-resolved by 1.7X, and downscaled to 1920 x 1080:

http://members.iinet.net.au/~carllooper ... edTest.png

But there is a very slight loss of focus towards the edges (when focused on the centre of the image) indicating the lens is not as planar as I might have otherwise appreciated.

And so am considering getting a better quality objective - one that is more planar than the one I already have.

Do you know what the difference is between objectives rated as "HI PLAN", "N PLAN" and "FL PLAN". And which would be more planar/flatter?

cheers
Carl
Carl Looper
http://artistfilmworkshop.org/

christoph
Senior member
Posts: 2486
Joined: Fri Jul 25, 2003 2:36 pm
Location: atm Berlin, Germany

Re: Super8 signal processing

Post by christoph » Wed Dec 15, 2010 12:52 am

hiya,

save the money and get a used enlarger lens. like a componon-s or rodagon APO if you want to be fancy. mount it reversed if you have a large sensor camera.

it will cost you about 50-100EUR on ebay and they are as good as it gets.
++ christoph

User avatar
peaceman
Posts: 105
Joined: Wed Feb 09, 2005 12:17 am
Real name: Friedemann Wachsmuth
Location: Germany
Contact:

Re: Super8 signal processing

Post by peaceman » Wed Dec 15, 2010 11:40 am

christoph is so 100% right. :)

carllooper
Senior member
Posts: 1206
Joined: Wed Nov 03, 2010 1:00 am
Real name: Carl Looper

Re: Super8 signal processing

Post by carllooper » Fri Jan 07, 2011 7:04 am

peaceman wrote:christoph is so 100% right. :)
Ok. Sounds good to me. Will do that in due in course. :)

ps. The microscope objective I'm using at the moment cost about AUS 100 from Shanghai.
Carl Looper
http://artistfilmworkshop.org/

User avatar
Nicholas Kovats
Posts: 772
Joined: Sat Mar 25, 2006 7:21 pm
Real name: Nicholas Kovats
Location: Toronto, Canada

Re: Super8 signal processing

Post by Nicholas Kovats » Fri Jan 07, 2011 12:56 pm

According to Peaceman....specifically the 50mmf2.8 Rodenstock APO Rodagon N. I just missed one on eBay which sold for approximately $80 US.

http://www.bhphotovideo.com/c/product/4 ... gon_N.html
Nicholas Kovats
Shoot film! facebook.com/UltraPan8WidescreenFilm

carllooper
Senior member
Posts: 1206
Joined: Wed Nov 03, 2010 1:00 am
Real name: Carl Looper

Re: Super8 signal processing

Post by carllooper » Sun Jan 09, 2011 12:39 am

freedom4kids wrote:According to Peaceman....specifically the 50mmf2.8 Rodenstock APO Rodagon N. I just missed one on eBay which sold for approximately $80 US.
http://www.bhphotovideo.com/c/product/4 ... gon_N.html
Cool. I'm going to keep an eye out for one of those. So, the lens will need to be mounted "back to front" so to speak.

The bellows I'm using has the so called "universal thread" for which there were readily available adapters to the microscope objective I was using.

But with the enlarger lens I imagine I'll need a somewhat weird adapter - assuming the lens has a thread on the front side (for filters?). So the adapter would need to be a filter-thread-to-universal-thread adapter.

Would such adapters exist?

Or will it need to be a custom DIY mounting job required.

Carl
Carl Looper
http://artistfilmworkshop.org/

Post Reply