Large format negative from digital display

[salvatore]

I am interested in dichromated gum and carbon techniques, both needing large negatives.
In this thread I like to express my ideas and get some suggestions on a way to obtain them in case of a digital photograph, and avoiding the use of an ink jet printer, due to the ease with which they get clogged and rapidly useless unless continuosly used.
I like to build a device which will project the image on a lcd screen, say that of a computer or of a TV set of high definition (think of a 4K!) onto a black and white photographic film.
The size of the lcd could be from say 15 by 20 inches to 45 to 60 inches.
My basic question is: Which lens to use? Which is the optimum focal length? Any model in mind?

[ndg]

About 2 years ago, one of my colleagues saw some of my wet plates and wondered if I could do a portrait of his horse. Wet plate portrait of a horse? I laughed it off bit kept thinking. This what I did:
- I shot a digital image of the horse and saved it on a jump drive.
- I have a 42" LG 120 Hz TV in the studio that I screened the image on.
- I placed the TV with the image of the horse on it in front of my camera with my 16" Tessar lens and focussed the image on the GG.
- I prepared a 12" x 12" plate and captured this image:
127121
(bad scan; scan it in 2 parts and tried to stitch it!)
It came out so well, they have it hanging in their home!

So I tried a second portrait but this time on 11x 14 photo paper using a Verito.
I developed the photo paper and contact-printed it on Ilford 300 paper.
127122

I thought of writing it up but never can make the time. Beside I haven't used that method much anymore because I didn't really like the results.
The images have the signature of the first lens used and not so much of the second lens. The images also look like copies.
I prefer digital negatives muh more than this method.

So if you real want to, get an LCD or LED screen with as much brightness as possible. The size will depend on how big the negatives should be. You can go wrong with a 42".
Any lens will do. You can meter the brightness of the screen to give you an approximate ballpark for exposure.

[Tin Can]

About 2 years ago, one of my colleagues saw some of my wet plates and wondered if I could do a portrait of his horse. Wet plate portrait of a horse? I laughed it off bit kept thinking. This what I did:
- I shot a digital image of the horse and saved it on a jump drive.
- I have a 42" LG 120 Hz TV in the studio that I screened the image on.
- I placed the TV with the image of the horse on it in front of my camera with my 16" Tessar lens and focussed the image on the GG.
- I prepared a 12" x 12" plate and captured this image:
127121
(bad scan; scan it in 2 parts and tried to stitch it!)
It came out so well, they have it hanging in their home!

So I tried a second portrait but this time on 11x 14 photo paper using a Verito.
I developed the photo paper and contact-printed it on Ilford 300 paper.
127122

I thought of writing it up but never can make the time. Beside I haven't used that method much lately.

So get an LCD or LED screen with as much brightness as possible. The size will depend on how big the negatives should be. You can go wrong with a 42".
Any lens will do. You can meter the brightness of the screen to give you an approximate ballpark for exposure.

There goes the neighborhood. :)

Great idea!

[ndg]

I haven't used this method much anymore because I don't really like the results.
The images have the signature of the first lens used and not so much of the second lens. The images also look like copies.
I prefer digital and in-camera negatives much more than this method.

[nede]

Did it this week with my 8x10 camera on my imac27 screen
I pût some low contrast on the image and shoot it, measuring with my spotmeter.
Nice result but you can Count the pixel on the négative .... Not a very good idea in the end...

[ndg]

I got the best results when I used the 36 mp images form my Nikon d800E. With those, even the 11x14 images were satisfactory.
Lower res images tend to show pixelation. One could also match the size of the image to that of the screen to get optimal results, much like when printing digitally.

[salvatore]

When using the 36 mp image, which screen and which lens did you use?
Thanks.

[ndg]

When using the 36 mp image, which screen and which lens did you use?
Thanks.

The screen was an LG 42" LED 120Hz screen. If I remember right, one lens was 16" and the other 18" for 11x14. For 8x10, I tried a 14".

[salvatore]

The screen was an LG 42" LED 120Hz screen. If I remember right, one lens was 16" and the other 18" for 11x14. For 8x10, I tried a 14".

Thanks for the information.
The "lenses" were just enlarging glasses (one diopter) or a composite chromatic and distortion corrected objectives (of cameras, of enlarger, of projection)?
With some plywood I built a very rough optical bench composed of an image source (the screen of a Google Nexus 7 (about 12 by eighteen centimeters) a simple enlarging lens of about 10 cm focal length, diaphragmed by an 8 mm hole, then a plexiglas plate.
From the image source to the lens about 20 cm and same from lens to plexiglas.
I generated a test pattern and observed the projected image with a small enlarging glass made for the purpose of focus optimization by Kodak, by which I could observe, in the central portion of the image of about 3 by 3 centimeters a very sharp definition showing clearly the pixels.
Of course I would like to use some more efficient optical device than a simple lens, and for that I plan to try the following optics I have, remnants of my past activity in analogic photography:

Enlargement lens: Rodenstock Rodagon 50mm
Camera lens 1: Nikkor 35mm for Nikkormat camera
Camera lens 2: Nikkor 110mm for Nikkormat camera.

However, I would like to have the best for my application, and I would like some suggestion do build something which could give from a source ranging from 10 by 15 cm to 30 by 45 cm to an image of same size (possibly also a reduction or enlargement in the given range.
Which lens for this purpose? May be a large format camera lens of aroun 200-300 mm focal or an enlarging optics?
Lenses made for grqphic arts seem the best but prohibitive in price as new.

[ndg]

Thanks for the information.
The "lenses" were just enlarging glasses (one diopter) or a composite chromatic and distortion corrected objectives (of cameras, of enlarger, of projection)?
With some plywood I built a very rough optical bench composed of an image source (the screen of a Google Nexus 7 (about 12 by eighteen centimeters) a simple enlarging lens of about 10 cm focal length, diaphragmed by an 8 mm hole, then a plexiglas plate.
From the image source to the lens about 20 cm and same from lens to plexiglas.
I generated a test pattern and observed the projected image with a small enlarging glass made for the purpose of focus optimization by Kodak, by which I could observe, in the central portion of the image of about 3 by 3 centimeters a very sharp definition showing clearly the pixels.
Of course I would like to use some more efficient optical device than a simple lens, and for that I plan to try the following optics I have, remnants of my past activity in analogic photography:

Enlargement lens: Rodenstock Rodagon 50mm
Camera lens 1: Nikkor 35mm for Nikkormat camera
Camera lens 2: Nikkor 110mm for Nikkormat camera.

However, I would have the best for my application, and I would like some suggestion do build something which could give from a source ranging from 10 by 15 cm to 30 by 45 cm to an image of same size (possibly also a reduction or enlargement in the given range.
Which lens for this purpose? May be a large format camera lens of aroun 200-300 mm focal or an enlarging optics?
Lenses made for grqphic arts seem the best but prohibitive in price as new.

What about an Artar?

[salvatore]

What about an Artar?

I just bought an Artar lens, non red dot (uncoated), 12" focal just to start playing with this kind of lens.
Do you think that I can use it to cover an 8x10 " film?
It was rather cheap, but if works fine enough I will then go to a red dot, coated and more modern lens (obviously more expensive).
Thanks for the advise, anyhow.

[salvatore]

What about an Artar?

I like to thank you for your advice and to have let me know the existence of this class of optics I did not know before.
As I read from the web they seem the best suited for photoreproduction, which I like to practice.
I ask you again some help in deciding which type of Artar lens to buy.
I have a budget of around 500 US$ and I saw some red dot Artars in this range.
There are different focal length, say 16 1/2 " and 19", letting apart the 24" that I believe more difficult to deal with.
I like to cover a film of 18x24cm (8x10"), but I like to leave some room for slightly larger films, always of the radiographic type.
What is your suggestion?

[ndg]

Replied to your PM.


I like to thank you for your advice and to have let me know the existence of this class of optics I did not know before.
As I read from the web they seem the best suited for photoreproduction, which I like to practice.
I ask you again some help in deciding which type of Artar lens to buy.
I have a budget of around 500 US$ and I saw some red dot Artars in this range.
There are different focal length, say 16 1/2 " and 19", letting apart the 24" that I believe more difficult to deal with.
I like to cover a film of 18x24cm (8x10"), but I like to leave some room for slightly larger films, always of the radiographic type.
What is your suggestion?

[salvatore]

Hi Nana Dadzie,
A few words on the progress of my project.
I bought a 19" Artar red dot lens and build a primitive camera mounting such lens.
A computer screen was projected (1 to 1) on a radiographic Fuji green film and developed with HC110 1:50 for 10 minutes.
I could not get a large optical density, even by extending exposure time.
By inspecting the negative with a magnifier I could clearly detect the single pixels, and I understood why I was limited in o.d.
Since the film is sensitive only to green (and perhaps a little to blue, but not to red), only a part of the pixel impresses the film.
Actually, the transmission of the impressed film could range from 66%(only green) to 33%(green and blue), and even more, considering the inactive part of the pixel.
This means an o.d. of 0.5 to 0.18, a very poor result.
Such a situation is very bad for the direct production of a negative, and acceptable to produce an intermediate positive to contact print to get the desired negative.
To overcome this problem (I do not like to need two films) I plan to build a device which moves the image up down and left right to the extent of a pixel (about 0.2 mm) to impress homogenously the whole pixel with the green one.
If it works fine I will describe it in detail.
Does anybody know of a similar approach?

[jp]

Really sounding like the wrong tool. Like when I need a hammer and only find my pipe wrench which is also capable of pounding in a staple because the stapler is not working.

[mdarnton]

Doesn't one of the new large Macs have a Retina display? Big, combined with tiny pixels-- that would be perfect for this.

[ndg]

Hi Nana Dadzie,
A few words on the progress of my project.
I bought a 19" Artar red dot lens and build a primitive camera mounting such lens.
A computer screen was projected (1 to 1) on a radiographic Fuji green film and developed with HC110 1:50 for 10 minutes.
I could not get a large optical density, even by extending exposure time.
By inspecting the negative with a magnifier I could clearly detect the single pixels, and I understood why I was limited in o.d.
Since the film is sensitive only to green (and perhaps a little to blue, but not to red), only a part of the pixel impresses the film.
Actually, the transmission of the impressed film could range from 66%(only green) to 33%(green and blue), and even more, considering the inactive part of the pixel.
This means an o.d. of 0.5 to 0.18, a very poor result.
Such a situation is very bad for the direct production of a negative, and acceptable to produce an intermediate positive to contact print to get the desired negative.
To overcome this problem (I do not like to need two films) I plan to build a device which moves the image up down and left right to the extent of a pixel (about 0.2 mm) to impress homogenously the whole pixel with the green one.
If it works fine I will describe it in detail.
Does anybody know of a similar approach?

Salvatore, like I stated in an earlier post, you need a big screen with lots of brightness. Back then, I used a 42" 120Hz display NOT a computer monitor. Also, the digital image has to be very high res. I used images from the Nikon 800e.

[salvatore]

Really sounding like the wrong tool. Like when I need a hammer and only find my pipe wrench which is also capable of pounding in a staple because the stapler is not working.

May be, but the only alternative is to use an inkjet printer of good quality and good quality (and expensive) cartridges, to be used almost continuously in order to avoid ink drying and related problems.
If anyone likes the inkjet approach, neglect my way.
But I think could be anyhow nice to have different approaches to the problem of obtaining a negative from digital photograph with optic methods.

[salvatore]

Salvatore, like I stated in an earlier post, you need a big screen with lots of brightness. Back then, I used a 42" 120Hz display NOT a computer monitor. Also, the digital image has to be very high res. I used images from the Nikon 800e.

Hi Nana Dadzie, for my approach I do not need a large screen, but a high resolution screen, and my ones are all 1080x1920 (full HD). Moreover 4K screens are becoming common with a resolution of 2160x3840 pixels.
However, the problem I mentioned in my past message is always present, and intrinsically bound to the pixel composition (green,red, blue) and the ortochromatic characteristic of the radiological films.
Work is in progress.

[Tin Can]

Are you using a CRT or LCD? This film is designed to capture CRT screens. http://www.zzmedical.com/analog-x-ray-supplies/x-ray-film/kodak-x-ray-film/8x10-in-carestream-kodak-ektascan-b-ra-single-emulsion-video-film.html

I shoot my 48" LCD LED BL HD TV with DSLR and 35 mm film. I found results usable for some things.

LCD has no scan lines like a CRT capture which may appear, dependent upon shutter speed and timing. Different technology.

[salvatore]

Are you using a CRT or LCD? This film is designed to capture CRT screens. http://www.zzmedical.com/analog-x-ray-supplies/x-ray-film/kodak-x-ray-film/8x10-in-carestream-kodak-ektascan-b-ra-single-emulsion-video-film.html

I shoot my 48" LCD LED BL HD TV with DSLR and 35 mm film. I found results usable for some things.

LCD has no scan lines like a CRT capture which may appear, dependent upon shutter speed and timing. Different technology.

I am not using any CRT at the moment, only full HD screens LED LCD.
I also made some pictures on a computer screen with 1080x1920 pixel by a Nikkormt camera and 110 mm lens. Good negatives, which can be reasonably enlarged.
However at present I like to extract all possible from a large format lens camera.
I will let know my results when available.
Thanks to all for the attention and the comments.

[VictoriaPerelet]

I am not using any CRT at the moment, only full HD screens LED LCD.
I also made some pictures on a computer screen with 1080x1920 pixel by a Nikkormt camera and 110 mm lens. Good negatives, which can be reasonably enlarged.


This topic was discussed here some time ago. If you use modern macro lens to take pic of LCD screen
This is what you'll get:

http://victoriasphoto.com/Notes/MacScreen/screen_1x1.jpg

Enlarging this will produce rather interesting "pixelation":)

(crop from 6MP Pic of Apple LCD display using Canon 100mm - more here (http://www.largeformatphotography.info/forum/showthread.php?46876-scanning-with-dslr&p=449744&viewfull=1#post449744))


Another problem with using screen (lcd or crt) for contact prints is that you will include few mm of glass between focused image and receiver (film or photo paper). This will defocus your final image.

Victoria

[Henrim]

I was goin to sleep but instead decided to try this out quickly. Shot a 4x5 negative from MacBook Pro Retina display. As the screen and the negative have different aspect ratio I tried to frame the shot to fill as much of the screen as possible. I developed the neg to rather high contrast and cut out a piece to be scanned with Nikon ED 9000.

Sorry for all the dust but here's a 100% crop of the scan @4000DPI. Shows a strong canvas like pattern.

http://melaanvuo.com/share/led_screen_1.jpg

Reduced size crop still shows banding.

http://melaanvuo.com/share/led_screen_2.jpg

Just wanted to see what it looks like. Could possibly be used for as effect. I guess banding could be minimised with low contrast development, optimising the exposure and what not. Possibly no use for me so I may not continue experimenting.

[salvatore]

I was goin to sleep but instead decided to try this out quickly. Shot a 4x5 negative from MacBook Pro Retina display. As the screen and the negative have different aspect ratio I tried to frame the shot to fill as much of the screen as possible. I developed the neg to rather high contrast and cut out a piece to be scanned with Nikon ED 9000.

Sorry for all the dust but here's a 100% crop of the scan @4000DPI. Shows a strong canvas like pattern.

http://melaanvuo.com/share/led_screen_1.jpg

Reduced size crop still shows banding.

http://melaanvuo.com/share/led_screen_2.jpg

Just wanted to see what it looks like. Could possibly be used for as effect. I guess banding could be minimised with low contrast development, optimising the exposure and what not. Possibly no use for me so I may not continue experimenting.
Quite interesting result.
Apart from the possible interest in a "canvas effect", obviously this technique cannot resolve the image to details smaller than a pixel.
The best resolution you can expect is the one offered by common full HD screens, which however is not too bad, especially if the negative obtained is used for gum printing, with its intrinsic limited resolution.
Moreover the use of a 4K and higher screen will reduce the size of a pixel to the limiting resolution of the naked eye (about 0.1 mm).