Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Any advantages shooting astrophotgraphy with film over digital?

  • Please log in to reply
104 replies to this topic

#76 SandyHouTex

SandyHouTex

    Skylab

  • *****
  • Posts: 4127
  • Joined: 02 Jun 2009
  • Loc: Houston, Texas, USA

Posted 13 August 2018 - 01:48 PM

There are multiple reports that, as of a few days ago, Ektachrome Elite 100 in 135-36 rolls is in the hands of beta-testers.

That would be a big FINALLY.  Geez it's taking Kodak forever to get this stuff to market.


  • Michael Covington likes this

#77 Todd N

Todd N

    Mariner 2

  • *****
  • Posts: 236
  • Joined: 07 Dec 2007

Posted 18 August 2018 - 04:32 AM

There are multiple reports that, as of a few days ago, Ektachrome Elite 100 in 135-36 rolls is in the hands of beta-testers.

 

Why wait for Ektachrome? There is rebranded Kodak motion picture film called CineStill in ISO 50 and 800. Red sensitivity in both peak at around 650nm and both can be pushed. C41 compatible. I haven't heard of anyone trying it.

 

Todd



#78 Alen K

Alen K

    Apollo

  • -----
  • Posts: 1093
  • Joined: 25 Nov 2009

Posted 18 August 2018 - 07:27 AM

Why wait for Ektachrome? There is rebranded Kodak motion picture film called CineStill in ISO 50 and 800. Red sensitivity in both peak at around 650nm and both can be pushed. C41 compatible. I haven't heard of anyone trying it.

There might be a good reason for that. As motion picture film, I have to wonder what the minimum buy quantity is. Even if it is a single reel, that would be a lot of film. There's a reason they talk about movie images as "footage."
  • LkUpRusty! likes this

#79 Michael Covington

Michael Covington

    Author

  • *****
  • Vendors
  • Posts: 5523
  • Joined: 13 May 2014
  • Loc: Athens, Georgia, USA

Posted 18 August 2018 - 10:02 AM

There might be a good reason for that. As motion picture film, I have to wonder what the minimum buy quantity is. Even if it is a single reel, that would be a lot of film. There's a reason they talk about movie images as "footage."

He said "rebranded."  I assume that means someone is packaging it in 35-mm cartridges.  Worth looking at. 



#80 SandyHouTex

SandyHouTex

    Skylab

  • *****
  • Posts: 4127
  • Joined: 02 Jun 2009
  • Loc: Houston, Texas, USA

Posted 18 August 2018 - 10:29 AM

Why wait for Ektachrome? There is rebranded Kodak motion picture film called CineStill in ISO 50 and 800. Red sensitivity in both peak at around 650nm and both can be pushed. C41 compatible. I haven't heard of anyone trying it.

 

Todd

I checked and CineStill is a color negative film.  Ektachrome is a color reversal film (slide).

 

While I’m waiting for Kodak Ektachrome, I’m going to use Fuji Velvia.


Edited by SandyHouTex, 18 August 2018 - 10:30 AM.


#81 Alen K

Alen K

    Apollo

  • -----
  • Posts: 1093
  • Joined: 25 Nov 2009

Posted 18 August 2018 - 02:15 PM

He said "rebranded."  I assume that means someone is packaging it in 35-mm cartridges.  Worth looking at.

l missed the meaning of that. I forgot that film photography is one of the recent retro crazes, like vinyl records and tape. We'll see if there is enough of a market for that to survive past a few years. I wonder what the significance of the "unique halation effect" is on astrophotography?
  • Michael Covington likes this

#82 chriscorkill

chriscorkill

    Explorer 1

  • *****
  • Posts: 67
  • Joined: 02 Jan 2013
  • Loc: Italy

Posted 19 December 2018 - 12:58 PM

I've never taken film photography but I would imagine that advantages include that you can blow the picture up as large as you want it with film and also that you don't need to lug around massive batteries. Also you wouldn't have to worry about your storage media being degaussed or destroyed, with film it should be pretty easy to handle the negatives. Personally I prefer electronic photography but I would love to test out film to say I've done it.

 

Christopher


Edited by chriscorkill, 19 December 2018 - 01:00 PM.


#83 Michael Covington

Michael Covington

    Author

  • *****
  • Vendors
  • Posts: 5523
  • Joined: 13 May 2014
  • Loc: Athens, Georgia, USA

Posted 19 December 2018 - 02:39 PM

"You can blow the picture up as large as you want it" -- No.  Ever heard of grain?  Film is intrinsically less sharp than digital sensors (by quite a bit) and has random grain all over it.  There's a reason Ansel Adams used huge sheet-film cameras when he wanted to make big enlargements. You can't enlarge film more than about x10 and expect it to look decent.

"You don't need to lug around massive batteries" -- A film SLR has batteries in it, and/or heavier mechanical parts; total weight about the same as a DSLR.  The telescope mount, of course, requires the same batteries regardless of the type of camera.

I could say that instead of a computer, you need to "lug around" (or rather go back to) a darkroom with running water and electricity...  Some great landscape photographers had these in their vans, especially in the early days.

"Wouldn't have to worry about your storage media being degaussed or destroyed" -- not electrically, but by heat or chemicals.  Developed and processed film doesn't keep forever.  It's *possible* to process black-and-white film carefully so it will last more than a century, but color slides and prints fade in 10 to 50 years.

There's a reason we've gone digital.



#84 Michael Covington

Michael Covington

    Author

  • *****
  • Vendors
  • Posts: 5523
  • Joined: 13 May 2014
  • Loc: Athens, Georgia, USA

Posted 19 December 2018 - 02:41 PM

Also, you can back up digital images by copying a disk file.  To "back up" film images requires rephotographing them onto more film, and there's always some loss of detail and dynamic range.  Copying film images was a challenge!



#85 TxStars

TxStars

    Surveyor 1

  • *****
  • Posts: 1875
  • Joined: 01 Oct 2005
  • Loc: Lost In Space

Posted 20 December 2018 - 11:33 PM

I have been working on the slow process of scanning all my film images.

This way I can more easily view them and have prints made.

As well as making some Blue-Ray DVD storage files.



#86 SandyHouTex

SandyHouTex

    Skylab

  • *****
  • Posts: 4127
  • Joined: 02 Jun 2009
  • Loc: Houston, Texas, USA

Posted 22 December 2018 - 11:26 AM

"You can blow the picture up as large as you want it" -- No.  Ever heard of grain?  Film is intrinsically less sharp than digital sensors (by quite a bit) and has random grain all over it.  There's a reason Ansel Adams used huge sheet-film cameras when he wanted to make big enlargements. You can't enlarge film more than about x10 and expect it to look decent.

"You don't need to lug around massive batteries" -- A film SLR has batteries in it, and/or heavier mechanical parts; total weight about the same as a DSLR.  The telescope mount, of course, requires the same batteries regardless of the type of camera.

I could say that instead of a computer, you need to "lug around" (or rather go back to) a darkroom with running water and electricity...  Some great landscape photographers had these in their vans, especially in the early days.

"Wouldn't have to worry about your storage media being degaussed or destroyed" -- not electrically, but by heat or chemicals.  Developed and processed film doesn't keep forever.  It's *possible* to process black-and-white film carefully so it will last more than a century, but color slides and prints fade in 10 to 50 years.

There's a reason we've gone digital.

Current digital sensors have nowhere near the resolution of film.  In the 35mm format, to equal film, you would need a sensor with 175 megapixels.



#87 Michael Covington

Michael Covington

    Author

  • *****
  • Vendors
  • Posts: 5523
  • Joined: 13 May 2014
  • Loc: Athens, Georgia, USA

Posted 22 December 2018 - 01:46 PM

Current digital sensors have nowhere near the resolution of film.  In the 35mm format, to equal film, you would need a sensor with 175 megapixels.

How do you figure that?  And do you have any experimental support?  Also, be sure to take the MTF of film into account.  Film resolution depends on contrast.

What I actually want to see, though, are pictures that are sharper on film than on the same-size digital sensor.  In my experience it's the other way around, by about a factor of 5.


Edited by Michael Covington, 22 December 2018 - 01:47 PM.


#88 Todd N

Todd N

    Mariner 2

  • *****
  • Posts: 236
  • Joined: 07 Dec 2007

Posted 22 December 2018 - 06:12 PM

SandyHouTex, on 22 Dec 2018 - 08:26 AM, said:

    Current digital sensors have nowhere near the resolution of film.  In the 35mm format, to equal film, you would need a sensor with 175 megapixels.

How do you figure that?  And do you have any experimental support?  Also, be sure to take the MTF of film into account.  Film resolution depends on contrast.

What I actually want to see, though, are pictures that are sharper on film than on the same-size digital sensor.  In my experience it's the other way around, by about a factor of 5.

 

The one scenario I think where film might have better resolution than digital would be a very fine grain film/analog print vs. a digital image/ inkjet print. I'm not aware if some film scanners  these days have the resolution that can exceed that of the finer grain films. Scanning film can lower the resolution while an inkjet printing likewise. There is an obvious drop in quality with inkjet prints from the original digital image on a monitor.



#89 Michael Covington

Michael Covington

    Author

  • *****
  • Vendors
  • Posts: 5523
  • Joined: 13 May 2014
  • Loc: Athens, Georgia, USA

Posted 22 December 2018 - 06:20 PM

Film resolution tops out at about 125 lines/mm.  The Nyquist sampling theorem says that's equivalent to 250 pixels/mm, i.e., 4-micron pixels (which is what we have).  But there's a catch.  Film resolution is 125 lines/mm only if you are willing to accept very low-constrast rendering of high-contrast detail -- that is, blur.  The Nyquist theorem pertains to rendering with no loss of contrast.  The MTF of film is far worse than putative digital equivalents estimated simply from lines/mm.

I want to see actual images.  In fact I plan to do a side-by-side test of Ektachrome vs. a digital sensor some time this year and rephotograph the Ektachrome in a microscope so that the resolution of a scanner is not a factor.  My earlier experiments indicate that fine-grain film is about 5 times worse than present-day digital, i.e., it is the equivalent of 20-micron pixels, roughly.

 

BTW, there is an often cited comparison of a digital camera to 4x5 sheet film.  That is not what I have in mind.  I want a comparison where the film image and the image on the sensor are the same size -- in fact the same image with the same lens.


Edited by Michael Covington, 22 December 2018 - 06:20 PM.


#90 Alen K

Alen K

    Apollo

  • -----
  • Posts: 1093
  • Joined: 25 Nov 2009

Posted 23 December 2018 - 01:47 AM

Current digital sensors have nowhere near the resolution of film.  In the 35mm format, to equal film, you would need a sensor with 175 megapixels.

It's amazing. Every time I hear a number for the digital equivalent of 35mm film resolution, it goes up. This one is way higher than I have heard before. If this trend continues, boys and girls, we're going to hear that it requires an infinite number of pixels to equal 35mm film. Where that will leave medium and large format film I don't know. 


  • Michael Covington likes this

#91 Michael Covington

Michael Covington

    Author

  • *****
  • Vendors
  • Posts: 5523
  • Joined: 13 May 2014
  • Loc: Athens, Georgia, USA

Posted 23 December 2018 - 12:17 PM

It's amazing. Every time I hear a number for the digital equivalent of 35mm film resolution, it goes up. This one is way higher than I have heard before. If this trend continues, boys and girls, we're going to hear that it requires an infinite number of pixels to equal 35mm film. Where that will leave medium and large format film I don't know. 

Indeed!  I can remember when it was 8 megapixels or even less... and my own feeling is that 8 is about right.  And that we seldom actually view pictures with a total resolution greater than about 2 megapixels; higher resolution is for blowing up small areas.

Consider an 8x10 print.  Sharp inkjet prints have 75 to 150 dots per inch.  Say 150.  Then 8 x 150 x 10 x 150 = 1.8 megapixels.



#92 SandyHouTex

SandyHouTex

    Skylab

  • *****
  • Posts: 4127
  • Joined: 02 Jun 2009
  • Loc: Houston, Texas, USA

Posted 23 December 2018 - 01:45 PM

How do you figure that?  And do you have any experimental support?  Also, be sure to take the MTF of film into account.  Film resolution depends on contrast.

What I actually want to see, though, are pictures that are sharper on film than on the same-size digital sensor.  In my experience it's the other way around, by about a factor of 5.

It’s somewhere in kenrockwell.com.  I’ve linked it before.  If I have time I’ll try to find it.

 

Found it.  Here's the link:

 

https://kenrockwell....-resolution.htm

 

Go down to the "Digital Resolution of Film",  paragraph 8.

 

What a lot of people forget, is that for a digital sensor, each pixel only records a single color.  To determine what the digital photograph actually looks like, all the pixels have to be interpolated to come up with the actual colors.  Or in other words, you need at least three pixels to determine the color at a single point.  So technically, the resolution of a digital sensor that's 30 megapixels only resolves at a 10 megapixel level.


Edited by SandyHouTex, 23 December 2018 - 02:15 PM.


#93 Michael Covington

Michael Covington

    Author

  • *****
  • Vendors
  • Posts: 5523
  • Joined: 13 May 2014
  • Loc: Athens, Georgia, USA

Posted 23 December 2018 - 03:09 PM

I remember reading Ken Rockwell, and it looked to me like he was comparing 4x5 film to a small digital sensor.  Do I need to look back?

 

Also, as regards the Bayer matrix, it is true that chrominance information requires adjacent pixels, but luminance information largely does not.  

 

What I want to see, from the "film is better" crowd, is actual experimental demonstration with the *same* subject and *same* lens.



#94 Michael Covington

Michael Covington

    Author

  • *****
  • Vendors
  • Posts: 5523
  • Joined: 13 May 2014
  • Loc: Athens, Georgia, USA

Posted 23 December 2018 - 03:23 PM

This is a case of theory needing to be tested against actual practice.

(1) Velvia does not resolve 160 lp/mm in any practical use.  Look at its MTF curve.  The MTF at 160 lp/mm is not even plotted on the graph; it must be 2% or less, if you extrapolate from the curve as plotted.  At 160 lp/mm there is a barely discernible image.  He takes this and compares it to the Nyquist theorem, which tells you there is a full response with no loss of contrast at 160 lp/mm if you have 320 pixels/mm.  Those are simply not comparable.

The Velvia 50 data sheet is here:
http://www.fujifilm....0_datasheet.pdf

See the MTF curve, lower left corner of last page.

 

The 50% point on the MTF curve is about 45 lp/mm.  That's with an appreciable contrast reduction.  If you really want full contrast, you have to work at about 20 lp/mm.  Doing the math, those are equivalent to, respectively, 4.7 megapixels and 0.9 megapixel.

 

(2) Bayer interpolation is not as bad as is being assumed here.  The pixels are not simply binned 3 or 4 to 1.  Instead, think about "virtual pixels," the intersection points of R, G, and B in the matrix.  These are as numerous as real pixels (except at the edges).  But each of them is based on combined information from three colors. 

Admittedly, this is not as good as separate R, G, and B images would be, but I don't think it's as bad as 4-to-1 (2x2) binning either.  And even if it were, compared to the actual performance of film, 24-megapixel digital sensors would still win.

The bottom line is that if film were really that much better, we'd see it in the pictures.  Where theory disagrees with observation, I go with observation.

A lot of the arguments against digital were cogent when digital cameras were 1 or 2 megapixels, but not today.



#95 SandyHouTex

SandyHouTex

    Skylab

  • *****
  • Posts: 4127
  • Joined: 02 Jun 2009
  • Loc: Houston, Texas, USA

Posted 23 December 2018 - 10:31 PM

I remember reading Ken Rockwell, and it looked to me like he was comparing 4x5 film to a small digital sensor.  Do I need to look back?

 

Also, as regards the Bayer matrix, it is true that chrominance information requires adjacent pixels, but luminance information largely does not.  

 

What I want to see, from the "film is better" crowd, is actual experimental demonstration with the *same* subject and *same* lens.

The resolution of film is size independent.  Velvia 50 does about 160 lines per millimeter whether it’s 35mm, 4X5, 6X7, etc.  Digital depends on pixel size.



#96 SandyHouTex

SandyHouTex

    Skylab

  • *****
  • Posts: 4127
  • Joined: 02 Jun 2009
  • Loc: Houston, Texas, USA

Posted 23 December 2018 - 10:36 PM

This is a case of theory needing to be tested against actual practice.

(1) Velvia does not resolve 160 lp/mm in any practical use.  Look at its MTF curve.  The MTF at 160 lp/mm is not even plotted on the graph; it must be 2% or less, if you extrapolate from the curve as plotted.  At 160 lp/mm there is a barely discernible image.  He takes this and compares it to the Nyquist theorem, which tells you there is a full response with no loss of contrast at 160 lp/mm if you have 320 pixels/mm.  Those are simply not comparable.

The Velvia 50 data sheet is here:
http://www.fujifilm....0_datasheet.pdf

See the MTF curve, lower left corner of last page.

 

The 50% point on the MTF curve is about 45 lp/mm.  That's with an appreciable contrast reduction.  If you really want full contrast, you have to work at about 20 lp/mm.  Doing the math, those are equivalent to, respectively, 4.7 megapixels and 0.9 megapixel.

 

(2) Bayer interpolation is not as bad as is being assumed here.  The pixels are not simply binned 3 or 4 to 1.  Instead, think about "virtual pixels," the intersection points of R, G, and B in the matrix.  These are as numerous as real pixels (except at the edges).  But each of them is based on combined information from three colors. 

Admittedly, this is not as good as separate R, G, and B images would be, but I don't think it's as bad as 4-to-1 (2x2) binning either.  And even if it were, compared to the actual performance of film, 24-megapixel digital sensors would still win.

The bottom line is that if film were really that much better, we'd see it in the pictures.  Where theory disagrees with observation, I go with observation.

A lot of the arguments against digital were cogent when digital cameras were 1 or 2 megapixels, but not today.

Actually some of the arguments against digital do still apply.  And MTF curves are actual test results, not theory.



#97 Michael Covington

Michael Covington

    Author

  • *****
  • Vendors
  • Posts: 5523
  • Joined: 13 May 2014
  • Loc: Athens, Georgia, USA

Posted 23 December 2018 - 11:21 PM

All I can say is, show me the pictures.  We need comparisons of the same subject photographed with the same lens on film and with a digital sensor.



#98 TxStars

TxStars

    Surveyor 1

  • *****
  • Posts: 1875
  • Joined: 01 Oct 2005
  • Loc: Lost In Space

Posted 11 January 2019 - 08:10 PM

Film may be a thing of the past, but without it there would be no visual record of past events

 

Here is a quick slide dupe of three shots of Mars in 1988..

 

 post-12231-0-33489900-1547254234.jpg

 

Show me your digital from 1988 ??  lol


Edited by TxStars, 11 January 2019 - 08:11 PM.

  • Joe F Gafford, SandyHouTex and Miguelo like this

#99 Joe F Gafford

Joe F Gafford

    Soyuz

  • ****-
  • Posts: 3958
  • Joined: 15 Dec 2006
  • Loc: Denver, Colorado, US

Posted 13 January 2019 - 06:39 PM

TxStars, good images from your old slide copier. I do have images from 1988. Also 1988 was the year that Kodak brought out the Gold series in their print film on 35mm format. The Gold-1 and 2 emulsions was available until 2000. These were good emulsions until the Gold 400-3 came out. Then the red response dropped before the Ha line. I was ticked when that happened. I went to TSP that year and found a poor response when I came back and developed the negatives. I did do a test at our new DSS and found the same with fresh developer. The response curves were then published from Kodak showing the bad news. Kodak's reason was the flesh tones were too red, especially around sunset from customer complaints. 

 

Image is from 1988 negatives. The VR series negatives is at left. This was old at the time and the VR Gold 400 series were in as transition. The first try at the Gold 400-1 is at right with the 2 different night exposures with the rest on the leftover 12 exposure roll of VR film. Images taken over 2 nights 6 days apart. Note the background color differences of the negatives. Same developer batch. Click to enlarge

P1136201_web.jpg

 

Sadr region taken on 8-13-1988 9:36UT 10" F/4.5 Newtonian. Kodak VR Gold-400 film. 60 minute exposure. Kenosha Pass, CO.

08_21_1988001_Ab_smWeb.jpg

 

Joe

 

Edit: The single spike is from the single bar used as Meade's secondary holder. The smaller spike is probably from the Lumicon OAG pick-up prism edge. 


Edited by Joe F Gafford, 13 January 2019 - 06:47 PM.

  • SandyHouTex and Miguelo like this

#100 Joe F Gafford

Joe F Gafford

    Soyuz

  • ****-
  • Posts: 3958
  • Joined: 15 Dec 2006
  • Loc: Denver, Colorado, US

Posted 13 January 2019 - 09:28 PM

Also 1988 was the year I got a Mamiya RZ-67 medium format camera with a 110mm lens, Lumicon off axis guider, field flattener and a low profile non-rotating helical focuser for the 10" F/4.5 Newtonian. Here's my planetary tries with the negatives in the prior post. The first try with the Kodak Royal Gold-1 and a Mars 6 days later with the leftover roll of VR-400 film.

 

Jupiter with a hint of the GRS in the lower band, Mars 6 days apart. New eyepiece projection adapter setup for my Olympus OM-1 with a 25mm Kellner. 

Planet-Coll-Aa.jpg

 

The first test with the new optical setup and the Royal Gold 400-1 film. 10-16-1988 9:01 UT, 20 minutes. My backyard. Click to enlarge.

_10_30_1988D003Ac_DBE.jpg

 

Joe


  • SandyHouTex and tuc like this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics