Megapixel to film equation - DPChallenge Forums

01/18/2006 03:55:37 AM · #1

info

Best site I've found so far. And please don't post any of those skewed luminous lanscape links. I hate those.

01/18/2006 04:13:30 AM · #2

Originally posted by Brent_Ward:

info

Best site I've found so far. And please don't post any of those skewed l luminous landscapes links. I hate those.

The information is fuzzy at best. Not only that but he bases his assumptions on images taken with a D60 shot in JPG at 1.6 times the distance of the shots taken with the film camera. I'd say about as reliable as Luminous Landscapes or Ken Rockwell.

01/18/2006 04:24:33 AM · #3

Originally posted by nsbca7:

Originally posted by Brent_Ward:
info

Best site I've found so far. And please don't post any of those skewed l luminous landscapes links. I hate those.

The information is fuzzy at best. Not only that but he bases his assumptions on images taken with a D60 shot in JPG at 1.6 times the distance of the shots taken with the film camera. I'd say about as reliable as Luminous Landscapes or Ken Rockwell.

Looks like that was just one of the tests. I'm more interested in the equations, not his tests. his number of 10 MP = 35mm in detail is not far off.

Message edited by author 2006-01-18 04:25:25.

01/18/2006 04:26:23 AM · #4

Originally posted by Brent_Ward:

Originally posted by nsbca7:

Originally posted by Brent_Ward:
info

Best site I've found so far. And please don't post any of those skewed l luminous landscapes links. I hate those.

The information is fuzzy at best. Not only that but he bases his assumptions on images taken with a D60 shot in JPG at 1.6 times the distance of the shots taken with the film camera. I'd say about as reliable as Luminous Landscapes or Ken Rockwell.

Looks like that was just one of the tests. I'm more interested in the equations, not his tests.

Where do you think he derives his equations from? If that is the sloppy manner in which he gathers evidence, what could his equations be worth?

01/18/2006 04:34:00 AM · #5

Originally posted by nsbca7:

Originally posted by Brent_Ward:

Originally posted by nsbca7:

Originally posted by Brent_Ward:
info

Best site I've found so far. And please don't post any of those skewed l luminous landscapes links. I hate those.

The information is fuzzy at best. Not only that but he bases his assumptions on images taken with a D60 shot in JPG at 1.6 times the distance of the shots taken with the film camera. I'd say about as reliable as Luminous Landscapes or Ken Rockwell.

Looks like that was just one of the tests. I'm more interested in the equations, not his tests.

Where do you think he derives his equations from? If that is the sloppy manner in which he gathers evidence, what could his equations be worth?

The only test I saw with a d60 was an outdated one from 2002. The other tests don't seem to be based on D60 trials. He goes indepth more than any other site I've seen. His calculations seem pretty on.

01/18/2006 04:50:39 AM · #6

Originally posted by Brent_Ward:

The only test I saw with a d60 was an outdated one from 2002. The other tests don't seem to be based on D60 trials. He goes indepth more than any other site I've seen. His calculations seem pretty on.

Perhaps in theory because that is all I see presented, theory. Give me a link to these other image tests taken with modern digital cameras shot in RAW. I failed to find any.

I shot 35mm and medium format using Velvia, Provia, T-Max and Kodachrome and I can tell you that in theory perhaps his calculations may have some relevance to reality, but in actual usage pushing the formats to their maximum prints size there is no 35mm or medium format film that will enlarge better then today's top line 35mm digital cameras. In fact I'll back date that three years.

Message edited by author 2006-01-18 04:51:18.

01/18/2006 05:04:54 AM · #7

Originally posted by nsbca7:

Originally posted by Brent_Ward:

The only test I saw with a d60 was an outdated one from 2002. The other tests don't seem to be based on D60 trials. He goes indepth more than any other site I've seen. His calculations seem pretty on.

Perhaps in theory because that is all I see presented, theory. Give me a link to these other image tests taken with modern digital cameras shot in RAW. I failed to find any.

I shot 35mm and medium format using Velvia, Provia, T-Max and Kodachrome and I can tell you that in theory perhaps his calculations may have some relevance to reality, but in actual usage pushing the formats to their maximum prints size there is no 35mm or medium format film that will enlarge better then today's top line 35mm digital cameras. In fact I'll back date that three years.

I'll put a 40x50 from one of my scans up to your 1Ds Mark II any day. If you print below 20x30, you might have me.

My old 1ds files res'd upto 70mb are pretty much equal to my 35mm scans @ 70 mb.

01/18/2006 05:09:25 AM · #8

Originally posted by Brent_Ward:

Originally posted by nsbca7:

Originally posted by Brent_Ward:

The only test I saw with a d60 was an outdated one from 2002. The other tests don't seem to be based on D60 trials. He goes indepth more than any other site I've seen. His calculations seem pretty on.

Perhaps in theory because that is all I see presented, theory. Give me a link to these other image tests taken with modern digital cameras shot in RAW. I failed to find any.

I shot 35mm and medium format using Velvia, Provia, T-Max and Kodachrome and I can tell you that in theory perhaps his calculations may have some relevance to reality, but in actual usage pushing the formats to their maximum prints size there is no 35mm or medium format film that will enlarge better then today's top line 35mm digital cameras. In fact I'll back date that three years.

I'll put a 40x50 from one of my scans up to your 1Ds Mark II any day. If you print below 20x30, you might have me.

My old 1ds files res'd upto 70mb are pretty much equal to my 35mm scans @ 70 mb.

Can you see grain on that 40x50? And how does printing at a smaller size put me at an advantage?

01/18/2006 05:13:51 AM · #9

Originally posted by nsbca7:

Originally posted by Brent_Ward:

Originally posted by nsbca7:

Originally posted by Brent_Ward:

The only test I saw with a d60 was an outdated one from 2002. The other tests don't seem to be based on D60 trials. He goes indepth more than any other site I've seen. His calculations seem pretty on.

Perhaps in theory because that is all I see presented, theory. Give me a link to these other image tests taken with modern digital cameras shot in RAW. I failed to find any.

I shot 35mm and medium format using Velvia, Provia, T-Max and Kodachrome and I can tell you that in theory perhaps his calculations may have some relevance to reality, but in actual usage pushing the formats to their maximum prints size there is no 35mm or medium format film that will enlarge better then today's top line 35mm digital cameras. In fact I'll back date that three years.

I'll put a 40x50 from one of my scans up to your 1Ds Mark II any day. If you print below 20x30, you might have me.

My old 1ds files res'd upto 70mb are pretty much equal to my 35mm scans @ 70 mb.

Can you see grain on that 40x50? And how does printing at a smaller size put me at an advantage?

A little yes, It's very pleasent. Smaller size you would have the advantage due to the lack of grain or noise. By the time you either res it up or the RIP interpolates it, you'll see either jaggies or noise. And if your using a canon WA you'll have soft edges to deal width. While my pentax glass is pretty sharp all across.

I'm to the point that I'll be selling off my pentax gear much like you did. For wildlife I'll use digital. For landscapes I'll use an 8x10 ( or 4x5) view camera.

01/18/2006 05:32:09 AM · #10

Originally posted by Brent_Ward:

It may seem very pleasant in your images, but in mine it is nonexistent. If I want grain I can simply add it.

I have three images in my gallery at this moment that are over 40 inches wide, two of them to 48 inches and all of them are crops. I've also had a couple of people from the University of South Alabama archives and photography departments in here with magnifying glasses looking for "grain" and "pixels". They seemed utterly amazed when they couldn't find any. BTW, the image they were inspecting was a cropped image 48" wide (framed to 54) taken with my 11Mp 1Ds.

There are nails on the tin roof of a building in this image that are actually smaller then the grain size of an image enlarged to the same size had I taken it with a fine grain film using my old 6x7.

Nikon AF-S NIKKOR 24-70mm f/2.8G ED

01/18/2006 05:57:40 AM · #11

The cropped image above.

A one inch square from the enlarged 48x17 inch image. This is a detail of the wall next to the Coca-Cola sign. The nails are visible under the paint.

It would literally take a 4x5 to reproduce this with film.

Camera data:
Canon EOS-1DS
Shooting Date/Time
3/7/2004 2:48:55 PM
Shooting Mode
Aperture-Priority AE
Tv( Shutter Speed )
1/1000
Av( Aperture Value )
5.6
Metering Mode
Evaluative
Exposure Compensation
0
ISO Speed
160
Lens
85.0 mm
Focal Length
85.0 mm

Message edited by author 2006-01-18 06:04:59.

Nuno

Nikon D700

01/18/2006 05:59:42 AM · #12

Well I must say that at first I was intreeged by this pixer versus film discussion. When I decided to do a small business regerding photograpg I searched the web looking for explanations like this. And I'm still not convinced with the high qualaty and detail of film. Specially 35mm.

A few days ago I was at a friend house and we where talking about wedding photos and I looked at his prints from his wedding. I could see the grain and softness in the images in all 5x7 prints. I was shoked. If I did a wedding and I got that qualaty to present to a client I would be ashame of myself. Even in more high dimmension prints I could not see any visible grain in my prints. A few month ago I photographed for a magazine an inndor motorcicle event. As I don't have fast lenses (except my 105mm f2.8) I did most of the photos at iso1600. I did some 30x20cm prints to see if the noise was too evidnet in prints, because I could see it in photoshop at 100% view. I was extremely pleased because I could'nt tell any major issues withought a magnifying glass.

And in that review why does the iso400 film goes downt to 4mp comparison???

I think that 35mm cameras and films can't evolute much more and any new good digital slr can reveal as much or better detail and sharpness than 35mm. And Top qualaty dslr are miles away better than 35mm film.

I always found litle wise to compare the gigantic megapixel files from scaned film and the much smaller files from even a high megapixel count top dslr. I think this is a fool comparison. One of this days I'll see someone digitalizing in high resolution a 5x7 print and telling how mugh more megapixels it have!

01/18/2006 06:21:24 AM · #13

BTW, this image was not sharpened in camera or in PS. Other then cropping and slight adjustments to Hue/Saturation the image is straight from a 1Ds set at default settings. This is the first digital image I had ever taken.

01/18/2006 11:01:27 AM · #14

Well I guess a picture is worth a thousand words, or in this case a thousand line graphs and equations.

01/18/2006 02:31:03 PM · #15

Originally posted by Nuno:

And in that review why does the iso400 film goes downt to 4mp comparison???

His is basically saying that 35mm ISO 400 film is equal to a 4mp camera in terms of resolution.

01/18/2006 02:33:41 PM · #16

Originally posted by nsbca7:

The cropped image above.

A one inch square from the enlarged 48x17 inch image. This is a detail of the wall next to the Coca-Cola sign. The nails are visible under the paint.

It would literally take a 4x5 to reproduce this with film.

Camera data:
Canon EOS-1DS
Shooting Date/Time
3/7/2004 2:48:55 PM
Shooting Mode
Aperture-Priority AE
Tv( Shutter Speed )
1/1000
Av( Aperture Value )
5.6
Metering Mode
Evaluative
Exposure Compensation
0
ISO Speed
160
Lens
85.0 mm
Focal Length
85.0 mm

I've done tests against a 1ds before, my 6x7 blew it away at large sizes. I'm not tlaking traditional printing, I'm talking about digitally printing both files. The scanned file was much better. IT looks like a 4x5 traditional print.

And in know way can you compare the detail in a 1Ds file with a 4x5 scan. There is no comparison.

01/18/2006 02:43:17 PM · #17

Originally posted by Brent_Ward:

I've done tests against a 1ds before, my 6x7 blew it away at large sizes.

The grain from a print enlarged from a 6x7 to the same extent as what I have done with this shot from the 1Ds would be about the same size as the nails in this picture.

petrakka

01/18/2006 02:50:56 PM · #18

To put this out there, I'm taking a class which we will print 10 40x50 (or 40x40) images throughout the course of the class. We are not allowed to use the 1ds that they have available (unless we are putting together a composite), We must shoot 6x6, 6x7 or 4x5, because quality is an issue. We are scanning of a Nikon coolscan 8000 and an Imacon.

01/18/2006 03:06:06 PM · #19

Originally posted by petrakka:

because quality is an issue.

If you can't get up close and personal quality with the 1Ds set at low ISO then perhaps it is because too many hands have been or that particular camera.

01/18/2006 03:18:38 PM · #20

Originally posted by nsbca7:

Originally posted by Brent_Ward:

I've done tests against a 1ds before, my 6x7 blew it away at large sizes.

The grain from a print enlarged from a 6x7 to the same extent as what I have done with this shot from the 1Ds would be about the same size as the nails in this picture.

Maybe on an enlarger, but with a digital print it's much, much better. Do I have to ftp you a 200mb scan to print out?

01/18/2006 03:34:24 PM · #21

Originally posted by Brent_Ward:

Maybe on an enlarger, but with a digital print it's much, much better. Do I have to ftp you a 200mb scan to print out?

It would probably be easier if you just blew the unsharpened image up to 48 inches in one direction and cropped out a one inch square like I did in a previous post.

01/18/2006 03:50:28 PM · #22

These are the only images I have online currently. Shot with NPS 160, so the grain is way more pronounced than my Velvia scans.

About 75% of orginal film area.
//www.brentwardphoto.com/pug.jpg

100% crop
//www.brentwardphoto.com/pugcrop.jpg

Message edited by Manic - please post thumbs or links, not large images.

petrakka

01/18/2006 03:58:15 PM · #23

Originally posted by nsbca7:

Originally posted by petrakka:
because quality is an issue.

If you can't get up close and personal quality with the 1Ds set at low ISO then perhaps it is because too many hands have been or that particular camera.

no I think it's because 11 mp on 40x50 isnt as good as 6x7. on 16x20, and maybe even 20x30....but when your goin that huge, I've read even the mamiya ZD and hasselblad H2D aren't quite as good as film....though I admit I haven't seen comparisons.

I don't care too much thuogh, the rest of the world won't know the difference.

anyways, for most occasions I think it's kind of a moot point, because it's not too often that you print 40x50. It's my opinion that at sizesl ike 11x14 or 16x20 sometimes a digital print can look better than scanned film.

01/18/2006 04:18:45 PM · #24

Originally posted by petrakka:

anyways, for most occasions I think it's kind of a moot point, because it's not too often that you print 40x50. It's my opinion that at sizesl ike 11x14 or 16x20 sometimes a digital print can look better than scanned film.

Perhaps it is not too often that you print to that size. I love to print big and have had no problems printing sharp images to 4 feet wide from the 1Ds and 1Ds Mk II.