V
...looks like our film Camera is here to stay .....
- Thread starter sulhan
- Start date
- Status
It's different. While a 1400dpi scan of 35mm film has no "visible" grain, the image isn't as clean as the output from a good DSLR. So when you interpolate the thing to 20 x 30, it isn't going to be as good. You're probably better off scanning at 4000dpi, whereby the enlargement factor is smaller.
I do not know if you have seen large digital prints up close. I've seen both 35mm prints and digital prints, 10 x 15, 20 x 30, etc up close. While it's an unscientific observation, the film print has grain which breaks up detail. The digital one, which is cleaner, looks a lot better. And when you go up the ISO ladder, say, up to ISO 1600, film is going to fail, and a nice clean ISO 1600 file from say a D1H, EOS 1D, etc is going to beat it flat.
I shot dance events alongside a friend who used a D30 - I used Press 800 pushed a stop as well as Superia XTRA 1600 on 2 different occassions. The D30 output simply surpassed the pushed film, which shows grain even at 4R.
On another occassion, he shot the bridge at a wedding, at ISO 800. I took the RAW file, converted to 16-bit TIFF, interpolated to 10 x 15" @ 240dpi, Unsharp Masked the thing, then printed the file on an Epson Stylus Photo EX printer on normal glossy paper. Result is amazing, even up close. Confident of doing the same thing on film?
Being a film shooter, I was skeptical like you, until I saw the results for myself on so many occassions. Digital is better in most cases.
Obviously, you can't compare 35mm with 4x5, 6x6, 6x7, etc as they obviously hold more detail and better tonality, but at least for 35mm, digital has won film in most aspects.
Regards
CK
If you want to compare the medium, you have to use a hi-end scanner (drum scanner if possible) to minimize the artifact generated by the scanning equipment.
As far as I know, a film such as Provia 100, able to record up to 140 lp/mm. While the sensor of D1X is only 126 pixel/mm or 63 lp/mm. Just from that technical figure, we can tell that the Provia will hold more details.
The only limiting factor I can see, is the lens, which may not be able to resolve more than the Provia can hold. If the film is exposed with a lens which only able to resolve up to 60 lp/mm, then the result will not be different than digital.
In the real life, I ever saw a poster-size print from digital, yes, it is impressive. But seems as something lacking, similar to those audio CD and LP, the CD is clean without noise but flat, while the LP sound lively and natural but has noise with it.
Digital has since benefit from advance interpolation algorithm to beef up its resolution, I can't comment much about this, just quoting a submission criteria from a stock-photo agency. Clearly they prohibit interpolation for some reason:
i) If your images are to be considered for one of our paper catalogs or the duping program you must be able to provide us a minimum file size of 50MB. Hi-Res files must be non-interpolated, so don't use Genuine Fractals, PhotoShop bi-cubic interpolation, etc. to artificially bump up your scans, as the results are always very noticeable when used in large, color separations.
going by the gist that digital is evolving (and improving) i believe one day it will definitely surpass film as a more convenient and visually better (although the film purists will still keep arguing)
however, it could very well turn out to be what happened when the Compact Disc format first came out and replaced the Tape and vinyl. today we even have super CD, but vinyl is very much still alive and people who appreciate it swear that it sounds a lot better (esp oldies and songs that feature vocals prominently)
perhaps because when you have analog you introduce noise; artifacts, everything is not so clean and filtered (like comparing minidisc recordings to tape recordings) everything does seem a lot better.
in fact if we were built like robots the compatibility of our digital eyes would be close to or even 100% with that of the media it was looking at - thus achieveing what we call 'perfect' image seen with our eyes.
but because as humans, we have faillings - our eyesight, our hearing fail. and then, our brain is a fascinating thing - it takes in sight, sound and introduces other things into the mix - such as emotions, feelings, thereby clouding logic. when our eyes fail, the other 'eyes' inside (up in the head; down in the heart) see what we don't anymore, or least feeling or something like that.
so in a way, sometimes sharp to the point of sharp might not be the be all end all. when they say 0% error rate, you know even some the technology, futurist advocates have matured beyond the understanding that it does not mean perfect. when there is error - it is not a dead end. just like in life - when there is a mistake - we learn.
going the big circle back to film vs digital - digital will have a better visual quality going by discrete values and measurements, if not today, it will someday. but if the love of film is there, grain and all ... it will always exist for the people that love them and believe in their beauty.
Hi
is it my turn to inject my opinions yet? ;p
Originally i wanted to do another one of my long digital vs film 'thesis' but i'm getting a bit tired of that. Those who hang out here long enough know my stand - i'm a pro digital person. The D30 wielding friend that CK mentioned is me. We have done numerous real world comparisons, shot from ISO 100 to 1600, in a variety of situations.
Without going too much into the nitty gritty details (again! we need some kind of FAQ for this!), the clean grainless and noiseless files of digital clearly surpasses any major wins in resolution film has. By using our good old eyeballs, examining prints of different sizes, the digital output exhibits superior color reproduction, color accuracy in mixed lighting conditions without the distraction of grain. Grain in film when viewed up close turns details into muddy pools of slush, especially at high ISO, while the digital files retain detail just as well as at ISO 100.
Film does have more resolution (more on this later); however, how much of that is contained in the grain, and how much of the details can be resolved visibly without being masked by grain?
My own subjective judgement of image quality - the print has to resemble as closely as possible what the eye sees at the time of light capture. And in the real world, there is no grain.
You can either take my word for it, or perform the tests yourself. ANd pls, no consumer digital camera output - use a real 35mm based, digital SLR for your comparisons. Our tests uses no theoritical, mathematical formulas or MTF charts. Visual inspection of big prints at high magnifications and normal viewing distances shows that digital is at least as good as 35mm film, if not more so. Show those prints to layman in blind tests, and their conclusions do not differ.
With the advent of true 6 megapixel digtal SLRs, i'm afraid the game is over for 35mm film (where image quality is concerned).
http://www.luminous-landscape.com/d60-first.htm
I have a few high resolution D60 files with me right now (shot by another friend who owns a D60). I cannot disagree with that article's conclusion. If the D30 is good, the D60 brings quality to another higher level (caveat: with good lenses)
Now, let's move on to something more interesting.
Vadium brought on the topic film resolution in terms of lp / mm. (line pairs per mm). Since black and white TMAX film is mentioned, let's talk about black and white processing. Document films like Kodak Technical Pan, Agfa Copex Rapid AHU, Maco Orthopan 25 and several others, do indeed have resolution figures of 350 to 600 lp/mm according to the manufacturer's datasheet. BUT let us point out the caveats:
The given resolution is measured on negatives that have been developed to the extremely steep characteristic curve needed for halftone (black-white) reproduction in special developer chemicals.
Also, the resolution figures have been produced by contact printing a glass plate with finely etched black/white lines with the emulsion. For very fine structures an even finer pattern can be used, again on a glass plate, with a chromium layer in which the patterns are etched.
How many of us actually develop our own black and white film, let alone adhere to stick standards to achieve the highese possible image quality?
Secondly, the resolution of very fine detail is limited by optical aberrations and the limiting value of diffraction. A lens that is diffaction limited is so well corrected in its optical aberrations, that only the diffraction of light does influence the quality of the image.
Thirdly, the viewing medium. Laying even 200 lp/mm of image quality into the best photographic paper is at best a tough job for even the best printers. And the contrast of the paper - different papers are going to give you different results. So how do we translate that supposedly 150 lp/mm of TMAX 100 into a high quality print? To view slides as they are meant to be viewed, at close to quality given by figures on manufactuers' MTF charts, you need VERY good loupes, with well calibrated lightboxes.
A chain is as strong as its weakest link. Every step in the imaging chain degrades the reproduction of the original object.
The taking lens is the obvious candidate as optical aberrations do reduce the accurate reproduction of the object. Less wellknown are the accurate focus and the movement of the camera body. Then the film emulsion adds to the degrading of the image, not only by contrast reduction and grain pattern, but also scattering of light in the emulsion. Then we have the exposure and development, which also introduce effects of reduction.
Then we have the enlargement lens and the conrast of the paper. At every stage there are many factors to control.
What am i trying to point out here? The theoritical resolution power of film at even 150 lp/mm is highly overrated. With digital, assuming good photo taking techniques, the imaging chain is simplified simplified, from taking lens (diffraction limited) to imaging sensor (CCD or CMOS), you cut away the many variables inherent in film.
The outout of the CCD or CMOS IS the negative and print both at the same time. This is a unique concept - in film, you still have to worry about nudging the last ounce of image quality from a negative. Not so with digital, unless u're alluding to the idea of processing RAW files from digital cameras. They are not quite the same thing.
Taking into account all of the above, assuming similar techniques, the actual resolution of a digital SLR output may not be that far from the resolution of film. True, the ultimate resolution of film is going to be higher, but grain, no matter how fine, distracts from the overall apparent image quality. Just look at pictures with clear blue skies for an example.
to be cont'd....
is it my turn to inject my opinions yet? ;p
Originally i wanted to do another one of my long digital vs film 'thesis' but i'm getting a bit tired of that. Those who hang out here long enough know my stand - i'm a pro digital person. The D30 wielding friend that CK mentioned is me. We have done numerous real world comparisons, shot from ISO 100 to 1600, in a variety of situations.
Without going too much into the nitty gritty details (again! we need some kind of FAQ for this!), the clean grainless and noiseless files of digital clearly surpasses any major wins in resolution film has. By using our good old eyeballs, examining prints of different sizes, the digital output exhibits superior color reproduction, color accuracy in mixed lighting conditions without the distraction of grain. Grain in film when viewed up close turns details into muddy pools of slush, especially at high ISO, while the digital files retain detail just as well as at ISO 100.
Film does have more resolution (more on this later); however, how much of that is contained in the grain, and how much of the details can be resolved visibly without being masked by grain?
My own subjective judgement of image quality - the print has to resemble as closely as possible what the eye sees at the time of light capture. And in the real world, there is no grain.
You can either take my word for it, or perform the tests yourself. ANd pls, no consumer digital camera output - use a real 35mm based, digital SLR for your comparisons. Our tests uses no theoritical, mathematical formulas or MTF charts. Visual inspection of big prints at high magnifications and normal viewing distances shows that digital is at least as good as 35mm film, if not more so. Show those prints to layman in blind tests, and their conclusions do not differ.
With the advent of true 6 megapixel digtal SLRs, i'm afraid the game is over for 35mm film (where image quality is concerned).
http://www.luminous-landscape.com/d60-first.htm
I have a few high resolution D60 files with me right now (shot by another friend who owns a D60). I cannot disagree with that article's conclusion. If the D30 is good, the D60 brings quality to another higher level (caveat: with good lenses)
Now, let's move on to something more interesting.
Vadium brought on the topic film resolution in terms of lp / mm. (line pairs per mm). Since black and white TMAX film is mentioned, let's talk about black and white processing. Document films like Kodak Technical Pan, Agfa Copex Rapid AHU, Maco Orthopan 25 and several others, do indeed have resolution figures of 350 to 600 lp/mm according to the manufacturer's datasheet. BUT let us point out the caveats:
The given resolution is measured on negatives that have been developed to the extremely steep characteristic curve needed for halftone (black-white) reproduction in special developer chemicals.
Also, the resolution figures have been produced by contact printing a glass plate with finely etched black/white lines with the emulsion. For very fine structures an even finer pattern can be used, again on a glass plate, with a chromium layer in which the patterns are etched.
How many of us actually develop our own black and white film, let alone adhere to stick standards to achieve the highese possible image quality?
Secondly, the resolution of very fine detail is limited by optical aberrations and the limiting value of diffraction. A lens that is diffaction limited is so well corrected in its optical aberrations, that only the diffraction of light does influence the quality of the image.
Thirdly, the viewing medium. Laying even 200 lp/mm of image quality into the best photographic paper is at best a tough job for even the best printers. And the contrast of the paper - different papers are going to give you different results. So how do we translate that supposedly 150 lp/mm of TMAX 100 into a high quality print? To view slides as they are meant to be viewed, at close to quality given by figures on manufactuers' MTF charts, you need VERY good loupes, with well calibrated lightboxes.
A chain is as strong as its weakest link. Every step in the imaging chain degrades the reproduction of the original object.
The taking lens is the obvious candidate as optical aberrations do reduce the accurate reproduction of the object. Less wellknown are the accurate focus and the movement of the camera body. Then the film emulsion adds to the degrading of the image, not only by contrast reduction and grain pattern, but also scattering of light in the emulsion. Then we have the exposure and development, which also introduce effects of reduction.
Then we have the enlargement lens and the conrast of the paper. At every stage there are many factors to control.
What am i trying to point out here? The theoritical resolution power of film at even 150 lp/mm is highly overrated. With digital, assuming good photo taking techniques, the imaging chain is simplified simplified, from taking lens (diffraction limited) to imaging sensor (CCD or CMOS), you cut away the many variables inherent in film.
The outout of the CCD or CMOS IS the negative and print both at the same time. This is a unique concept - in film, you still have to worry about nudging the last ounce of image quality from a negative. Not so with digital, unless u're alluding to the idea of processing RAW files from digital cameras. They are not quite the same thing.
Taking into account all of the above, assuming similar techniques, the actual resolution of a digital SLR output may not be that far from the resolution of film. True, the ultimate resolution of film is going to be higher, but grain, no matter how fine, distracts from the overall apparent image quality. Just look at pictures with clear blue skies for an example.
to be cont'd....
Hi
plainsman brought on a very interesting point about emotional attachment to film. This is a very real issue, and one which i believe influences a large number of people away from the digital trend. I do have that attachment, since I still shoot film, but it has nothing to do with image quality.
My preference for wide angles ensures that at least a part of what i produce will still be on 35mm film. (However, i'm always thankful that the multiplier of DSLRs saves me money in getting a 300 f2.8 lens, a focal length and speed which i now enjoy with my current lens setup on a digital SLR).
More significantly, the manufactuer of my favourite camera system (no not Canon! ;P) is unlikely to come up with the digital equivalent of their cameras any time soon, one of which i recently acquired and am actively using now. And hence, in this aspect, i don't have a choice of the medium. (which is a touch unfortunate)
anyway like wat some have pointed out, all these are merely tools for image capture. Use the one that YOU LIKE!
plainsman brought on a very interesting point about emotional attachment to film. This is a very real issue, and one which i believe influences a large number of people away from the digital trend. I do have that attachment, since I still shoot film, but it has nothing to do with image quality.
My preference for wide angles ensures that at least a part of what i produce will still be on 35mm film. (However, i'm always thankful that the multiplier of DSLRs saves me money in getting a 300 f2.8 lens, a focal length and speed which i now enjoy with my current lens setup on a digital SLR).
More significantly, the manufactuer of my favourite camera system (no not Canon! ;P) is unlikely to come up with the digital equivalent of their cameras any time soon, one of which i recently acquired and am actively using now. And hence, in this aspect, i don't have a choice of the medium. (which is a touch unfortunate)
anyway like wat some have pointed out, all these are merely tools for image capture. Use the one that YOU LIKE!
V
Vadim
Guest
Red Dawn,
I heard about "game is over for 35mm film" several years ago from D30 users, but it appears that game is really over for D30 (sorry for sarcasm, but I couldn't miss that). I bet I'll hear the same old tune "game over for 35mm" in 10-15 years to come.
As I understood from your post - you are mostly concerned by grain. Well, thanks God, grain is the last thing I worry about, and I never had problems with smoothness of blue sky or any other issue raised by you. Believe me or not, I can obtain better results from a roll of 35mm Velvia or T-Max than any DSLR can do. May I suggest you to give film another try?
Any picture taken with digital camera, can be taken on film. Most pictures taken on film can be taken with digital camera (why most - see my previous posts). Will the film image be better or worse to digital one? I believe that only depends on person who is behind the camera.
The given resolution is measured on negatives that have been developed to the extremely steep characteristic curve needed for halftone (black-white) reproduction in special developer chemicals.
Not true. At least for Kodak tech publications. Kodak resolution tests comply with ISO 6328 that implies normal processing.
Secondly, the resolution of very fine detail is limited by optical aberrations and the limiting value of diffraction. A lens that is diffaction limited is so well corrected in its optical aberrations, that only the diffraction of light does influence the quality of the image.
What's your point? Diffraction limit at f/5.6 is 268 lp/mm, at smaller aperture it is even higher. Some Zeiss CF lenses do reach diffraction limit at f/5.6. Which means 250lp/mm is reachable on practice (although you rarely need that).
Thirdly, the viewing medium. Laying even 200 lp/mm of image quality into the best photographic paper is at best a tough job for even the best printers. And the contrast of the paper - different papers are going to give you different results. So how do we translate that supposedly 150 lp/mm of TMAX 100 into a high quality print? To view slides as they are meant to be viewed, at close to quality given by figures on manufactuers' MTF charts, you need VERY good loupes, with well calibrated lightboxes.
Well, you can extract 150lp/mm from film with drum scanner, if you really need that. Secondly, speaking about transparencies, any good projection lens offers adequate resolution.
What am i trying to point out here? The theoritical resolution power of film at even 150 lp/mm is highly overrated. With digital, assuming good photo taking techniques, the imaging chain is simplified simplified, from taking lens (diffraction limited) to imaging sensor (CCD or CMOS), you cut away the many variables inherent in film.
You missed one important thing about film. You can apply digital workflow to film as well. Scan. If worried about scanner lens - get the one without the lens - the drum scanner it is. Do digital manipulations; yes, including Genuine Fractals. Output.
Even with 35mm film, the good one: Velvia, Provia, T-Max and alike, with proper techniques, you'll get better results than from any DSLR. Believe me or not.
I heard about "game is over for 35mm film" several years ago from D30 users, but it appears that game is really over for D30 (sorry for sarcasm, but I couldn't miss that). I bet I'll hear the same old tune "game over for 35mm" in 10-15 years to come.
As I understood from your post - you are mostly concerned by grain. Well, thanks God, grain is the last thing I worry about, and I never had problems with smoothness of blue sky or any other issue raised by you. Believe me or not, I can obtain better results from a roll of 35mm Velvia or T-Max than any DSLR can do. May I suggest you to give film another try?
Any picture taken with digital camera, can be taken on film. Most pictures taken on film can be taken with digital camera (why most - see my previous posts). Will the film image be better or worse to digital one? I believe that only depends on person who is behind the camera.
The given resolution is measured on negatives that have been developed to the extremely steep characteristic curve needed for halftone (black-white) reproduction in special developer chemicals.
Not true. At least for Kodak tech publications. Kodak resolution tests comply with ISO 6328 that implies normal processing.
Secondly, the resolution of very fine detail is limited by optical aberrations and the limiting value of diffraction. A lens that is diffaction limited is so well corrected in its optical aberrations, that only the diffraction of light does influence the quality of the image.
What's your point? Diffraction limit at f/5.6 is 268 lp/mm, at smaller aperture it is even higher. Some Zeiss CF lenses do reach diffraction limit at f/5.6. Which means 250lp/mm is reachable on practice (although you rarely need that).
Thirdly, the viewing medium. Laying even 200 lp/mm of image quality into the best photographic paper is at best a tough job for even the best printers. And the contrast of the paper - different papers are going to give you different results. So how do we translate that supposedly 150 lp/mm of TMAX 100 into a high quality print? To view slides as they are meant to be viewed, at close to quality given by figures on manufactuers' MTF charts, you need VERY good loupes, with well calibrated lightboxes.
Well, you can extract 150lp/mm from film with drum scanner, if you really need that. Secondly, speaking about transparencies, any good projection lens offers adequate resolution.
What am i trying to point out here? The theoritical resolution power of film at even 150 lp/mm is highly overrated. With digital, assuming good photo taking techniques, the imaging chain is simplified simplified, from taking lens (diffraction limited) to imaging sensor (CCD or CMOS), you cut away the many variables inherent in film.
You missed one important thing about film. You can apply digital workflow to film as well. Scan. If worried about scanner lens - get the one without the lens - the drum scanner it is. Do digital manipulations; yes, including Genuine Fractals. Output.
Even with 35mm film, the good one: Velvia, Provia, T-Max and alike, with proper techniques, you'll get better results than from any DSLR. Believe me or not.
Time for a few facts here folks.
Vadim:
Your arguement is flawed, unsound and unsupportable regarding lpm and film resolution, diffraction limits and so on.
According to Kodak's own literature T-max 100 (which is being discontinued this year by the way) has the following Diffuse RMS Granularity and resolving powers.
RMS Granularity index = 8
TOC 1:6 - resolution is 63 lpm
TOC 1000:1 - resolution is 200 lpm
Kodak Professional source document
Now what this means is that at normal shooting conditions with a dynamic range of 6:1 the film resolves 63 lpm, not 150 as you dream about. 200 lpm is achieved using high contrast imaging with a dynamic ratio of 1000:1.
1000:1 This is typical of shooting test targets. Note: both are in accordance with ISO 6328 which I doubt you've read.
For further information on film resolution and diffraction limiting please refer to the following websites as it might help you to better understand why your arguement is fallacious.
Discussion by Patrick Bartek - includes data for most film emulsions
creekin's film resolution table (old and new film emulsions)
tsdh
Velvia is capable of just over 80 lpm under typical lighting conditions of 6:1 or thereabouts and 140 lpm when shooting high contrast images such as test targets with a dynamic range of 1000:1. Provia on the other hand cracks a mere 63 lpm at 6:1.
From memory the actual figure for Velvia is 83 lpm in the real world. The only emulsions to score higher than 100 lpm from Kodak, Fuji or Ilford are Kodak's Tech Pan and Ilford Pan F plus, both of which run out at 100+ lpm at 6:1 contrast levels. This matches very closely to best generally available B/W papers by the way.
Now that means that by your own figures the Nikon D1x which you cite as having a 126 lpm resolution has nearly double the real world 6:1 resolution of the majority of film emulsions.
Have a nice day.
Vadim:
Your arguement is flawed, unsound and unsupportable regarding lpm and film resolution, diffraction limits and so on.
According to Kodak's own literature T-max 100 (which is being discontinued this year by the way) has the following Diffuse RMS Granularity and resolving powers.
RMS Granularity index = 8
TOC 1:6 - resolution is 63 lpm
TOC 1000:1 - resolution is 200 lpm
Kodak Professional source document
Now what this means is that at normal shooting conditions with a dynamic range of 6:1 the film resolves 63 lpm, not 150 as you dream about. 200 lpm is achieved using high contrast imaging with a dynamic ratio of 1000:1.
1000:1 This is typical of shooting test targets. Note: both are in accordance with ISO 6328 which I doubt you've read.
For further information on film resolution and diffraction limiting please refer to the following websites as it might help you to better understand why your arguement is fallacious.
Discussion by Patrick Bartek - includes data for most film emulsions
creekin's film resolution table (old and new film emulsions)
tsdh
Velvia is capable of just over 80 lpm under typical lighting conditions of 6:1 or thereabouts and 140 lpm when shooting high contrast images such as test targets with a dynamic range of 1000:1. Provia on the other hand cracks a mere 63 lpm at 6:1.
From memory the actual figure for Velvia is 83 lpm in the real world. The only emulsions to score higher than 100 lpm from Kodak, Fuji or Ilford are Kodak's Tech Pan and Ilford Pan F plus, both of which run out at 100+ lpm at 6:1 contrast levels. This matches very closely to best generally available B/W papers by the way.
Now that means that by your own figures the Nikon D1x which you cite as having a 126 lpm resolution has nearly double the real world 6:1 resolution of the majority of film emulsions.
Have a nice day.
V
Vadim
Guest
Hey Red Dawn,
you forgot the other Luminous Landscape Article
http://www.luminous-landscape.com/dq.htm
That places a 35mm scan ahead of D60. Even on the same website, you have contributors with different points of view....
Of course, Norman Koren, who co-authored the above article, had this to say on his website:
"June 14, 2002 I didn't have to wait long for change; the pace is accelerating. The Fujifilm S2 Pro is on its way with US price set at $2399. The S2 Pro is an SLR with a Nikon lens mount and a 6.17 megapixel "Super CCD" that produces up to 12.1 Million (4256x2848) recorded pixels using on-chip interpolation. I don't know how to simulate it. DPReview.com has the URLs of two Japanese sites with sample images: Miscall and Yamada Kumio. It's well worth comparing the ASA 100 images from the EOS D60, D100, and S2 Pro on Kumio's site. Note particularly the block wall just below the clock. The S2 Pro is clearly superior; the images on both sites are amazing. The S2 Pro may well equal 35mm. "
you forgot the other Luminous Landscape Article
http://www.luminous-landscape.com/dq.htm
That places a 35mm scan ahead of D60. Even on the same website, you have contributors with different points of view....
Of course, Norman Koren, who co-authored the above article, had this to say on his website:
"June 14, 2002 I didn't have to wait long for change; the pace is accelerating. The Fujifilm S2 Pro is on its way with US price set at $2399. The S2 Pro is an SLR with a Nikon lens mount and a 6.17 megapixel "Super CCD" that produces up to 12.1 Million (4256x2848) recorded pixels using on-chip interpolation. I don't know how to simulate it. DPReview.com has the URLs of two Japanese sites with sample images: Miscall and Yamada Kumio. It's well worth comparing the ASA 100 images from the EOS D60, D100, and S2 Pro on Kumio's site. Note particularly the block wall just below the clock. The S2 Pro is clearly superior; the images on both sites are amazing. The S2 Pro may well equal 35mm. "
Yes, I know that. The figure of 140 lp/mm was taken from Fuji data-sheet, at contrast 1000:1
I agree that this kind of high contrast is only happen for testing purposes, not in the real world. But it did show the ability of film to record such details.
As RedDawn mentioned, in daily usage, we rarely (or even never) achieve that resolution because of other factors involved. That's true. My point is: at its maximum condition, film can hold more details than digital. But we rarely can come close to maximum.
It is similar as: I'm driving a Toyota, its max speed 180km/h, and my friend is driving a BMW, max speed 250 km/h. But the fastest both of us can drive is only 120 km/h on the highway due to limiting factors such as traffic, speed limit, etc. So at the end both cars show the same speed performance.
Ian, you misread my post. I wrote; D1X sensor has 126 pixel/mm, that's equal to 63 lp/mm.
regards.
Okay guys. I've seen many, many, many threads in my time. This one takes the cake. Prizes for reaching the end as always.
Here we go again, in chronological order:
[1] Tsdh, kudos to you. You're the first person in my long time in this business who has managed to completely leave my speechless by your posts. Well done, congratulations, and truce (cause yes, I really don't know what more to say)?
[2] Least of intention in getting my name printed in National Geographic thank you very much. I suppose when I was first starting out in photography it had a certain lure, but now after talking to more people and getting more into the industry, that glow has faded just a bit. I've got goals of my own, so I'll transfer your good wishes to them, thanks.
[3] Vadim, as has already been pointed out, let's try to be sure about what you're saying before you state it as fact. As others have already pointed out, the D1x is indeed sensitive to the infrared spectrum. As indeed most CCDs are, and require IR blocking filters to preserve normal image quality. Sure you can simply switch a roll of film. But what if you want that roll of film and you don't have it? By the time you account for every possibility, you'll be packing 50 different films into a portable fridge with a portable generator. Second question, no need to answer because the answer is obvious; do you carry around IR film with you everywhere you go?
[4] As to your resolution figures, I think enough has been said directly.
[5] I'm not following your argument about Photoshop being available before digital cameras. Certainly I'm aware of that fact. The point again I'm making is, what difference whether you use a specific film, a specific filter, or a specific Photoshop action to achieve your desired result? It shouldn't make any difference. A compositionally good picture taken on a digital camera and then Velvia-ed in Photoshop should be regarded as a better picture than a compositionally less good picture taken with a film camera loaded with Velvia. And two pictures with the same compositional strength, one taken with Velvia and one digital shot Velvia-ed later, should be regarded exactly the same way.
[6] Yes you can crop film to get the same benefits as a digital camera's focal length multiplier. But you also lose 40% of your total resolution, leaving you with only 60% of your film's capabilities. Now, given that it's already arguable that digital has overtaken film on a 1:1 level within the same formats, losing 40% of your film resolution is hardly desireable. I know you don't agree with the fact that digital is on 1:1, but others have already made this point, I'm making it now, and more on this later.
[7] Vadim says with current DSLRs you'll get better results with lenses at about 40lpmm to serve as an anti-aliasing filter. Here's the bad news; current digital cameras already have low pass anti-aliasing filters built in, so this would be double unsharpening of the picture. Not a good idea.
Now, and I speak from personal experience. Looking at my pictures now, I can tell the difference between each and every one of my lenses. I'm not saying that I can tell straightaway which lens took what, but I can easily see the difference between two lenses taking shots of the same subject in terms of what detail they record, and more importantly fail to record. This goes for primes and zooms, including between two prime lenses of similar class (a 300/2.8 and a 400/2.8). Now this suggests several things. First, it's clear that the camera records enough resolution to be able to distinguish the optical resolution of the lenses. Which in turn leads to the deduction that if this is the case, then IF film contains more resolution, it becomes a completely moot point as the lenses are starting to limit the image quality.
[8] Tsdh, as I said, I'm with you on the what tools you use to achieve the job shouldn't matter, only the result. It's the image that matters.
[9] You prefer a tool that is optimised for one goal rather than a generic tool for all things? Great. But I thought it's the image that matters? Incidentally the Swiss Army Knife is a very good bit of kit, not quite as good as the photographer's standard bit of Leatherman, but very good nonetheless). And while we're on the topic, no photographer worth his salt should ever be caught without a good multi-purpose tool. Or I suppose you can carry a full set of pliers, scissors, blades, saw, file, screwdrives etc if you prefer. (Tongue in cheek for the latter part, but the first part is good serious advice).
[10] Not all landscape photographers use ultrawide lenses. In fact, most serious landscape photographers use medium format and large format. Where there really isn't anything wider than 20mm equivalent anyway. They don't seem to complain. A lot of beginners (not saying that you are one) think that landscape photography is all about wide angle lenses. First they try to get everything into the frame, then they get beyond that and progress to including foreground interest. Then quickly they realise that ultrawides are not your workhorse landscape lenses at all. Many, many people far more into landscapes than me value the short telephoto above all else.
[11] Fair enough digital doesn't help you because you don't need the telephoto end. I never said it would. I said it's not necessarily a drawback, and that there are plus points to it. Things are not always absolute, just because it doesn't work for you doesn't mean it mightn't work for someone else, which is all I'm saying. So our two positions can co-exist.
[12] Yes digital sensors do get dust left on them. No it's not a major problem.
[13] Back to Vadim again. With film you have a lot more options. Yes. Agreed. If you want stylish grain you can push TMZ to EI 6400. If I want stylish grain I push sensitivity to EI 6400. Isn't that exactly the thing? Except that I will never be stuck in a situation where I might want stylish grain and not have the TMZ with me. Let's see... HIE, Konica Infrared, SFX all give very different responses. There's also EIR. And then there's PanF+ if you want fine grain, APX25 if you want a bit more actuance with that fine grain, Delta 100 if you want a really sharp picture with the modern feel, FP4+ if you want the conventional feel, Delta 400 if you want speed, HP5+ if you want that conventional grain feel, Tri-X if you want that taken to a greater extreme. *Deep breath* TMZ if you want that special grainy feel, Delta 3200 if you want that special grainy feel but want more contrast. Reala if I want realistic colour, Astia if I want realistic colour and slides. Sensia if I'm on a budget. Provia if I'm wanting fine grain without Velvia colours. Velvia if I want, erm, Velvia colours. Kodachrome if I want archivability. E100VS if I want Velvia with a bit more grain and a bit less polish. E100SW if I want really warped browns. Etc, etc. Now, the point is, I can approximate more than half of those "needs" fairly quickly and easily on my digital camera (I'm serious here). It won't be the exact same result, but it will be a similar result. And to me anyway it doesn't matter because it's the result that I want, not the exact HP5 look or whatever... but something similar. The rest I can probably also get fairly close to if I knew before hand (like you would need to to get the film and bring it along) by prepping custom curves.
[14] The point about colour spaces is that they are the camera's colour palette. A bit like Adobe RGB is closer to Astia as sRGB would be closer to Velvia.
[15] It's not such a simple thing as converting profiles around in Photoshop either. You can convert a scanned image into any colour space, sure, but you're missing the whole point of it. For starters, converting sRGB into Adobe RGB would be a complete waste of time, and if you need to ask why, then you obviously need to read up more. OTOH if you do know why, then your statement about converting colourspaces is very confusing.
[16] Interesting that you find sRGB crapy (sic). I assume you mean crappy?
[17] Vadim mentions that he can de-Velvia Velvia. Not really. If Velvia blocks up your shadow detail, then there's no way you can unblock it after the fact. Arguing that you can go from Astia to Velvia I would accept, but the other way around? Sorry. Sure you can use Astia instead of Velvia, but like I said before, it would mean carrying an endless supply of film around the place.
And what if one shot you want Astia, the next shot you want Velvia, the next shot you want Astia, the next you want HIE, the next you want TMZ at EI 6400? Enjoy using your proper film, and hope you don't miss any shots while you change films and do midroll rewinds.
(Cont'd)
Here we go again, in chronological order:
[1] Tsdh, kudos to you. You're the first person in my long time in this business who has managed to completely leave my speechless by your posts. Well done, congratulations, and truce (cause yes, I really don't know what more to say)?
[2] Least of intention in getting my name printed in National Geographic thank you very much. I suppose when I was first starting out in photography it had a certain lure, but now after talking to more people and getting more into the industry, that glow has faded just a bit. I've got goals of my own, so I'll transfer your good wishes to them, thanks.
[3] Vadim, as has already been pointed out, let's try to be sure about what you're saying before you state it as fact. As others have already pointed out, the D1x is indeed sensitive to the infrared spectrum. As indeed most CCDs are, and require IR blocking filters to preserve normal image quality. Sure you can simply switch a roll of film. But what if you want that roll of film and you don't have it? By the time you account for every possibility, you'll be packing 50 different films into a portable fridge with a portable generator. Second question, no need to answer because the answer is obvious; do you carry around IR film with you everywhere you go?
[4] As to your resolution figures, I think enough has been said directly.
[5] I'm not following your argument about Photoshop being available before digital cameras. Certainly I'm aware of that fact. The point again I'm making is, what difference whether you use a specific film, a specific filter, or a specific Photoshop action to achieve your desired result? It shouldn't make any difference. A compositionally good picture taken on a digital camera and then Velvia-ed in Photoshop should be regarded as a better picture than a compositionally less good picture taken with a film camera loaded with Velvia. And two pictures with the same compositional strength, one taken with Velvia and one digital shot Velvia-ed later, should be regarded exactly the same way.
[6] Yes you can crop film to get the same benefits as a digital camera's focal length multiplier. But you also lose 40% of your total resolution, leaving you with only 60% of your film's capabilities. Now, given that it's already arguable that digital has overtaken film on a 1:1 level within the same formats, losing 40% of your film resolution is hardly desireable. I know you don't agree with the fact that digital is on 1:1, but others have already made this point, I'm making it now, and more on this later.
[7] Vadim says with current DSLRs you'll get better results with lenses at about 40lpmm to serve as an anti-aliasing filter. Here's the bad news; current digital cameras already have low pass anti-aliasing filters built in, so this would be double unsharpening of the picture. Not a good idea.
Now, and I speak from personal experience. Looking at my pictures now, I can tell the difference between each and every one of my lenses. I'm not saying that I can tell straightaway which lens took what, but I can easily see the difference between two lenses taking shots of the same subject in terms of what detail they record, and more importantly fail to record. This goes for primes and zooms, including between two prime lenses of similar class (a 300/2.8 and a 400/2.8). Now this suggests several things. First, it's clear that the camera records enough resolution to be able to distinguish the optical resolution of the lenses. Which in turn leads to the deduction that if this is the case, then IF film contains more resolution, it becomes a completely moot point as the lenses are starting to limit the image quality.
[8] Tsdh, as I said, I'm with you on the what tools you use to achieve the job shouldn't matter, only the result. It's the image that matters.
[9] You prefer a tool that is optimised for one goal rather than a generic tool for all things? Great. But I thought it's the image that matters? Incidentally the Swiss Army Knife is a very good bit of kit, not quite as good as the photographer's standard bit of Leatherman, but very good nonetheless). And while we're on the topic, no photographer worth his salt should ever be caught without a good multi-purpose tool. Or I suppose you can carry a full set of pliers, scissors, blades, saw, file, screwdrives etc if you prefer. (Tongue in cheek for the latter part, but the first part is good serious advice).
[10] Not all landscape photographers use ultrawide lenses. In fact, most serious landscape photographers use medium format and large format. Where there really isn't anything wider than 20mm equivalent anyway. They don't seem to complain. A lot of beginners (not saying that you are one) think that landscape photography is all about wide angle lenses. First they try to get everything into the frame, then they get beyond that and progress to including foreground interest. Then quickly they realise that ultrawides are not your workhorse landscape lenses at all. Many, many people far more into landscapes than me value the short telephoto above all else.
[11] Fair enough digital doesn't help you because you don't need the telephoto end. I never said it would. I said it's not necessarily a drawback, and that there are plus points to it. Things are not always absolute, just because it doesn't work for you doesn't mean it mightn't work for someone else, which is all I'm saying. So our two positions can co-exist.
[12] Yes digital sensors do get dust left on them. No it's not a major problem.
[13] Back to Vadim again. With film you have a lot more options. Yes. Agreed. If you want stylish grain you can push TMZ to EI 6400. If I want stylish grain I push sensitivity to EI 6400. Isn't that exactly the thing? Except that I will never be stuck in a situation where I might want stylish grain and not have the TMZ with me. Let's see... HIE, Konica Infrared, SFX all give very different responses. There's also EIR. And then there's PanF+ if you want fine grain, APX25 if you want a bit more actuance with that fine grain, Delta 100 if you want a really sharp picture with the modern feel, FP4+ if you want the conventional feel, Delta 400 if you want speed, HP5+ if you want that conventional grain feel, Tri-X if you want that taken to a greater extreme. *Deep breath* TMZ if you want that special grainy feel, Delta 3200 if you want that special grainy feel but want more contrast. Reala if I want realistic colour, Astia if I want realistic colour and slides. Sensia if I'm on a budget. Provia if I'm wanting fine grain without Velvia colours. Velvia if I want, erm, Velvia colours. Kodachrome if I want archivability. E100VS if I want Velvia with a bit more grain and a bit less polish. E100SW if I want really warped browns. Etc, etc. Now, the point is, I can approximate more than half of those "needs" fairly quickly and easily on my digital camera (I'm serious here). It won't be the exact same result, but it will be a similar result. And to me anyway it doesn't matter because it's the result that I want, not the exact HP5 look or whatever... but something similar. The rest I can probably also get fairly close to if I knew before hand (like you would need to to get the film and bring it along) by prepping custom curves.
[14] The point about colour spaces is that they are the camera's colour palette. A bit like Adobe RGB is closer to Astia as sRGB would be closer to Velvia.
[15] It's not such a simple thing as converting profiles around in Photoshop either. You can convert a scanned image into any colour space, sure, but you're missing the whole point of it. For starters, converting sRGB into Adobe RGB would be a complete waste of time, and if you need to ask why, then you obviously need to read up more. OTOH if you do know why, then your statement about converting colourspaces is very confusing.
[16] Interesting that you find sRGB crapy (sic). I assume you mean crappy?
[17] Vadim mentions that he can de-Velvia Velvia. Not really. If Velvia blocks up your shadow detail, then there's no way you can unblock it after the fact. Arguing that you can go from Astia to Velvia I would accept, but the other way around? Sorry. Sure you can use Astia instead of Velvia, but like I said before, it would mean carrying an endless supply of film around the place.
And what if one shot you want Astia, the next shot you want Velvia, the next shot you want Astia, the next you want HIE, the next you want TMZ at EI 6400? Enjoy using your proper film, and hope you don't miss any shots while you change films and do midroll rewinds.
(Cont'd)
Yes, the digital sensor limits you. But it also frees you up to do the alterations that you can do (possibly not as flexible as film) quickly and freely after any shot. Don't belive that you mind need that diverse combination of film in a short space of time? IF (I stress the if) so then you obviously you don't shoot enough. So in a sense, the film also limits you. Yes you can have the "proper" film for one shot, but you might not for the next 35.
[18] Sigh. Unsharp lenses are there for a reason and that's because of the enlarging principles behind digital work. You have to remember also that these lenses are for large format work. Firstly, the average large format lens also has a lower resolving power than a smaller format lens. Secondly, at those sizes it matters less.
[19] As to Schneider charging a lot, that's another argument altogether. Which would make this thread even more exciting. People lust for Zeiss 100/3.5 CFi or APO Symmar 110XL lenses because lust is a primal thing, not a rational thing. People lust after D100 or 1D digital cameras despite not needing them at all. As to the price, your guess is as good as mine. I've maintained for ages that Schneider is laughing all the way to the bank. Compare an R8 and an F5 for approximately the same price, since that's at least a vaguely tangible comparison. The R8 has about a tenth the features of an F5, at the same cost. Draw your own conclusions about the difference in price between their lenses.
The next issue is, do we need all this extra resolution? How big do you print? I print big, and have my work printed big. I'm happy enough. Maybe I'm stupid, but I'm also practical. I take pictures with what equipment I have, rather than worrying about an extra 10lpmm that a German lens might or might not offer me. And usually my equipment that might or might not have 10lpmm less also enables me to get a shot that I would have missed with a German camera.
[20] "Any audiophile will tell you that SACD (96kHz sampling rate) rules and CD sucks." I had to quote that, because it bears striking resemblance to the following line: "Any leicaphile will tell you that NCMP (japanese low lpmm) sucks and Leitz rules." That's just the kind of statement that typifies a cult following.
[21] Now, to this issue about resolution. You're an engineer, you know the numbers and the whats and whys and wherefores. I know a bit about them, enough to get me by, but clearly not as much as you. So that's your turf, you win on that front. But I'm a photographer, and I know my pictures and the whats and whys and wherefores. Throw me all the numbers in the world, at the end of the day, I'll take the 20"x30" from my camera hanging on my wall as opposed to the Velvia enlargement to the same size. No, I'm not making predictions, I have actually done this. While it might not convince you, in fact I would be really surprised if it did, this is my bottom line. Everyone (layman and photographer alike) I've showed comparisons to have picked out the print from the digital file as the print they think is better. As far as I'm concerned this is all that matters. And 20x30 is not the limit of what I have done with my files either. I have less than 3mp files done up to about 6m x 4m. Sure they fall apart when you look at them from a foot away. But trust me they look unbelievably good at normal viewing distances. So the question then is, when do you need more resolution? This mind is a smaller than 3mp file.
[22] Hang on, aliasing and colour moire are not the main problems of digital photography. And as already mentioned, yes those lenses work great with less than 40lpmm, but only with larger sensors because with smaller sensors the loss in resolving power would far outweight any potential aliasing and moire; which as I've already said is not a major problem in digital photography at the moment. Sure if you can eliminate it, by all means do, but not at the expense of resolution unless you have the resolution to do so. You seem competent enough with numbers, you should have thought this one out.
[23] A Hassie with several lenses costs as much as a D1x alone. Question. What Hassie? If you're going to take the most expensive Nikon camera then I'd compare it with the most expensive Hassie. And certainly several lenses is a stretch with any Hassie body. And that is a Hassie without any film, and a Hassie without any of the features found on the D1x. We go back to what I said about an R8. If you take the most expensive Hassie then it costs about the same as a D1x, both without lenses. But the D1x has far more features than the Hassie. So why is the Hassie so expensive?
[24] Yes the Hassie will keep it's value while the D1x won't. So? Cameras are bought to be used, not re-sold. More importantly, you will still be paying for film and processing in 2010, 2020, while you won't for the D1x. Part of the price to pay for any digital SLR is the built in assumption of wiping out film bills. You see that's the difference, for some people buying cameras is an all round experience. Some numbers mean more than others, like lpmm as opposed to fps for example. And resale value as opposed to 20000 frames shot in 2 months. Or whatever.
[25] Oh you do a lot of digital editing and prefer to stick to one profile? Remember that bit above where I was wondering if you knew much about colour spaces? Well I apologise, apparently you do. So you do know also then that switching from profile to profile is undesireable. Which also moots your point about freely switching from one profile to the other.
[26] (Being stuck with the wrong film) has (yet) to happen to (Vadim) yet. Well good for him. He would also pull process to reduce contrast - Velvia handles pull very well. Then color compensate if needed. Very very good. But reducing Velvia's contrast doesn't make it any more suitable for portraiture, it'd still suck. And if you pull and then compensate, doesn't that just completely moot the point about having a "proper" film and using it? If you're so willing to compromise on this issue, pulling and then compensating if necessary, why are you so unwilling to use a flexible image CCD?
[27] "If you output images to quality CRT monitor and RBG printer - than you'd better stick to Adobe 1998 or similar profiles (BruceRGB is better IMHO)." Actually. if you output images to quality CRT monitors and RGB prints then you're better off sticking with sRGB. Not unless you really understand colour profiling, and if you did then you wouldn't need this bit of advice. This advice only really applies to beginners, and in which case, they are better off sticking to sRGB, making it potentially damaging advice.
[28] Erm, now I'm lost. In the very next paragraph you say, (it's better sticking to Adobe or BruceRGB) because if you use a wide gamut profile then you are wasting a colour space... but... Adobe and Bruce RGB profiles ARE wide gamut profiles. And hence why I said if you don't know what's going on you're best off sticking to sRGB.
[29] Yes sRGB is a crap profile, and is very confining, but the whole point is that it works with crap CRTs and printers, and if you don't know your way around colour management. If they did, they wouldn't need your advice, and if they don't, you're giving the wrong advice. Stick to sRGB if you don't know what's going on.
[30] Scanning film at only 3mp to reduce the impact of grain and hence achieve a picture similar in quality to a 3mp digital file. Sorry, doesn't work. Don't ask me for some theory to explain this, but I've worked side by side with film and digital for more than 2 years now. And I mean work; I shoot more in a week than most people shoot in a year. And I don't mean happy snappers, I mean amateurs. In the same way that lab tests sometimes don't reflect reality, I'm telling you your calculations aren't all that reliable.
[31] Yes 6x7 and 5x4 film has more details than digital. Of course they do! I shoot 645 and 5x4 myself and I know what they're capable of doing. But the discussion was on 35mm film vs 35mm digital. If you want to compare 6x7 and 5x4 then compare a digital sensor of the same size please. For someone who can grasp Nyquist theorems and lines per millimetre, you're having surprising difficulty understanding the difference between 16mm x 24mm and 120mm x 100mm.
[32] I've already dealt with the reason for stock agents not accepting digital files elsewhere on the forum.
[33] Correction to Red Dawn. Digital files at high ISO do not retain the same amount of detail as at lower ISO. Sad but true.
[34] Red Dawn has touched nicely on the lpmm issue. Just to add that a lot of the figures you are throwing around are at the high end of the contrast scale. In the real world contrast is often a lot lower which drops the films resolving capabilities down tremendously. I believe Ian has touched on this later although I've yet to fully read his post.
(Cont'd)
[18] Sigh. Unsharp lenses are there for a reason and that's because of the enlarging principles behind digital work. You have to remember also that these lenses are for large format work. Firstly, the average large format lens also has a lower resolving power than a smaller format lens. Secondly, at those sizes it matters less.
[19] As to Schneider charging a lot, that's another argument altogether. Which would make this thread even more exciting. People lust for Zeiss 100/3.5 CFi or APO Symmar 110XL lenses because lust is a primal thing, not a rational thing. People lust after D100 or 1D digital cameras despite not needing them at all. As to the price, your guess is as good as mine. I've maintained for ages that Schneider is laughing all the way to the bank. Compare an R8 and an F5 for approximately the same price, since that's at least a vaguely tangible comparison. The R8 has about a tenth the features of an F5, at the same cost. Draw your own conclusions about the difference in price between their lenses.
The next issue is, do we need all this extra resolution? How big do you print? I print big, and have my work printed big. I'm happy enough. Maybe I'm stupid, but I'm also practical. I take pictures with what equipment I have, rather than worrying about an extra 10lpmm that a German lens might or might not offer me. And usually my equipment that might or might not have 10lpmm less also enables me to get a shot that I would have missed with a German camera.
[20] "Any audiophile will tell you that SACD (96kHz sampling rate) rules and CD sucks." I had to quote that, because it bears striking resemblance to the following line: "Any leicaphile will tell you that NCMP (japanese low lpmm) sucks and Leitz rules." That's just the kind of statement that typifies a cult following.
[21] Now, to this issue about resolution. You're an engineer, you know the numbers and the whats and whys and wherefores. I know a bit about them, enough to get me by, but clearly not as much as you. So that's your turf, you win on that front. But I'm a photographer, and I know my pictures and the whats and whys and wherefores. Throw me all the numbers in the world, at the end of the day, I'll take the 20"x30" from my camera hanging on my wall as opposed to the Velvia enlargement to the same size. No, I'm not making predictions, I have actually done this. While it might not convince you, in fact I would be really surprised if it did, this is my bottom line. Everyone (layman and photographer alike) I've showed comparisons to have picked out the print from the digital file as the print they think is better. As far as I'm concerned this is all that matters. And 20x30 is not the limit of what I have done with my files either. I have less than 3mp files done up to about 6m x 4m. Sure they fall apart when you look at them from a foot away. But trust me they look unbelievably good at normal viewing distances. So the question then is, when do you need more resolution? This mind is a smaller than 3mp file.
[22] Hang on, aliasing and colour moire are not the main problems of digital photography. And as already mentioned, yes those lenses work great with less than 40lpmm, but only with larger sensors because with smaller sensors the loss in resolving power would far outweight any potential aliasing and moire; which as I've already said is not a major problem in digital photography at the moment. Sure if you can eliminate it, by all means do, but not at the expense of resolution unless you have the resolution to do so. You seem competent enough with numbers, you should have thought this one out.
[23] A Hassie with several lenses costs as much as a D1x alone. Question. What Hassie? If you're going to take the most expensive Nikon camera then I'd compare it with the most expensive Hassie. And certainly several lenses is a stretch with any Hassie body. And that is a Hassie without any film, and a Hassie without any of the features found on the D1x. We go back to what I said about an R8. If you take the most expensive Hassie then it costs about the same as a D1x, both without lenses. But the D1x has far more features than the Hassie. So why is the Hassie so expensive?
[24] Yes the Hassie will keep it's value while the D1x won't. So? Cameras are bought to be used, not re-sold. More importantly, you will still be paying for film and processing in 2010, 2020, while you won't for the D1x. Part of the price to pay for any digital SLR is the built in assumption of wiping out film bills. You see that's the difference, for some people buying cameras is an all round experience. Some numbers mean more than others, like lpmm as opposed to fps for example. And resale value as opposed to 20000 frames shot in 2 months. Or whatever.
[25] Oh you do a lot of digital editing and prefer to stick to one profile? Remember that bit above where I was wondering if you knew much about colour spaces? Well I apologise, apparently you do. So you do know also then that switching from profile to profile is undesireable. Which also moots your point about freely switching from one profile to the other.
[26] (Being stuck with the wrong film) has (yet) to happen to (Vadim) yet. Well good for him. He would also pull process to reduce contrast - Velvia handles pull very well. Then color compensate if needed. Very very good. But reducing Velvia's contrast doesn't make it any more suitable for portraiture, it'd still suck. And if you pull and then compensate, doesn't that just completely moot the point about having a "proper" film and using it? If you're so willing to compromise on this issue, pulling and then compensating if necessary, why are you so unwilling to use a flexible image CCD?
[27] "If you output images to quality CRT monitor and RBG printer - than you'd better stick to Adobe 1998 or similar profiles (BruceRGB is better IMHO)." Actually. if you output images to quality CRT monitors and RGB prints then you're better off sticking with sRGB. Not unless you really understand colour profiling, and if you did then you wouldn't need this bit of advice. This advice only really applies to beginners, and in which case, they are better off sticking to sRGB, making it potentially damaging advice.
[28] Erm, now I'm lost. In the very next paragraph you say, (it's better sticking to Adobe or BruceRGB) because if you use a wide gamut profile then you are wasting a colour space... but... Adobe and Bruce RGB profiles ARE wide gamut profiles. And hence why I said if you don't know what's going on you're best off sticking to sRGB.
[29] Yes sRGB is a crap profile, and is very confining, but the whole point is that it works with crap CRTs and printers, and if you don't know your way around colour management. If they did, they wouldn't need your advice, and if they don't, you're giving the wrong advice. Stick to sRGB if you don't know what's going on.
[30] Scanning film at only 3mp to reduce the impact of grain and hence achieve a picture similar in quality to a 3mp digital file. Sorry, doesn't work. Don't ask me for some theory to explain this, but I've worked side by side with film and digital for more than 2 years now. And I mean work; I shoot more in a week than most people shoot in a year. And I don't mean happy snappers, I mean amateurs. In the same way that lab tests sometimes don't reflect reality, I'm telling you your calculations aren't all that reliable.
[31] Yes 6x7 and 5x4 film has more details than digital. Of course they do! I shoot 645 and 5x4 myself and I know what they're capable of doing. But the discussion was on 35mm film vs 35mm digital. If you want to compare 6x7 and 5x4 then compare a digital sensor of the same size please. For someone who can grasp Nyquist theorems and lines per millimetre, you're having surprising difficulty understanding the difference between 16mm x 24mm and 120mm x 100mm.
[32] I've already dealt with the reason for stock agents not accepting digital files elsewhere on the forum.
[33] Correction to Red Dawn. Digital files at high ISO do not retain the same amount of detail as at lower ISO. Sad but true.
[34] Red Dawn has touched nicely on the lpmm issue. Just to add that a lot of the figures you are throwing around are at the high end of the contrast scale. In the real world contrast is often a lot lower which drops the films resolving capabilities down tremendously. I believe Ian has touched on this later although I've yet to fully read his post.
(Cont'd)
[35] "Believe me or not, I can obtain better results from a roll of 35mm Velvia or T-Max than any DSLR can do." You can? Right, colour portraits then, let's go. Or actually, let's shoot pictures where we're looking for special grain effect? Actually, forget that silly jibe, but no, I don't believe you. Through this, erm, discussion you have refrained from stating at any point that you have actually made comparisons with real results, but are just sticking to the theory front. Seems curious that you would not have mentioned it as an additional arrow in your quiver. And hence, if you haven't much practical experience in the area, and to state unconditionally that you can outshoot a DSLR is highly dubious.
[36] Under another barrage of figures about diffraction limits... so I suppose Vadim would also be the kind of photographer (or is that statistician?) who would advocate never shooting at anything smaller than f5.6?
[37] "Any good projection lens offers adequate resolution." Don't you think it strange that camera lenses must have every available ounce of resolution, preferably be German, but when it comes to projection lenses...
Okay okay okay. If you've read everything till this point then you deserve a star. Heck, make it five *****s. I think the point is that neither camp is going to agree with the other camp, so let's let the matter rest shall we? Agree to disagree. As in, I'm not saying that you're wrong, so you don't have to take offence and reply, just like you're not saying we're wrong, so we won't have to take offence and reply.
I'm happy with the pictures I create, whether digital or film (yes, I still use both), and I suppose that's good enough at the end fo the day.
[36] Under another barrage of figures about diffraction limits... so I suppose Vadim would also be the kind of photographer (or is that statistician?) who would advocate never shooting at anything smaller than f5.6?
[37] "Any good projection lens offers adequate resolution." Don't you think it strange that camera lenses must have every available ounce of resolution, preferably be German, but when it comes to projection lenses...
Okay okay okay. If you've read everything till this point then you deserve a star. Heck, make it five *****s. I think the point is that neither camp is going to agree with the other camp, so let's let the matter rest shall we? Agree to disagree. As in, I'm not saying that you're wrong, so you don't have to take offence and reply, just like you're not saying we're wrong, so we won't have to take offence and reply.
I'm happy with the pictures I create, whether digital or film (yes, I still use both), and I suppose that's good enough at the end fo the day.
:what:
Glad that this topic has settled down into a more level-headed discussion.
There is LOTS of information contained here, most of which sorta flew over my head and passed me by. Good factual information coupled with personal experiences. This is what discussions should be like in forums.
;p
Glad that this topic has settled down into a more level-headed discussion.
There is LOTS of information contained here, most of which sorta flew over my head and passed me by. Good factual information coupled with personal experiences. This is what discussions should be like in forums.
;pJed,
phew.... if anything, that latest 37-point post was the one that floored me. i will probably read it bit by bit and try to digest the info at some time, but yeah, i think its good we reach this point where we realise irregardless of image quality, convenience etc. factors film and digital does have its interest and people to pursue it.
have a good day all
phew.... if anything, that latest 37-point post was the one that floored me. i will probably read it bit by bit and try to digest the info at some time, but yeah, i think its good we reach this point where we realise irregardless of image quality, convenience etc. factors film and digital does have its interest and people to pursue it.
have a good day all
Originally posted by Goondu
Hey, irrespective, not irregardless.
This must have been the most commonly mis-used double-negative word around.
Just "regardless" will do.
Regards
CK
Originally posted by ckiang
This must have been the most commonly mis-used double-negative word around.
Just "regardless" will do.
Regards
CK
Sorry, way OT here, but got to say this.
'regardless of' is also wrong in this case. We can have 'regardless what....', 'regardless who....' but not 'regardless of....'.
However, 'irrespective of....' could be used in this case.
Just my 90 rupiah worth of thoughts.
hey thanks for enlightening, even if it was OT .... looks like im even more goondu than goondu!!!! :what: :bsmilie:
- Status
Similar threads
