...But...
Everyone who is saying what it is better than darkroom prints is just blind. Like really or for whatever reason under self illusion...
Yes indeedThen there are the hybrid possibilities, such as creating an inkjet negative and contact printing that onto silver gelatin paper (or alt processes). I agree with Richard Man about it being two different media, but I look at digital photography as being a sub-set of digital art as we begin to leave digital art's infancy. It will be interesting to look back at the development of photography and digital art during the first two decades of the 21st century...once enough years pass to get a decent view.
The technological advances in lenses due to computer aided design largely just apply to zoom lenses. Prime lenses, like those found on an enlarger, have seen very little, if any, improvements since the introduction of the computer. Their formulas are much simpler since you only have to correct at one focal length. This is why most macro photographers who are serious about quality still use enlarging lenses attached to bellows on their digital cameras. Even the best modern macro lenses made to mount directly on to digital cameras fall short of good enlarging lenses. The reason is because most macro lenses are made with compromises so they will produce usable images at long focal distances, as well as short ones, whereas an enlarging lens is designed for short focal distances only.Thanks for your comments. Let me put aside questions of rarity or "value" and address some of the mechanistic points you raised.
I agree that the lens and sensor used in the scan will all degrade image quality. On the other hand, so will the enlarging lens, and most of the enlarging lenses in use today were made 30+ years ago when optical engineering was not nearly as good as it is now (due to increased available computational power). So it's conceivable to me that a negative scan from a modern macro lens could be indistinguishable, especially on something like a D850 with 40+ MP. The hybrid process can also access various digital sharpening techniques (e.g. digital unsharp mask and Richardson–Lucy deconvolution) unavailable in the analog process.
That said, I'd still bet on the analog print being sharper in the best possible circumstances, but I'd be surprised if the difference between that and an expert digital print were perceptible under normal viewing circumstances.
I agree, though I'm not sure at normal viewing differences the silver halide particles offer a perceptible advantage over a 600 dpi print. Even 300 vs 600 dpi is barely noticeable under magnification, if I remember an article by Jeff Schewe correctly.
But—thank you. You raise good points, and it does seem inkjets can't truly match these characteristics.
The technological advances in lenses due to computer aided design largely just apply to zoom lenses. Prime lenses, like those found on an enlarger, have seen very little, if any, improvements since the introduction of the computer. Their formulas are much simpler since you only have to correct at one focal length. This is why most macro photographers who are serious about quality still use enlarging lenses attached to bellows on their digital cameras. Even the best modern macro lenses made to mount directly on to digital cameras fall short of good enlarging lenses. The reason is because most macro lenses are made with compromises so they will produce usable images at long focal distances, as well as short ones, whereas an enlarging lens is designed for short focal distances only.
Also, I'm not aware of any commercially available printer that can print at 600 dpi. Many will claim high numbers, but they're using fuzzy math to achieve them. Basically, they'll claim it can lay down 600, 1,600, 2,440 dots per inch, but what they don't tell you is that those dots are laid down on too of each other, and are indistinguishable. Thus, usually 300-360 is around the highest number of individual dots most inkjet printers can lay down in an inch without making a messy blob. Look for the maximum number of dots a printer can put down in black ink. That will usually tell you the actual dpi resolution of the printer. The higher number they use to sell the printer to lay people is usually that number multiplied by the total number of ink cartridges the printer can use.
Okay. There are some issues here. First off, yes some newer prime lenses will be sharper. Yes coatings have improved. But so have manufacturing processes and the need for sharper lenses with higher resolution digital sensors. It's not that they couldn't have made sharper lenses in the past. It's that there was no need to. The formulas for most prime lenses have remained unchanged for the last 50 years or so. What's changed is the precision with which they are made. Computer controlled machines can maintain much higher tolerances than even the most skilled Craftsmen.I am fairly certain your claim is wrong. Certainly the zooms have gotten better. But the primes have too, especially away from the telephoto end (where they are easier to design).
For example, compare third party lens reviews and MTF charts for the Nikon 50mm 1.8 G to older 50mm lenses. It's just sharper and better performing. Admittedly it distorts a bit more than the AF-D 1.8, but presumably this is a purposeful compromise because we have digital distortion correction now. (And this was designed relatively long ago. The new 50mm for the Z mount is amazing and essentially as good as a Zeiss lens that costs multiples of its price, but it's taking advantage of both increased design power and the shorter flange distance of the new mount, so unfortunately it doesn't really prove the superiority of modern design per se. But it is certainly suggestive.)
Also, design aside, everything else that goes into the construction of lenses is better today, including the coatings, the glass, and the quality control. (You might raise the possibility that the superiority of contemporary prime lenses I noted above is due to these factors and not better computer-aided design. I don't have any hard evidence this objection is wrong but I'd be highly suspicious. In any case, the exact reason why they're better is not really relevant to this thread.)
I haven't checked the wide angles out extensively, but I'm willing to be the difference is even more obvious there. If I recall correctly, the 28mm AF-D Galen Rowell used is way worse than what we have available today. Same for the vintage 35s.
I don't know anything about enlarging lenses, but assuming they are superior to modern macro lenses, I'd bet it comes down more to purposeful compromises in lens design (as you indicated) than a lack of technological progress. I would also be curious to see MTF testing of the enlarging lenses to see how big this purported gap is.
I think it's fairly well known the Epsons can print at 720 dpi, and that this makes a perceptible difference. See for instance this testing by Schewe: https://www.digitalphotopro.com/technique/photography-workflow/the-right-resolution/.
I think the Canons can print at 600 dpi with the right settings, but this is less well documented.
Bingo.It doesn't matter how you slice it, scanning a negative will always reduce the quality of the image because you're taking a photo of a photo. If inkjet is all you ever want to print with, and you're concerned about things like resolution and color accuracy, then it's best to take your photos with a digital camera and never enter into the analog realm until your final print.
Okay. There are some issues here. First off, yes some newer prime lenses will be sharper. Yes coatings have improved. But so have manufacturing processes and the need for sharper lenses with higher resolution digital sensors. It's not that they couldn't have made sharper lenses in the past. It's that there was no need to. The formulas for most prime lenses have remained unchanged for the last 50 years or so. What's changed is the precision with which they are made. Computer controlled machines can maintain much higher tolerances than even the most skilled Craftsmen.
Also, keep in mind digital sensors operate differently than film. They require a sheet of glass to be placed directly over the sensor (which is usually a visible light bandpass filter to filter out IR and UV light) to protect their delicate parts. These filters tend to skew indirect rays of light, which isn't a problem with film, since film doesn't require the filter. So some lens designs, like the Biogon, can produce very sharp images on film, but will produce much softer images on a digital sensor. Modern lenses are made with this phenomenon in mind, so often comparing an old lens designed for a film camera, used on a digital camera with a new lens designed for a digital camera, used on a digital camera will give you the false impression that the new lens is sharper, when in reality, the new lens is just sharper on a digital camera. It may actually be softer when used on an old film camera.
But none of that matters, because of dithering, sensor noise, blah, blah, blah. It doesn't matter how you slice it, scanning a negative will always reduce the quality of the image because you're taking a photo of a photo. If inkjet is all you ever want to print with, and you're concerned about things like resolution and color accuracy, then it's best to take your photos with a digital camera and never enter into the analog realm until your final print
And ignore dpi. Dpi is a marketing term. It means how many dots they can fit in an inch. It doesn't mean how many individual dots you can discern in an inch after they've all been squeezed in there. You could put 10,000 dots in an inch, but if it looks like one massive blob, then you've only really got 1 dpi, though they'll sell it to you as 10,000 dpi. It's useless as far as resolution is concerned. What matters is lpi. Lpi means lines per inch. That will tell you how many dots you can place in an inch, without having them overlap. Lpi gives you an idea of the size of the dot. I'm not aware of an inkjet printer that can surpass about 175 lpi, which is about 350 true dpi (175x2=350). Photographic darkrrom papers can sometimes go above 2,000 lpi in the right conditions. Now I'm not saying these numbers are set in stone. But we're talking about numbers with orders of magnitude of a difference. They're not even close when it comes to resolution. Hold a printer's loupe up to the best inkjet print you can find and against an average silver gelatin print, and the differences should be obvious.
I think you will find that the highest end enlarging lenses have not been substantially improved upon - for the purposes that they are optimized for.
One thing that should be noted is that there is a tendency to try to evaluate the capabilities of each process by converting one into the other and then taking measurements of the respective results.
Whether you are converting analogue into digital by scanning, or converting digital into analogue using a film recorder, in almost all cases the mode of conversion has the greatest influence on the quality of the result.
Essentially, you distort the comparison if you try to evaluate either process using the measurement systems applicable to the other.
The lenses are still being made and are still usable for enlargers - they just don't bother marketing them in photographic markets.I can certainly believe this, if only because I don't know of any company manufacturing modern high end enlarging lenses in 2019.
The problem that condemns most inkjet printing for me, is they are vastly over sharpened.
I think the Canons can print at 600 dpi with the right settings, but this is less well documented.
You seem to seek validation rather than enlightenment. If you want the truth, it's as easy as comparing the two print methods yourself.Thank you, that is a lot of interesting information. But there are still a few things I'm not totally on board with.
So we agree that modern lenses are sharper, but not about the reason why. You claim in particular that lens designs have remained unchanged for the last 50 years. This seems demonstrably false to me, since Nikon, Canon, and others keep filing new patents for designs. I don't doubt that the new designs resemble old ones, but my understanding is that there are many parameters (distances, curvatures, etc.) to optimize in lens construction, and they're better able to tweak these using the increased computing power that's become available in the past ~20 years. I have even read a lens designer state that the choice of particular software package and numerical methods it makes available matters for the quality of the ultimate design.
Another data point: Brandon Dube (an optical engineer) has written that the lens patents often have key parameters "fuzzed"/obfuscated to protect intellectual property, since without these very precise measurements, which were determined with a great deal of computing power, one cannot fully recreate the lens (with optimal performance).
I would be interested if you could point to technical references that contradict these claims, since I'd love to read them (and keep buying vintage lenses...).
Thanks, I'm mostly familiar with this phenomenon in the context of the corners of the image going to hell when people attempt to adapt vintage lenses to modern DSLRs. Do you know where I could find data on this? In particular I'm interested in buying some lenses for my Nikon F100, and if modern lenses would be noticeably less sharp on it than modern MTF testing on DSLRs indicates, that would very, very interesting to me.
I love printing the darkroom. I just have to go travel a bit and rent one, and it's generally inconvenient. (Developing film in a small grad student apartment is not so bad, but optical printing is another mater...) My motivation in making this thread is really to judge the convenience/quality tradeoff, not trash darkroom printing.
Again, I agree with your main point about the order of magnitude difference in lpi. But—I just provided a link that sure suggests certain printers can get 720 dpi. If you have references that suggest this is false, I'd really love to see them, since it's "common wisdom" on both the internet and in inkjet printing books that Epson drivers operate natively at 720 and do nearest-neighbor interpolating when down/upsizing to accommodate input at different resolutions.
You seem to seek validation rather than enlightenment. If you want the truth, it's as easy as comparing the two print methods yourself.
John, I want to understand the technical properties of each printing medium better. I print both ways and have no plans to stop.
This is not really about seeking a magic solution. I know what I need to improve in my printing, and it's sure not the process being used.
As Clark Gable said in a 1939 Oscar Winning motion picture, (slightly changed), "frankly I don't give a da!!!!!, darn.". I prefer to shoot B&W where sometimes pre-WW2 uncoated lenses can yield superior results both on the camera and in the darkroom. Of course many of them have a "natural" coating..........Regards!I've read a few sources (e.g. Barry Thornton) that suggest that modern inkjet printing can reliably produce results that are technically and aesthetically superior to analog printing by optical enlargement. Is this true?
First, how does the dynamic range of photographic paper compare to that of consumer or "prosumer" inks and inkjet paper? (I'm sure this data is readily available online, but I can't find it.) I think this is really the key issue, though of things like avoiding color casts in the inkjet prints also matter. Are there are other measurable and objective technical factors to take into consideration?
Second, to get specific, let's compare a Canon Pixma Pro-100 (or something in that price bracket) printing monochrome on 8x10 semigloss to an optical enlargement on Ilford multicontrast 8x10 pearl RC paper with a good lens and enlarger. The prints are both 6x9. Suppose the Pro-100 is printing a good scan (from a 24MP DSLR) from the same negative being enlarged. Can the inkjet produce results that are as "good" as the Ilford print? If no, does the answer change if we use a better printer? (I'm purposely sticking to semigloss papers for the comparison since the Pro-100 is a dye ink printer and has trouble with matte papers.)
Thanks in advance for your comments. My suspicion is that inkjet technology has advanced to the point where it's superior, but I'm looking for data to substantiate (or disprove!) this claim.
i hate to say this, i hope you don't take me the wrong way, but people spent too much time digging deep into some sort of perfection realm ....
and looking for some magic solution for their ( taking, processing and ) printing needs. unless you will be looking at things under a magnifying glass / loupe i really wouldn't worry
about how much resolution you will be getting out of your printer. i just took a show down where i had 75 prints, all printed with an epson consumer grade
all in one scanner printer that cost about 85$. they looked like optical prints.
That is operator error. Good sharpening is easy to not see until it’s not there. Over sharpening is a really common problem in digital land and is a classic sign of somebody who isn’t very experienced.
I could not agree more, but I have even seen a good number of prints commercially made where the unsharp mask has been done to death. There really is no excuse. A few years ago when I could be bothered with a photographic club, mixing it with the prima donnas who HAD to have all the latest gadgets, longest lenses and usually most expensive cameras, I used to submit colour and black and white photographs (note - I did not call them images images) and one of these was in locomotive shed on a heritage railway.
The colour balance was good with no colour casts ( I have perfect colour vision) and the moody, lighting could not have been better suited. Grain was invisible, (I was using medium format) but the myopic judge criticised that it did not have the 'biting edge' sharpness of a digital print. She did not say the words 'digital prints' probably because that is all she had ever experienced. Now on this occasion, she was viewing a 12x16 print from no further away than 12" and a print this size is best appreciated from a greater distance.
After all you don't inspect an Old Master oil painting, even a small one, from as close as that, you stand back to get an overall view. This person obviously needed a little bit of tuition to appreciate what she was looking at because she could not even recognise the fact it was not a digital image, but a film based photograph. She was more concerned with the technical minutii than the visual content of what had been presented.If it had been unsharp then I would have agreed with her but it wasn't, the negative was bitingly sharp, it didn't need enhancing electronically.
The fact it was a film photograph was later made clear to her by one of the other members and I could tell that after the interval when judging recommenced, she was accutely embarrased and not so brashin what she said..
I later scanned the same negative and printed a section to the same releavant size and used the unsharp mask but not to excess, but in my view this destroyed it's charm and appearance. It is a case of each to their own. If I was a skilled carpenter, I would have loved to make my own furniture, but as I am not, I have to make do with something perhaps 'not quite as good'.
The photographic final print IMHO is the only thing that counts.. how we get there is irrelevant.
How does the final print look, how does it move you, that is what is important in my club ... not what stinkin equipment made it.
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?