Scanning 120 Chrome on V700 vs. Sony A900 DSLR

[john_bradshaw6]

<p>Hi - I am wondering if anyone could share some wisdom on this.<br>

I am shooting 120 Astia 100F with a Mamiya 7II at the moment and<br>

my question is, would I get better results ( 16" x 20" prints ) scanning<br>

the film on a Epson V700 with Vuescan or instead shoot the landscapes with<br>

a Sony A900 with a Sony/Zeiss or Sony G lens?</p>

<p>Cheers, John.</p>

[michael_burrows1]

<p>Is this Sony a 100+ megapixel camera? Mamiya 7II has some of the sharpest glass on the planet.</p>

[piotr_fracala1]

<p>Mamiya, but find better scanner, please...</p>

[gt1]

<p>Keep in mind you can only get an effective 2000dpi with the V700. Even with the V700 you will get better results with the Mamiya 7 plus V700. And if you want larger prints for your favorites get them scanned with a Nikon 9000 or a drum scanner.</p>

[john_bradshaw6]

<p>Thank you.</p>

<p>Cheers, John.</p>

[leicaglow]

<p>My guess is it would be a toss up, but if you scanned on a Nikon 9000, Hasselblad, or drum scanner, the odds would tip toward the MF negative. Though 16x20 is not that large of a print.</p>

[mitchfalk]

<p>Save your money, don't buy a scanner for anything smaller than a 4x5, you'll be dissapointed when you see the comparison between a pro-sumer scanner and a scan from a lab such as North Coast Photographic Services.<br>

They can scan your 120/6x7 film to equal a high end medium format digital back for about $11.00 bucks a roll.</p>

[mitchfalk]

<p>Check this out, zoom in ! Pentax 67 - 45mm lens at f.22, Fuji Velvia 50.<br>

They do a nice job.</p><div></div>

[john_fleetwood]

<p>North Coast Photographic Services does a good job with 35 slides, but I haven't been happy with 120 chromes they've scanned. Looks like film flatness issues to me.</p>

[tony_lastoria]

<p>Having worked with 4x5 and 6x7, owned a Howtek drum scanner and the Epson V750; I sold the drum scanner and camera and replaced it with the Sony A900 and never looked back. If you are printing no larger that 16x20 the Sony should meet your requirements. It also will allow you to shoot and experiment more than you probably would with a film camera. I would not recommend scanning with Epson for prints larger than 11x14.</p>

[dave_redmann]

<p>The scanner is only part of the limiting issue for the Mamiya. Fuji's data sheet for Astia 100F shows that the MTF curve is down to 50% response at about 35 lp/mm; if you accept the 50% MTF response point as the (semi-arbitrary, but widely-used) resolution,* that would translate to onlu 1778 ppi, or 3920 x 4865 for a 6x7 frame. And that's assuming your scanner has 100% MTF response at 1778 ppi.</p>

<p>And resolution is only part of the issue. The A900 will give much better color accuracy than any film, more dynamic range (ability to capture detail in highlights and shadows) than Astia or any other transparency film, and very likely less noise/grain. And achieving the A900's optimum results doesn't require a high-quality E-6 lab, a drum scanner, and skilled operators; just copy the files over.</p>

<p>I don't doubt that, in a really large print (larger than 16x20), using appropriate film (not Astia 100F), with a great scan, 6x7 film can give slightly better retention of fine detail than the A900 can. But will a reasonably skilled photographer, in the real world, for your stated 16x20 inch prints, I think the A900 is likely to look better most of the time, and it will certainly be a lot more flexible and a lot less of a hassle.</p>

<p>*There are those who do, or appear to, advocate for crediting film with higher resolutions, but there examples usually show a lot of extra grain/noise for a small amount of additional detail.</p>

 

[brettdeacon]

<p>John: I have a drum scanner and the A900, and just checked out some 100% crops of scanned MF Astia (center crops of 6x17 images taken with my 4x5 and the Rodenstock 150 APO Sironar-S, regarded as one of the sharpest 4x5 lenses available) and A900 images (Zeiss 24-70 lens, optimal aperture, tripod, mirror lock-up, etc.). At 16x20 and 300 dpi, there is more fine detail in the MF scan. Both images have essentially no grain at this size, at least prior to sharpening. At 50% magnification, which is in my experience the best magnification to judge the appearance of the final print, the difference between the two is negligible and in all likelihood irrelevant to all but the most extremely picky viewers.<br>

This comparison applies to drum scanned MF film. The Epson V700 has far less resolution and dynamic range than a drum scanner. With good glass, the Sony A900 will be very competitive with MF Astia scanned with the Epson in terms of resolution and will have superior dynamic range. Other advantages include better depth of field (if this is desirable) and far greater post-processing flexibility (HDR, focus stacking, etc.). If I were in your shoes, I'd most likely go with the A900 and Zeiss glass unless my client was insistent upon me shooting film. Plus, the A900 (and all DSLRs for that matter) is a lot more fun to use!<br>

This just my 2 cents of course; feel free to take or leave my advice. From my experience working in an art gallery I wonder if photographers like us care far more than 90%+ of consumers about the kind of very small differences in image quality evident in the scenario discussed in this thread?</p>

[Ed_Ingold]

<p>It's a wash in a 16x20 inch print, probably even larger. You can scan 6x7 cm film to any resolution you want, but a well-crafted 22 MP DSLR image with good glass is going to do as well with far less fuss and bother at that size of print. As a practical matter, each pixel in a DSLR is worth three from scanned film. Accutance counts - resolution is an academic issue.</p>

<p>A print which resolves 4 lp/mm will look critically sharp, so it simply doesn't matter if the source file is 8 or 16 lp/mm.</p>

[guido_h]

> [...] my question is, would I get better results ( 16" x 20" prints ) [...]

 

You need to define your understanding of "better results" more precisely. The aspects of resolution, dynamic range, cost/convenience, lenses and scanners available etc. each lead to a different answer to your question.

[roger_smith4]

<p>Hey Les, cann you add a slide film to your dynamic range test?</p>

[brettdeacon]

<p>Les, I'm curious as to whether you've printed images shot with a 20+ MP DSLR like the A900. I'd be interested in hearing your thoughts as to how prints at various sizes, say 12x18 and 16x20, compare to scanned 35mm and MF film.</p>

[brettdeacon]

<p>In all honesty, I don't have much interest in performing such a test because I am very satisfied with the dynamic range in Sony A900 RAW files at ISO 100, which is what I shoot 95% of the time. I'm used to shooting slide film (Astia) so perhaps my standards are not as exacting as others who use negative films. Your tests show that Kodak Ektar does a better job than some digital sensors in retaining detail in very overexposed highlights, but with instant feedback via histogram I can tell when I've blown out the highlights and just take another shot at a different exposure.<br>

The question I posed to you had more to do with the issue of print quality being discussed on this thread, which I take to mean issues such as resolution, noise/grain, etc. for properly exposed images. I take your non-answer to my question to mean that you haven't printed with images from the Sony A900 or Canon/Nikon 20+MP DSLRs. Those of us who have appear quite satisfied with the print quality relative to scanned film. Maybe it's just me, but I don't see how your blown highlights image is particularly relevant to the print quality of properly exposed photos. I'm honestly not trying to start a flame war here, but I think it's important to value the opinions of with personal experience related to the topic of print quality using the different systems mentioned by the OP.</p>

[brettdeacon]

<p>Sigh. I lack the time and interest for a pissing contest, so let me briefly respond and end my contributions to this thread.</p>

<p>Your statement that I "believe blown highlights to be irrelevant when it comes to prints" indicates that you didn't bother to read what I wrote. I said that the issue of blown highlights on film vs. digital is not particularly relevant to the print quality of *properly exposed photos.* I'll grant you the likelihood that were I to overexpose all my photographs by 3 stops, I'd get fewer blown highlights with Ektar than my A900. So what? I’m not even going to get into HDR photography, which I do quite often with my A900 to get dynamic range far beyond Ektar.</p>

<p>It’s great that you feel confident contributing to a thread on the print quality of scanned film vs. the A900 when you’ve never printed an image with the A900 or another 20+MP DSLR, and confident that the personal experiences of those of us who have are invalid and based on misinformation and ignorance. It’s not something I would choose to do, but to each his own. Taking me to task for not being interested in testing the dynamic range of my DSLR is, again, irrelevant to a discussion on the print quality of *properly exposed photographs.* I appreciate your efforts to “inform” me, but I’m not stupid enough to misunderstand that MF film has more resolution than 35mm film, or that scanned MF film has more resolution than the Sony A900. This thread is about discussing the quality of 16x20 prints, and a number of variables are relevant to this discussion that photographers like me who have printed with both scanned MF film and the Sony A900 are in a position to answer from personal experience. Despite the resolution advantage of MF film over the Sony A900, I guess the reality is that the digital images seem to be "good enough" for us, at least at print sizes of 16x20 and smaller. And let's keep in mind that the OP will be scanning on An Epson V700, which probably doesn't resolve even half of the detail of a Mamiya 7II photograph captured on Astia.</p>

[dave_redmann]

<p>Brett, you'll be happiest accepting that there's no convincing Les, and no point in arguing with him.</p>

<p>

<p>Les, as is so often the case, your claims don't bear close scrutiny, although I will leave it to others to opine whether the cause is lack of knowledge, less than full candor, or some combination of the two. For example, your assertion regarding "the 12.4MP Nikon D2X and the 24.6MP Sony A900, you will note that although the A900 has 98% more pixels, it only has 35% more vertical resolution and 13% more horizontal resolution," is misleading at best. If a camera has 98% more pixels, all else being equal, the <em>linear</em> resolutions would only go up a hair over 40% (1.407 x 1.407 = 1.98). Also, DPReview actually gives two separate resolution figures for each dimension ("absolute" and "extinction"), and depending on which ones you use, the A900's horizontal resolution is either 13% <strong>or 32%</strong> higher, and the A900's vertical resolution is either 35% <strong>or 42%</strong> higher. (The data is from <a href="http://www.dpreview.com/reviews/sonydslra900/page35.asp">http://www.dpreview.com/reviews/sonydslra900/page35.asp</a> and <a href="http://www.dpreview.com/reviews/nikond2x/page27.asp">http://www.dpreview.com/reviews/nikond2x/page27.asp</a>.) If you take the average of the two types of resolution, the test results imply that the A900's effective pixel count is about 71% higher than the D2x's, on 98% more pixels. Certainly we should expect diminishing returns on extra pixels at some point, and you will note that (IIRC) DPReview's tests show the A900 to have a slight resolution edge over the D3x and the 1Ds Mk. III.</p>

<p>But all that is mostly measurebation. The point about dyanmic range is that usually you just don't need it. Lots of enthusiast and professional photographers used to shoot most or all of their pictures on transparency film, and did fine with it. Anything you can capture on transparency film, from a dynamic range standpoint, you can capture on any modern DSLR. So the fact that negative film as more dynamic range is relevant only to a very small proportion of images.</p>

<p>And the point about resolution and print size is that, with almost any non-test-target subject, at a normal viewing distance, the A900 (or 1Ds Mk. III or D3x) supplies enough detail for a 16x20-inch print that the limiting factor is more your eyes than your capture. Also, while I don't doubt that, say, a 6x7 TMX frame from a M7II can contain a lot more detail than a file from an A900, if you compare the A900 to, say, a 645 frame of Astia, the film has little if any resolution advantage.</p>

 

</p>

[dave_redmann]

<p>Less, obviously it is physically impossible for me to post anything here that will demonstrate what is, and is not, visible in normal viewing of a 16x20-inch print. There's no way to compare what you see on a roughly 90 ppi trransmitted light image (like what you see on a typical monitor) with a roughly 300 ppi reflected light image (the typical print).</p>

<p>But consider this: I think if you research it, you will find that there's some consensus that the limits of "perfect" human vision are about one-half minute of angle (MOA), in other words, the finest detail we can see occupies 1/120th of a degree in our eyes' angle of view. (Yes, I realize a few rare people have better than "perfect" vision, e.g., 20/15.) This translates into just over half an inch (13.3 mm) at 100 yards (91 m). To pick out detail that fine, it has to be very high contrast, like pure black against pure white.</p>

<p>Now I don't know how you look at 16x20-inch prints, but to me, normal viewing distance is about 3 feet (0.9 m). At that distance, a 1/2 MOA detail is 1/191 inch wide. In other words, at a viewing distance of 3 feet, the finest resolution that "perfect" human vision can see is about 191 ppi. And again, that's pretty much for picking out black on white, or nearly so; you can't see, say, red from blue, or dark gray from light gray, in that fine detail.</p>

<p>Back to the A900. If we accept DPReview's "absolute" resolution of 2700 lines and "extinction" resolution of 3700 lines, on a 16x20-inch print, those translate to, respectively, 169 ppi and 231 ppi of effective resolution. So the point is, with a 16x20 at a normal viewing distance, even "perfect" human vision will be able to see no or virtually no more resolution than the A900 can deliver.</p>

<p>So if you want to view prints up close, with your nose to them, I don't doubt that you could see more detail than the A900 (or for that matter, 6x7 Astia) can deliver, although at some point the printer will be the limiting factor. And I don't doubt that some really big prints almost invite closer inspection of less than the whole print. But if we're talking about 16x20-inch prints at normal viewing distances, well, the problem isn't <em>my</em> eyes, it's the limits of human vision.</p>

[roger_smith4]

<p>"However, your opinion about any modern DSLR matching the dynamic range of transparency film goes against even Canon's declaration that only it's full frame sensors do and that the none full frames do not."</p>

<p>I'm still waiting for you to compare apples to apples incorporating a slide film into your DR test, as you already have a methodology. Astia would be a good choice- maybe include Velvia too as it's still popular. <br>

To be fair to the slide film you might need to scan twice and blend with your Nikon- I'm finding this to be necessary with my LS-5000 to preserve highlight and shadow detail as it appears on the slide.</p>

[Dave Luttmann]

<p >I’ve only played with online samples of the A900, so to be fair, I can’t really comment on those as I’m held hostage to the processing of others. The 5D2 however is pretty close, and I’ve had the opportunity to work with the camera a little and can state the following findings:</p>

<p > </p>

<p >When printing both 6x7 Astia, scanned on an Imacon at 3200ppi, (better than my V700 by a fairly decent margin.....even when fluid mounted in a proper holder) and comparing that to shots done on the 5D2 and processed in Capture One, at 16x20 (16x24) on an Epson 3800 with HM Photorag 308, I found them to be virtually identical when viewed from 2 to 3 feet. Maybe a lustre paper would show more….but I use HM308 for most of my printing. Maybe a drum scan would have been better….but based upon past experience, I don’t think the differences would be a deal breaker.</p>

<p > </p>

<p >The film print did indeed contain the slightest bit more detail, but the digital file maintained acutance better. The 5D2 contained more dynamic range than I’ve ever been able to obtain from Astia….maybe not from a color neg film, but compared to Astia, it was better by about 2 stops.</p>

<p > </p>

<p >Let it be said that in term of real detail, viewed on screen, the film won out easily. But on print at a small size like 16x24, the differences basically vanished. At 20x30, the film held the detail advantage, but grain started to rear its head on the film file that in my opinion, started to negate the advantage from a viewing standpoint. To obtain better results, I’d have to resort to 4x5 film.</p>

<p > </p>

<p >So yes, the film is technically better, but on print at that size, it really doesn’t matter. Considering the overwhelming majority of prints I sell, and most others here I presume, are smaller than 20x30, and maybe even 16x24, the latest crop of DSLRs from Canon, Sony & Nikon that exceed 20mp are sufficient to produce gallery quality prints, when processed carefully from Raw using a workflow to maximize detail.</p>

<p > </p>

<p >If you regularly print 20x30 and larger, then a MF digital back or 4x5 film will hold an advantage. If you print at 16x24 and smaller, then sheet film and MFDB will probably be considered overkill and not make any real difference.</p>

<p > <br>

This is based upon my viewing of prints, and just sums up my findings. And yes, I’m a lover of film…..but don’t mind stating when it really doesn’t matter between the two formats.</p>

[mattb1]

<p>Using an Epson on MF? More than likely you will be disappointed IMO, they scan 35mm much better. I have no MF scans from the time I had a Epson, nothing ever came out well. Pretty much just wasted my time using a Epson.</p>

<p>Get a better scanner if you want to shoot film.</p>

<p>The Sony will be much better in this comparion. </p>

<p>Digital is much cleaner and will provide very pleasing prints at large sizes. Detail is a touchy subject, I'd love to have a 100mp digital camera as I love detail. However, there is no denying how clean and sharp a DSLR image is or how pleasing they can be to the viewer.</p>

[Robin Smith]

<p>John,</p>

<p>I think you would be happier with the results from the Sony.</p>

[yoshio]

<p>when I read things like this:</p>

 

<blockquote>

<p>Using an Epson on MF? More than likely you will be disappointed IMO, they scan 35mm much better. I have no MF scans from the time I had a Epson, nothing ever came out well. Pretty much just wasted my time using a Epson.</p>

</blockquote>

<p>it is simply illogical.120 film has significantly more film area than 35mm does, so therefore to produce a 5000 x 4700 pixel image you only need 2000dpi scanning. This is within the range of the Epson, to scan 35mm at such you would only get 2700 x 1800 pixels. To get a result which makes a comparison you would need 4000dpi, at which the Epson is well past its optimal. The only way I can imagine it is if you scan fuzzy or out of focus MF and compare it to sharp 35mm.</p>

<p>Unlike a Flextight or other scanners the head does not zoom to different positions giving optical enlargement, so you will get better results from MF film than 35mm on the Epson. It makes me wonder if the poster has actually any experience on much equipment.</p>