It all comes down to the print

<p>The big variable that no one has mentioned is the resolving power of the enlarging lens. High resolving power on the film will be wasted if there is not an enlarging lens of as high a quality as the taking lens. Although my 50mm f/2.8 EL-Nikkor along with my 75mm f/4 EL-Nikkor are mid 70's vintage, they both have exceptional resolving power, every bit as good as my AIS Nikkor lenses. They only thing they do not possess is multi-coating, which I have never found to be an issue, especially since I print everything with a color head which has diffused lighting.</p>

<p>That being said, it still comes a no surprise to me that film is still king-a-da-heap.</p>

<p>Mauro:<br>

Could you clarify this, please? Are you saying that a 35mm frame enlarged and printed at 30x40 would look almost the same as a 120 frame enlarged and printed at the same size?</p>

<blockquote>

<p>If you want to know how 35mm film prints at 16x20, just look at the 30x40 print in your hand from MF film (approximately the same).</p>

</blockquote>

<p>Dave,</p>

<p>What I am saying is different. On the test sheet you just printed, look at the 30x40 samples from MF film, those look very similar to 35mm printed at 16x20.</p>

<p>Mike,<br>

When you look at the 30x40 sample on the sheet you just printed, disregard the DSLR for a moment. Look at how a home scan of MF holds detail on a print that large. It is fantastic. It is easier to explain when people have the print in their hands.</p>

<p> </p>

<p>Regarding the 5DII, I am trying to keep this article/post based on facts.</p>

<p>If you can share a test sheet from a 5DII at 30x40 I can print at home and look at it I will give you my opinion.</p>

<blockquote>

<p>On the other hand, a Canon 5DII is about the same size as a Canon 3-series film body, and the 5DII blows away the film body in terms of image quality.</p>

</blockquote>

<p>Counterpoint: not if you use good film and print, it doesn't, unless it's much better than the M9. It has advantages compared to even the best film, but the <em>categorical </em>superiority of its sensor is not one them.</p>

 

<blockquote>

<p>Compare a modern medium format digital back with its film equivalent, and they’re about the same size.</p>

</blockquote>

<p>... about the same size, not particularly more convenient, and one or more orders of magnitude costlier.</p>

<blockquote>

<p>But you have to go to large format film before the image quality compares with the medium format digital.</p>

</blockquote>

<p>Jeeze, how good could it be? It's an honest question. I rented a D3X one weekend and was underwhelmed; the most expensive Nikon that many attribute MF-quality to has nothing image-wise on my cheap 500CM or P67. I queried flickr for phase one and, meh, not THAT much better. I think I'll keep my tens of thousands of dollars in the bank, instead.</p>

<p>OK I don't have them to begin with.</p>

<p>But even if I did, film is a great, super, A+ technology, and for some people's (hi) pattern of use, it's just better, funner, cheaper.</p>

<blockquote>

<p>meh, not THAT much better</p>

 

</blockquote>

<p>Than the D3X!</p>

<p>Thanks for all this Mauro. It should provide a frame of reference for people when they ask how good these different formats are at different sizes. This type of reference would have been nice when the arguements about 3mp sensors beating 35mm scans was all the rage nearly a decade ago. </p>

<p>I need to figure out what type of scanning equipment I want to get to pull the most from my MF and 4x5.</p>

<p>You are welcome. </p>

<p>I have gotten several people thanking me for the test sheet so it looks like it provided value.</p>

<p>Thanks Mauro!<br>

And seeing what a 35mm at 16x20 would look like is a big help. I wasn't sure if it would be worth enlarging to that size and spending money on the paper in the process.</p>

<p>After reading the post again, I see how deprivation really <em>really </em>hurts reading comprehension ... oy, oy, oy!<br>

I need to go to bed!</p>

<p><em>Most people by now know that many 35mm films (especially B&W) outresolve DSLRs and Leica digitals.</em></p>

<p>You need to test modern bodies Mauro. My Canon 7D out resolves 35mm Velvia 50 and pretty much matches 35mm Tech Pan (no longer made) in Les Sarile's map test. (See attached.) Pretty much every other color film made resolves less than Velvia 50. As I recall, Velvia 50 is rated as an 80 lpmm film while just about everything else falls in the 50-60 lpmm range.</p>

<p>I think it's safe to say that 15 MP class DSLRs and higher match and exceed what 35mm color film can deliver in resolution, assuming average detail contrast. (You can always generate impressive film resolution numbers by using special, high contrast test targets. Even the B&W target you've selected has much higher contrast than fine detail has in an average scene.) Many B&W films are in the same range as far as resolution goes, though certain films and film/developer combinations can achieve numbers comparable to or better than Velvia.</p>

<p>A 7D cannot match MF with a single shot, but it comes surprisingly close considering the very large difference in sensor and film size. Certainly there are subject matter / print size combinations where the resolution of MF is beneficial and easily observed. But I'm finding that a simple 3 frame stitch gives me images equal to or better than what I could expect from MF. To each his own, but at this point I would rather invest money in my Canon system than invest in a CoolScan and additional MF equipment. I do still occasionally shoot B&W in both 35mm and MF and develop/print in a darkroom, but that doesn't require a CoolScan level investment. What interests me more than MF is LF, both because of the movements and because of the more complicated stitching required to match that level of detail.</p>

<p>I'll post additional samples after this post.</p><div></div>

<p>Canon 7D, single shot, vs MF, again using Les Sarile's map test.</p><div></div>

<p>Canon 7D, 3 frame stitch, vs the same MF shot.</p><div></div>

<p>I always include a link to Les Sarile's site when I mention it or show these samples. He put a lot of work into it. http://www.fototime.com/inv/E0D372FC8001820</p>

<p>He posted an 8000 ppi Imacon scan of 35mm Velvia 50 in one thread, but I don't think that scan has been posted to his site yet. On the Imacon 35mm Velvia 50 did better, but still did not have the resolution, sharpness, or lack of noise that the 7D sample has. Judging from the samples, particularly the Imacon sample, it looks like Velvia 50 might have comparable resolution on high contrast details, but falls short on low contrast details, which is consistent with what we know about film and digital sensors.</p>

<p>I should note that my 7D sample was shot with a Canon 50 f/1.8 at f/5.6. It's noted in the samples that I used ACR. On this test ACR was able to resolve noticeably more fine detail than DPP.</p>

<p>None of my posts should be taken as a knock against film or anyone who continues to use film. Film is a very capable medium and while I love my DSLRs (especially my 7D), I still enjoy the darkroom and would recommend a course in classic B&W film developing and printing to any photographer. Some of the color films have very unique characteristics which are also worth exploring. Knowing more tools and mediums can only be an advantage to an artist.</p>

<p>Thank you Daniel.</p>

<p>I like the 40D and the 7D but I'm not particularly fond of stitching. You should know though that the 7D resolves only 9 megapixels (2450x3750 - See DPreview) whereas 35mm TMAX resolves about 40 megapixels and 35 mm Techpan about 90 megapixels. They are world apart. A 4000dpi scan of 35 mm film is only going to give you some 20 megapixels - but there is a lot more information on film that a desktop scanner like the Coolscan can capture you. At lower contrast resolution decreases.</p>

<p>Another lesson learned on film, don't quote resolutions you haven't tested. The manufacturer's nominal resolution is just a guideline. For example TMAX is published at 200 lpmm and I tested 230lpm. Velvia 50 measured 180lpmm.</p>

<p>I really don't want to take this thread to a resolution contest that doesn't make sense. Nor it is the point of this test.</p>

<p>Can you please provide a link to the full shot of the map you tool with the 7D? I would like to look at it and also print it at 30"x40" to observe it.</p>

<p> </p>

<p>Daniel, I am also interested to know what was your appreciation after running the test in this thread. What did you think about the 30x40 print?</p>

<p>Les - I'll send it next week. This week is...well...tax season :-)</p>

<p>Daniel,<br>

As you point out, at least a Coolscan is needed (in my opinion as well), to scan film adequately in terms of resolution, color and density.</p>

<p>I am not sure what scanner you have, but this can easily be the cause of less than adequate results.</p>

<p>To each his own, I agree as well. My main print sizes are 24x30 and 24x36 on my Epson 7880 and at these size or larger I prefer the results from film (35mm and MF) than from a DSLR.</p>

<blockquote>

<p>A 7D cannot match MF with a single shot, but it comes surprisingly close considering the very large difference in sensor and film size.</p>

</blockquote>

<p>In the sense that a P&S comes surprisingly closer to a 5DII. <a href=" is what 67 can do. <a href="http://www.flickr.com/photos/manualcrank/4524991140/">Here</a> is what the K20 can do. The K20D is not the 7D (it's 0.4 DXO marks worse, whatever they measure) but I seriously doubt the 7D is light years ahead of another APS-C camera. The Howtek scan looks far better than the 7D, and something those crops can't show is the effect of visually integrating its superiority over the whole image, such that I'd be embarrassed to display the full size prints together. Also, what's with the missing lake in one of the comparisons?</p>

<p><em>You should know though that the 7D resolves only 9 megapixels (2450x3750 - See DPreview) whereas 35mm TMAX resolves about 40 megapixels and 35 mm Techpan about 90 megapixels. They are world apart.</em></p>

<p>Mauro - DPReview reports their resolution numbers in units of LPH, or Lines per Picture Height. LPH, as derived from their chart, is not equal to pixels and cannot be used to count megapixels in the manner you have assumed above. It's also important to remember, in any comparison to traditional tests, that DPReview reports both an "absolute" and an "extinction" number. Both terms are misleading IMHO, but the absolute number would be roughly comparable to the MTF50 result in classic tests, and the extinction number to MTF10 results.</p>

<p>Along those lines I would like to see your evidence and calculations regarding the resolutions listed for TMAX and Tech Pan. T-MAX 100 has a rating of 63 lpmm, which calculates to roughly 14 MP. Tech Pan was rated as an 85 lpmm film, which calculates to roughly 25 MP. (Both ratings for normal contrast scenes.) However, a strict conversion of lpmm to MP fails to take into account various issues which can result in a lower than expected performance when comparing to digital sensors on tests other than traditional line charts.</p>

<p><em>A 4000dpi scan of 35 mm film is only going to give you some 20 megapixels - but there is a lot more information on film that a desktop scanner like the Coolscan can capture you.</em></p>

<p>In all honestly I would expect the 35mm Tech Pan sample to edge out the 7D sample if scanned on an Imacon. But there's not that much difference between the CoolScan and top end units like the Imacon or a drum scanner. As I pointed out, 35mm Velvia on an Imacon still can't pull even to a 7D, and a straight conversion of Velvia's 80 lpmm rating yields 22 MP.</p>

<p>Under a high contrast test I would expect 35mm Velvia to edge out the 7D. But in the real world, where most fine detail occurs at low contrast, one would expect results similar to the map comparison.</p>

<p><em>Can you please provide a link to the full shot of the map you tool with the 7D? I would like to look at it and also print it at 30"x40" to observe it.</em></p>

<p>I'll post it for both you and Les ASAP.<em><br /></em></p>

<p>The printer resolution is the real limitation here. You don't see much different in quality between 35mm and larger format unless you make real large size print. If you want to make good quality print of moderate size (8x10, 11x14) from either medium or large format film then you should print it digitally but rather print it the old fashion way using an enlarger and print it optically. Otherwise you won't see much different in quality when using larger size negative.</p>

<p><em><a rel="nofollow" href=" target="_blank">Here</a> is what 67 can do. <a rel="nofollow" href="http://www.flickr.com/photos/manualcrank/4524991140/" target="_blank">Here</a> is what the K20 can do.</em></p>

<p>Your links are not to comparable pictures, and are not available at original sizes. There are no conclusions I can draw from this other than you chose to shoot a more interesting subject and lighting situation with the 6x7.</p>

<p><em>The Howtek scan looks far better than the 7D, and something those crops can't show is the effect of visually integrating its superiority over the whole image, such that I'd be embarrassed to display the full size prints together.</em></p>

<p>As I said earlier, to each his own, but I consider that an exaggeration. It's clear the Howtek MF scan is better, but I certainly would not be embarrassed to compare, say, 24" prints. (Depending on your monitor those crops are like looking at 60-70" prints.) Certainly the gap grows as the print size grows, which is why I sometimes stitch 3 frames with the camera at the opposite orientation of the scene I'm shooting. If I think I'm going to want a landscape scene printed larger than 24", I stitch and the results are excellent as can clearly be seen from that comparison.</p>

<p><em> Also, what's with the missing lake in one of the comparisons?</em></p>

<p>Same map "model", but not the same map. That said...what lake? There are some color and line differences, but I never noticed a lake difference.</p>

<p>I posted 100% crops at both links. I don't have comparable scenes but those two images are representative of either camera's ability to resolve detail. The difference is drastic and I don't know why you'd think shooting film in good natural light would make it less so. Never mind the lake, it's back again.</p>

<p>" (Depending on your monitor those crops are like looking at 60-70" prints.)"</p>

<p>Daniel,</p>

<p>You can't judge from a monitor. You need to make a print .</p>

<p>As you can see from the print test I posted, if you are holding it in your hand, the different MF films look excellent under nose scrutiny at 30x40.<br>

The 7D is not in the printed sample so you can't compare it without taking a very practical test into just a hypothetical discussion.</p>

 

<p>If you wish to compare your 7D to the print test I posted, you should shoot a tager sith the same composition and print it.</p>

<p>Then look at both together in front of you.</p>

<p>It is really not worth it in my opinion. The differences and too large between MF film and a DSLR, especially at larger sizes.</p>