Poor scans - May explain why some switch to Digital

[john_photo]

<p>The Nikon 9000 with glass holder runs around 2500 if you can find one.<br>

I'm amused at those, especially the original poster, who act as if this is the best one<br>

can get out their MF film.<br>

Far from it. That is why the Epsons are such outstanding alternatives for proof printing.<br>

I certainly wouldnt want a 30x40 printed from a low end nikon film scanner nor do most gallerys.<br>

If you shoot medium format in the first place, at least in the old days, you were after maximum quality. This requires a drum scan or imacon in the eyes of many professionals who regulary exhibit their work.The Nikon 9000 does NOT in any way fill this niche market.<br>

The problem is not the Epsons. Even 10 years ago when Medium Format was going strong very few would use an 8000 Nikon to blow up their finest work. The problem is that nothing touches a drum scan and drum scans are expensive (in the same way that Medium Format gear used to be exorbitantly expensive)</p>

<p> </p>

[roman_p]

<p>Rishi,<br>

it looks to me like color fringing or chromatic aberration. but it's optical issue.</p>

[rishij]

<p>wow, that's an L series 17-40 f/4 Canon lens. that hurts :(<br>

-Rishi</p>

<p>p.s. Don E that's not really a valid guess as to what format it is :P</p>

[swilson]

<blockquote>

<p ><a href="../photodb/user?user_id=423641">Landrum Kelly</a> <a href="../member-status-icons"><img title="Subscriber" src="http://static.photo.net/v3graphics/member-status-icons/sub8.gif" alt="" /><img title="Frequent poster" src="http://static.photo.net/v3graphics/member-status-icons/1roll.gif" alt="" /></a>, Mar 18, 2009; 08:15 p.m.</p>

 

<p>Oval smokestacks: There might be software solutions to such distortions. Maybe Scott will know.</p>

 

</blockquote>

<p>

<p>You get those kinds of ovals from any really wide angle image that is rectilinear, this is true of either digital or film. The effect can be reduced by not doing a rectilinear, going to somewhat of a fisheye. There is simply no way to go very wide angle without some kind of distortion. The image also has perspective correction, like a LF camera would, which exaggerates the wide-angle effects of the lens.</p>

<p>I have linked to a version that is a bit less wide angle and have no perspective control to it. The FOV is very close to what you would get with a 35mm camera with a 25mm lens on it. The tipping in of the stacks is from a wide-angle lens with no shift added to it, i.e. a normal 35mm lens. I set the resolution to match a full 35mm frame at 4000 ppi.</p>

<p>Next I will do a version that has the highlights toned down for Les.</p>

</p><div></div>

[StuartMoxham]

<p>The fairy falls shot looks nice. Personally I would have framed a bit tighter to lose most of the foreground and some of the right hand side. That would allow the falls to appear larger and look more dramatic. As for what it was shot on. It does not really matter. More intersesting would be how you might improve on that image if you were to go back again.</p>

[rishij]

<blockquote>

<p>How much creative you would be if your brain can recognize only two digits -1 and 0?</p>

</blockquote>

<p>What on earth are you talking about, Roman? As a 'design engineer', Roman, you should be ashamed for making a comment like that.</p>

<p>1s and 0s can be infinitely powerful if you increase the sampling dimension to infinity. Therefore, 1s and 0s are just as powerful as analog in the limit of infinite samples.</p>

<p>With all due respect, that's just a ridiculous statement. And I'm a biologist/film shooter. Sheesh.<br>

-Rishi</p>

[swilson]

<blockquote>

<p ><a href="http://www.photo.net/photodb/user?user_id=696354">Les Sarile</a> <a href="http://www.photo.net/member-status-icons"><img title="Subscriber" src="http://static.photo.net/v3graphics/member-status-icons/sub6.gif" alt="" /><img title="Frequent poster" src="http://static.photo.net/v3graphics/member-status-icons/2rolls.gif" alt="" /></a>, Mar 18, 2009; 06:00 p.m.</p>

 

<p>Scott Wilson,<br />Below is an example that I consider to be very sharp. Do you have a similar scene?<br /><a rel="nofollow" href="http://www.fototime.com/CE54B7638DEF5F2/orig.jpg" target="_blank"><img src="http://www.fototime.com/CE54B7638DEF5F2/thumbnail.jpg" border="0" alt="" /> </a>Click thumbnail for full res 2MB file.<br /><br />Zoom to the center of her vest and you can actually read the label. My real world may not be yours?</p>

 

</blockquote>

<p>

<p>I resized one of my 8MP photos to match your size and put them side by side. This is very typical of what I see between film and digital, a good film scan (which yours is) will have a bit more detail then an 8MP digital image, but you image still looks to my eye to be just a bit softer. FWIW I have seen scans from 35mm that can match what a 8MP DSLR can do for sharpness, but I have never seen one that goes past.</p>

<p>To say that people who have been shooting with 8MP DSLRs have been missing something over using 35mm film does not make sense in light of the 35mm photos I have been seeing.</p>

</p>

<blockquote>

 

</blockquote><div></div>

[rishij]

<p>OK, I'll make this one easy for ya, folks:</p>

<p>Here's a 12MP image file... is it from a 12MP dSLR or a 12MP scan of a 35mm frame of film?</p>

<p><img src="http://staff.washington.edu/rjsanyal/Photography/MtHood.jpg" alt="" width="800" /><br>

<a href="http://staff.washington.edu/rjsanyal/Photography/MtHood.jpg">Link to Full-Size Image</a></p>

<p>I want at least 4 votes... then I'll divulge which image is which format between the Fairy Falls image and this Mt. Hood one. Sorry to keep ya waitin' Lannie :)</p>

<p>Cheers,<br /> Rishi</p>

[StuartMoxham]

<p>I down sampled Les's canoe shot to 6mp saved it as a tif closed the image, reopened it upsampled to original size and compared it to the original image. I can't any find any differences between the two even pixel peepling at 100% in PS. Now it could be that the jpg compression has destroyed fine details that are there in the original scan but for me the image just does not look that sharp.</p>

[steve m smith]

<blockquote>

<p>Here's a 12MP image file... is it from a 12MP dSLR or a 12MP scan of a 35mm frame of film?</p>

</blockquote>

<p>The answer is yes.</p>

[StuartMoxham]

<p>Rishi you are using HTML to rescale your large images and display them inline. That means the browser is downloading a 12 mp image and having to resize it. It slows down scrolling and makes the browser sluggish. I would be better to rescale in PS and display that image inline.</p>

[swilson]

<p>Here is a close up of the top of Rishi's photo</p><div></div>

[swilson]

<p>Rishi,</p>

<p>On the mt hood photo I am guessing film, but I could well be wrong. It kind of looks like you did what I have done with film scans where the sky had more grain then I like, selected the sky and reduced the noise by a whole lot.</p>

<p>As I said this is just a guess.</p>

[pankaj purohit]

<p ><a href="../photodb/user?user_id=2381463">Rishi Sanyal</a> <a href="../member-status-icons"><img title="Frequent poster" src="http://static.photo.net/v3graphics/member-status-icons/2rolls.gif" alt="" title="Frequent poster" /> </a> , Mar 19, 2009; 04:15 a.m.<br>

OK, I'll make this one easy for ya, folks:<br>

Here's a 12MP image file... is it from a 12MP dSLR or a 12MP scan of a 35mm frame of film?<br>

 <br>

Answer : Digital Image..</p>

[rishij]

<p>Stuart -- shoot you're right it does slow down scrolling. Sorry, not much I can do now other than kill the link. Which I'll do after I get a few more votes, how does that sound? :)</p>

<p>Thanks Scott & Pankaj for the votes... that's 2 down, 2 to go!</p>

<p>Really, I want Mr. David Littleboy's vote... y'know, the guy who thinks that a 12MP dSLR surpasses 645, while scoffing at 35mm. C'mon man... which one of these is a <em>dSLR</em> & which one's just measly ol' <em>35mm-unworthy-of-landscapes</em> ?</p>

<p>-Rishi</p>

[rishij]

<p>P.S. Steve Smith your answer is about as ambiguous as a hermaphrodite's sexuality.</p>

[steve m smith]

<blockquote>

<p>P.S. Steve Smith your answer is about as ambiguous as a hermaphrodite's sexuality.</p>

</blockquote>

<p>It's the question that is ambiguous. It's like asking if a light switch is on or off. The correct answer is yes. The question you ask to get the response you want is "which position is the light switch set to?".<br>

Possibly far too pedantic. A bit like when my wife asks me: "do you want to make a cup of tea?" My answer is always "no" even though she's not really asking a question.</p>

<p> </p>

[steve m smith]

<p>Back to normality. The detail in the highlights in the mountain make me think it's a film image but the foliage of the foreground trees make me think digital or film not handled well by the scanner.</p>

<p>If I had to choose I would say film.</p>

[rishij]

<p>Stuart, fair points re: the composition. I would actually process the file more like this than what was posted above (too many blown highlights, I think):</p>

<p><img src="http://staff.washington.edu/rjsanyal/Photography/FairyFalls_reprocessed.jpg" alt="" /></p>

<p>Regardless, though, not too much shadow detail to work with...</p>

<p>Cheers,<br /> Rishi</p>

[rishij]

<p>Steve: the problem with the foliage detail either not captured well digitally or not scanned well by the scanner may result from the fact that this was a 30 second exposure on a gratuitously windy morning in the Columbia River Gorge area...</p>

<p>-Rishi</p>

[james_symington1]

<p>I'll have a go: the waterfall is a DSLR picture and the mountain 35mm with a good dose of noise reduction in the sky. Fingers crossed.</p>

[radek_pohl]

<p>imho, both waterfall and mountain are scanned film images</p>

[ilkka_nissila]

<p>The problem with doing drum scans is not the cost of individual scans as priced by a lab. It's that the scanner is too expensive to buy for personal use and commercial labs don't have operators that care one bit about the photographer's intentions. I couldn't care less if a 12000ppi scan of my film has more detail than my LS-9000 scans if the colors, tone curve, sharpening etc. are botched up by the operator that wants to be somewhere else and won't listen. That's why my LS-9000 is a better option as it allows me to have full control and wasn't too expensive for my limited number of images. An Epson flatbed makes 6x7 scans slowly and the image quality is hardly better than that from 6 MP DSLRs, and the process is far slower than using the Nikon. With the Nikon, color accuracy out of the box is far better. To each their own. I wouldn't use a lab made scan if a gun was pointed to my head.</p>

[matt_needham]

<p>"Poor scans may be the reason why some people believe DSLRs compare to scanned medium format film..."</p>

<p>In my case it's because I have large prints from both DSLRs and 6x7 Tmax 100 (not scanned; old fashioned, hand printed, optical prints) hanging next to each other on my walls. I don't bother with extreme magnification or worry about the math; I just view them with my eyeballs from about 4 or 5 feet away.</p>

[matt_needham]

<p>"...else you are just eyeballing the results, not really measuring them..."</p>

<p>Ha! That's hilarious! Guess what? Most of the world is only concerned about how prints look when being "eyeballed". </p>

<p><em>"Pictures, regardless of how they are created and recreated, are intended to be looked at. This brings to the forefront not the technology of imaging, which of course is important, but rather what we might call the eyenology.</em> " -Henri Cartier-Bresson</p>