Poor scans - May explain why some switch to Digital

<p>True, Mauro. And, like I said, in our free time we should build an 8000ppi MF/35mm desktop scanner :) I certainly have a couple ideas of how, and seeing how the new $100-$200 'digitize your memories!' film scanners these days flooding the market scan at some pathetic resolution of 1-2MP, I don't have any confidence that any company is working on releasing such an 8000ppi scanner in the future :(</p>

<p>For those interested in judging 'pixel sharpness' of film scanners in comparison to dSLR, here's a link to a full 20MP version of the Fairy Falls scan (originally 80MP straight out of the Imacon 848). It's only 17.1MP because I cropped it quite a bit (the damn shot wasn't straight, argh); trust me, it's equivalent to a 20MP scan, just cropped.</p>

<p><a href="http://staff.washington.edu/rjsanyal/Photography/FairyFalls_20MP.jpg">http://staff.washington.edu/rjsanyal/Photography/FairyFalls_20MP.jpg</a></p>

<p>I'd say that parts of it do look as 'pixel-sharp' as what a 20MP dSLR might put out (i.e. 5D Mark II, 12.6MP true resolution), but, then again, parts of it don't. Certainly, if you resize properly to 12MP, it looks 'pixel-sharp' as a dSLR across the frame (save for motion & lens effects). My guess is that it looks somewhere in between an equivalent 5D & 5D Mark II capture...</p>

<p>Mauro, you've looked at quite a bit of scanned 35mm & MF, using exquisite technique far as I can tell, so I'm particularly curious about your evaluation of the sharpness when viewed at 20MP size 1:1...</p>

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

<p>P.S. I'm quite confident the file linked above would print pretty well at 16x20... if anyone wants to try, be my guest, and let us know :)</p>

<p ><em ></em></p>

<p > <br>

 

<p ><em >Digital for their purposes is "good enough."</em></p>

<p >Well, but seems to me they just don’t have another choice…</p>

<br>

 

<p ><em >This will change if we can evolve intelligently to a more sustainable, and healthy economy, and lifestyle;</em></p>

<p >That’s true, Jeff. I think that the intelligence it’s what completely missed in our economy and unfortunately in someone’s lifestyle… </p>

</p>

<p >Rishi:<br>

 

<p ><em >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.</em></p>

<p >That’s exactly what I don’t like in binary digit theory. to store each analog signal you have to create infinity series of digital values which is nothing but combination of 1s and 0s. If my brain has stored all my knowledge, experience, pictures and music this way it would explode already. Such way of representing analog signals required enormous amount storage and processing. the film delivers data much more natural way – it’s just changing the properties of substances and represent analog signal (i.e. color) the way it naturally exists, without sampling it. That’s why we’re designing with our brain and use computers only for calculations. And I dont think it’s good for anything else. Obviously you might have different opinion. </p>

<p >With all respect to you, Rishi, but I’m going to stay where I am and not ashamed of what I said.</p>

<p >Now I have a question to you. I never used Velvia 50 to shot waterfall. I use Velvia 100F for this purpose. Is it old Velvia 50 or the “new” one?</p>

<p > </p>

</p>

<p > </p>

<p > </p>

<p>Well I guess I think the infinite series is 'good enough'... :) Well, not in all cases. For example I wouldn't sample music at a rate of 22kHz... if we hear up to 20kHz, you sample at least twice that frequency, and as you go to higher and higher sampling frequencies, it becomes hard to distinguish the analog signal from the digital representation of it. Digital cameras essentially are tryina do the same thing. That's all I was saying.</p>

<p>If you accept CDs, you should accept that digital image capture is or will get there just fine.</p>

<p>Apologies if I sounded offensive.</p>

<p>Re: the waterfall shot -- it's the new Velvia 50. I don't like 100F at all, for anything. Too bland. If I wanted blanded I'd shoot digital RAW :)</p>

<p>That being said, some color shifts with Velvia 50 and 100 (not 100F) can be annoying. Long exposures tend to show green shifts... perhaps the blue of the waterfall is also due to such a color shift? I don't really know. The scan is a pretty accurate representation of the slide, since I used a Hutch-color-target generated profile. Minus the slight diffuse glow I added, of course.</p>

<p>-Rishi</p>

<p>Hmmm... I think Mauro may be right & I've been short-selling myself here.</p>

<p>Take a look at these full-size JPEG crops from a 5D Mark II... at 1:1 viewing, their 'pixel sharpness', to my eyes, looks on par with the 'pixel sharpness' of my Fairy Falls 20MP scan also at 1:1 viewing, don't you think?<br>

<a href="http://web.canon.jp/imaging/eosd/eos5dm2/downloads/3_nightscene.jpg">http://web.canon.jp/imaging/eosd/eos5dm2/downloads/3_nightscene.jpg</a><br>

<a href="http://web.canon.jp/imaging/eosd/eos5dm2/downloads/2_landscape.jpg">http://web.canon.jp/imaging/eosd/eos5dm2/downloads/2_landscape.jpg</a></p>

<p>Perhaps the latter 5D Mark II file looking somewhat sharper. Now I'm thinking that in an objective test, 35mm may compete quite well with a 5D Mark II... man I'd really like to find out!<br>

-Rishi</p>

<blockquote>

<p ><a href="http://www.photo.net/photodb/user?user_id=3995956">Mauro Franic</a> <a href="http://www.photo.net/member-status-icons"><img title="Frequent poster" src="http://static.photo.net/v3graphics/member-status-icons/3rolls.gif" alt="" /></a>, Mar 19, 2009; 09:14 p.m.</p>

 

<p>Scott, what scanner did you use for the slide you posted?</p>

 

</blockquote>

<p>That was a Minolta DiMAGE SCAN Dual III, or perhaps my older HP Smart Scanner (long gong now).<br>

 

<p>In either case the scanner was not stellar but it could get much better looking scans then that, if the film was in good shape. The odd thing about that slide is there is likely more detail there then the scanner can capture, but the grain is so high I can even make a decent screen size image from it. Even downsizing to 1200 pixels wide this image looks like crap.<br>

 

<p>This is an extreme case where a photo can have a good amount of detail, but still look very soft. I the attached photo I have resized a photo from my old Nikon 995 to match the height of the scanned image and put a crop from each side by side. It is pretty clear to my eye that the film scan has captured way more detail, which it should have since the digital is only 3.2 MP But the film photo can make a sharp looking 4x6 in print whereas the digital makes a beautiful 4x6 inch print.<br>

This link shows the two photos both sized to a width of 1200 pixels, resized the digital image looks far sharper, this is roughly what you would see on a 4x6 inch print.<br>

<a href="http://www.pbase.com/konascott/image/110409845/original">http://www.pbase.com/konascott/image/110409845/original</a></p>

<p>

<p>Thankfully the roll of film that slide is from is not the norm, in no way are these image meant to show that a 3MP camera is better then 35mm film What I am showing is that detail is not all there is to what makes an image look sharp, therefore photos of test chart are of limited value.</p>

</p>

 

<p>

 

</p>

 

<p>

<p> </p>

</p>

 

<p>

<p>

<p> </p>

</p>

</p>

<br>

 

 

</p>

 

</p>

<div></div>

<blockquote>

<p ><a href="http://www.photo.net/photodb/user?user_id=3995956">Mauro Franic</a> <a href="http://www.photo.net/member-status-icons"><img title="Frequent poster" src="http://static.photo.net/v3graphics/member-status-icons/3rolls.gif" alt="" /></a>, Mar 19, 2009; 06:54 p.m.</p>

 

<p>I think scanned 35mm film (at least the ones I use) have better tonality than a 10MP DSLR. Roughly 20 million full RGB samples vs 10 million with just a 1/3 of a full RGB sample.<br>

Grain can always be removed and if you even downsample a 20MP 35mm scan to just 10MP no print would be able to show grain.<br>

Let me know what you whink of the results printing the links above at 16x20. The danger of knowing though.... is that once you know..... side by side.... you may go back to film (not your quick workflow work but the things you may consider placing on the wall at some point).</p>

 

 

</blockquote>

<p>I am not sure when I am going to get more ink, with luck in the next day or two.</p>

<p>Not much chance of me ever going back to 35mm, just does not cut it for large print IMO. This is what I would use for a 20x30 inch print (a nice size for the wall)<br>

<a href="http://sewcon.com/samples/30x20_inches_at_300ppi.jpg">http://sewcon.com/samples/30x20_inches_at_300ppi.jpg</a><br>

Print out some of your image at 20x30 and some of my and I think you will see why 35mm just does not hold much interest for me.</p>

<p>Assuming you scan the full 36x24mm area of 35mm film at 4000 ppi you would end up printing a 20x30 inch print at 198 ppi, so to match my image yours need to be scale up by 158.75%, I have put a crop in showing the two, print out at 300ppi to see what I mean.</p>

<p>Sure I had to stitch image to get to this point, but to get a good sharp image on film you really should use a tripod, and once the camera is on the tripod it is very easy to get the images needed to stitch.</p>

<p>

<p>If I was ever going to do film again it would be LF, but I don't like BW and there is no place I can drive to that will develop sheet film. And I can get whatever resolution I want anyway so I am not sure I would do it even if there was a place here that would develop the film.</p>

<p>As for tonality, it is the grain the limits in for film, which is way MF has such better tonality then 35mm. Like it or not digital has great tonality.</p>

</p><div></div>

<blockquote>

<p>What I am showing is that detail is not all there is to what makes an image look sharp, therefore photos of test chart are of limited value.</p>

</blockquote>

<p>Right, hence we talk of 'accutance'.</p>

<p>Actually, Scott, your examples go to show that a lot of those comparisons that people have done of how a digital SLR clearly beats 35mm, even MF, film are, in fact, invalid.</p>

<p>A cursory glance at these comparisons on the 'net with the film scan simply <em>done by a 'professional lab'</em> or just <em>run through an old film scanner the tester had lying around</em> <strong>are not valid</strong> . Because, as I've shown with the Fairy Falls picture, with enough care, attention to detail, high-resolution scanning, noise reduction, and selective sharpening, one can make a 20MP scan from 35mm that has quite sharp pixels that look 'digital SLR-like' at 1:1 viewing (as voted 5:2, so this is not just my opinion).</p>

<p>So, I think you just helped prove Mauro's point :)<br>

-Rishi</p>

<p>Scott-- what the heck? That's not a 1:1 crop of the leaf from the Fairy Falls photo! It's almost a 2:1 view of it. I certainly never claimed that Velvia 50 35mm is 40MP. It's entirely unfair to have that blow-up next to a 1:1 crop of your <strong>stitched image</strong> .</p>

<p>Not only are you not comparing apples to apples, you're more comparing apples to... I dunno, elephants?</p>

<p>Your stitched images are amazing, I admit. My super-resolution images with PhotoAcute blow me away too. Both are irrelevant to <strong>one</strong> film scan of <strong>one</strong> image. Especially when you show your digital image at 1:1 and then my film scan at 2:1, just to make the comparison look worse.</p>

<p>If you wanna do a fair comparison, take one of your 40MP images and stack it up 1:1 against Mauro's MF scan. My guess is that Mauro's MF scan will hold up. Certainly not my 35mm film scan. That's ludicrous to even have placed the two side-by-side. Not sure what you were trying to get at there.<br>

-Rishi</p>

<blockquote>

<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="" /></a>, Mar 20, 2009; 02:07 a.m.</p>

 

 

<blockquote>

<p>What I am showing is that detail is not all there is to what makes an image look sharp, therefore photos of test chart are of limited value.</p>

</blockquote>

<p>Right, hence we talk of 'accutance'.<br>

Actually, Scott, your examples go to show that a lot of those comparisons that people have done of how a digital SLR clearly beats 35mm, even MF, film are, in fact, invalid.<br>

A cursory glance at these comparisons on the 'net with the film scan simply <em>done by a 'professional lab'</em> or just <em>run through an old film scanner the tester had lying around</em> <strong>are not valid</strong> . Because, as I've shown with the Fairy Falls picture, with enough care, attention to detail, high-resolution scanning, noise reduction, and selective sharpening, one can make a 20MP scan from 35mm that has quite sharp pixels that look 'digital SLR-like' at 1:1 viewing (as voted 5:2, so this is not just my opinion).<br>

So, I think you just helped prove Mauro's point :)<br />-Rishi</p>

 

</blockquote>

<p>

<p>My point is that simply quoting line pair/mm or the smallest text that can be read does not tell the whole story on how sharp a print is going to look. Some on this thread seem to believe that is all you need to know.</p>

<p>As for your Fairy Falls photo, there is enough odd stuff going on that it is hard for me to really judge it. I looks like some of this might have been from leafs moving during the exposure, but hard to know.</p>

<p>The detail in the moss looks pretty good, not 20MP worth but not at all bad.</p>

<p>I have always figured that for a well done scan 35mm and digital are a pretty close match, and the line between them is fuzzy enough and subjective enough that there will be disagreements.</p>

<p>Within the digital comunity that are disagrments as to how large you can print with a given camera, in the film comunity there are disagreements as to how large you can print with 35mm before you should switch to MF. I tend to be on the side of wanting really good images for large print, which means either MF or stitched digital.</p>

<p> I am holding off judment on how large the New Sony or 5D II can print untill I see some more photos from them. I would not think either could go to 20x30 without starting to look soft, but I know 35mm film is going to look way soft at 20x30.</p>

<p>I really don't understand why dedicated film shooters who care about making large prints would mess around with 35mm.</p>

<p> </p>

 

<p>

 

</p>

</p>

 

<blockquote>

<p ><a href="http://www.photo.net/photodb/user?user_id=2381463">Rishi Sanyal</a> <a href="http://www.photo.net/member-status-icons"><img title="Frequent poster" src="http://static.photo.net/v3graphics/member-status-icons/2rolls.gif" alt="" /></a>, Mar 20, 2009; 02:19 a.m.</p>

 

<p>Scott-- what the heck? That's not a 1:1 crop of the leaf from the Fairy Falls photo! It's almost a 2:1 view of it. I certainly never claimed that Velvia 50 35mm is 40MP. It's entirely unfair to have that blow-up next to a 1:1 crop of your <strong>stitched image</strong> .<br>

Not only are you not comparing apples to apples, you're more comparing apples to... I dunno, elephants?<br>

Your stitched images are amazing, I admit. My super-resolution images with PhotoAcute blow me away too. Both are irrelevant to <strong>one</strong> film scan of <strong>one</strong> image. Especially when you show your digital image at 1:1 and then my film scan at 2:1, just to make the comparison look worse.<br>

If you wanna do a fair comparison, take one of your 40MP images and stack it up 1:1 against Mauro's MF scan. My guess is that Mauro's MF scan will hold up. Certainly not my 35mm film scan. That's ludicrous to even have placed the two side-by-side. Not sure what you were trying to get at there.<br />-Rishi</p>

 

</blockquote>

<p>I was giving an answer to your statment that I might decide to shoot 35mm, for photos that I plan to make large prints of and hang on the wall. I was just showing that I have no need for anything past what I already have. For me to shoot 35mm for large prints it would have to do way better then it does.</p>

<p>If you dont' mind all the scanning film can produce great lookint 8x12 inch prints, but that is about as far as I would use it, I also don't like to go past 8x12 with my digital. These sizes are pretty subjective, but once you get to something like an 20x30 print the lack of sharpness in either a 35mm print or digital becomes pretty clear.</p>

<p>Scott-- here's a more fair, yet still unfair, comparison of your stitched shot vs. my film scan:<br>

<img src="http://staff.washington.edu/rjsanyal/Photography/MoreFairComparison-Scott_vs_Rishi.jpg" alt="" /><br>

Still, unfair because any multiple-imaging method typically averages pixel data to generate even cleaner images... your image is much much cleaner than a 1:1 viewing of a RAW file out of my Panasonic LX3... so, something's going on there... either multi-image sampling or your actual resultant image was bigger and you downscaled it to give us the crop you show here... which, I might add, would be<em> really unfair</em> to place next to an upsampled film scan.</p>

<p>Actually, now I'm curious: is that really a 1:1 of the final stitched image, or is it downsampled from the final stitch?</p>

<p>Furthermore, digital camera have better accutance partially, aside from the fact they have lower noise, because they have increased microcontrast. Film is much more 'leaky' from one area of grain to another. This is fixed by proper sharpening methods. Digital still retains better microcontrast in small details b/c its MTF falloff is much more sharp than film's gradual MTF drop-off as you near extinction resolutions (in my understanding).</p>

<blockquote>

<p>The detail in the moss looks pretty good, not 20MP worth but not at all bad.</p>

</blockquote>

<p>OK, thanks for the opinion. I would also agree that it doesn't quite look like 20MP throughout the frame; however, some parts, like the rocks+moss, do look about as pixel-sharp as the 5D Mark II images I linked to. So it's up in the air. Your opinion is registered though :)</p>

<blockquote>

<p>I really don't understand why dedicated film shooters who care about making large prints would mess around with 35mm.</p>

</blockquote>

<p>Because:</p>

<ul>

<li>I've been shooting with it since I was kid... for the past 12 years. I have a lot of precious shots. I want them scanned, and some of 'em enlarged. Back in the day, 35mm was still state-of-the-art for what was affordable to me & my family at the time!</li>

<li>I don't really print larger than 13x19, on my Epson R2400. For a well scanned, cleaned, and sharpened 35mm frame, it really is quite impressive.</li>

<li>The versatility of Canon L-series lenses & 35mm bodies is unparalleled, save for by the same L-series lenses and their full-frame digital bodies. I have a 17mm-280mm zoom range, with graduated neutral density adapters, polarizer, etc. that all fit in a small backpack that I can whip out and place on my Gitzo tripod in a matter of seconds. And it all doesn't weight my small 130 pound frame down all too much, so I can hike with other gear or just actually enjoy my hike. You can't get that versatility with MF (of course, you could crop instead of zoom, but that's another story).</li>

<li>Investing in 35mm equipment, e.g. lenses and all, will allow me to step up to full-frame digital very easily. <em>Which I plan to do very soon</em> :)</li>

</ul>

<p>-Rishi</p>

<p>Also, as you can see, with the long exposure required to get the water-blur effect, I already have tons of leaves/branches blurred. Imagine taking a bunch of 30 second shots to stitch together, with those blurred elements. Stitching software would be hella confused, not to mention the time for all those shots at 30 seconds each. And, like I said before, sometimes nature (on the verge of changing light) doesn't give you all that time in the world :)</p>

<p>That being said, sure, I wish I'd invested in MF a long time ago!<br>

-Rishi</p>

<p>From the amount of cropping I have done from medium format, a 6x6 exposure can give somewhere close to 50MP resolution straight. In the current technology, you would need a 150MP sensor to do that. As far as larger formats are concerned, do the maths. I admit some films may have higher grains and all that stuff, but common... get the point here? Good luck fighting...</p>

<blockquote>

<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="" /></a>, Mar 20, 2009; 02:42 a.m.</p>

 

<p>That being said, sure, I wish I'd invested in MF a long time ago!</p>

 

 

</blockquote>

<p>You and me both. When I was shooting film, for many years, I mainly shot slides since I could not affort large prints ( it seems they cost a lot more 30 years ago)</p>

<p>There was no way I could afford a good MF slide projector, so I stuck with 35mm, or went too 35mm I should say since the camera I had as a kid was a cheap TLR 645.</p>

<p>For most of my film shooting days there was no thought of scanning, what could you do with 20-40 mbytes of image data when a 5 mbyte harddrive was running around $5,000.</p>

<p>But now I have all these photo from 35mm that I know could have looked far better if I had used MF. For me it is not some much about the extra detail in MF as it is the out clean of an image MF can deliver compared to MF.</p>

<p>Oh and if I had it to do over I would not have shot anything by kodachrome. </p>

<blockquote>

<p ><a href="http://www.photo.net/photodb/user?user_id=2399881">Debejyo Chakraborty</a> <a href="http://www.photo.net/member-status-icons"><img title="Subscriber" src="http://static.photo.net/v3graphics/member-status-icons/sub3.gif" alt="" /></a>, Mar 20, 2009; 02:45 a.m.</p>

 

<p>From the amount of cropping I have done from medium format, a 6x6 exposure can give somewhere close to 50MP resolution straight. In the current technology, you would need a 150MP sensor to do that. As far as larger formats are concerned, do the maths. I admit some films may have higher grains and all that stuff, but common... get the point here? Good luck fighting...</p>

 

</blockquote>

<p>I see 50MP for 6x6 and the alarm bells go off, sounds like you are scanning at 3200ppi, which sure sounds like a flatbed scanner to me. Go luck getting those 50MP to be sharp.</p>

<p>"If you dont' mind all the scanning film can produce great lookint 8x12 inch prints, but that is about as far as I would use it,"</p>

<p>Yep. That's why I didn't shoot 35mm; at even 11x14 MF looks way better. Before the 5D, the effort of scanning medium format seemed very much worth it.</p>

<p>" I also don't like to go past 8x12 with my digital."</p>

<p>Agreed again, for 8MP dSLRs. But the 5D is a different story; it holds up at 12x18 in a way that previously required 645 (and the 5DII goes one step beyond that). Rishi seems to be living in an alternate universe where Velvia doesn't have grain and film always magically lies flat in the scanner.</p>

 

<blockquote>

<p>Rishi seems to be living in an alternate universe where Velvia doesn't have grain and film always magically lies flat in the scanner.</p>

</blockquote>

<p>Nope, I'm just tryina create the world where it does lie flat in a scanner. Some of us actually build $hit. I believe they call us 'engineers'.</p>

<p>These so-called 'engineers' also resort to <em>measurements</em> and <em>objective analyses</em> , both of which have shown 35mm Velvia scanned properly to out-resolve a 5D. My real-world 35mm scans above also held up at 10-12MP sizes at 1:1 viewing.</p>

<p>So it is you, my friend, living in an alternate universe -- the one most people oblivious to methodical testing live in. Step up your game, or shut it.<br>

-Rishi</p>

<h2></h2>

<p>I know I'm going to get a barrage of counter arguments for suggesting this but shouldn't we take scanning out of the equation when comparing film and digital prints?</p>

<p>Film was always supposed to be printed optically. Therefore a fair test would be a totally optically produced print vs. a digitally produced print.</p>

<p>The scanner is usually the very weak link in the process and is a modification/compromise of the film process.</p>

<p>I am aware that the original intention of this thread was to discuss scanning quality.</p>

<p>"The scanner is usually the very weak link"</p>

<p>People who have actually compared wet projection prints to scans have found them to be very similar. Having done both darkroom and digital prints (although at very different times), I'm not surprised. This page has some examples of scanned film vs. scanned optical prints.</p>

<p><a href="http://www.terrapinphoto.com/jmdavis/">http://www.terrapinphoto.com/jmdavis/</a></p>

 

<p>I have also seen some very good scanned images which compete favourably with optical prints but at a general consumer level, the scanner is still the weak link.<br>

A lot of these comparisons are flawed (I'm not saying that your specific link is) as the very best possible from one system is compared with the average results from another.<br>

To get a valid comparison either the very best of both systems should be compared or the industry standard of both should be compared. And preferably not via the medium of a computer monitor.<br>

Or even better - lets just go out and take pictures of things we like and print them any way we please without worrying about how others are doing it or how others think we should be doing it.</p>

 

<blockquote>

<p>"Nope, I'm just tryina create the world where it does lie flat in a scanner." --Rishi Sanyal</p>

</blockquote>

<p>Well, I don't build things (except for boats), and so I have to try to find other solutions. I just looked up Aztek starter kits: $639 + exorbitant shipping = close to $700.</p>

<p>I have had this recommended to me. All that I can say is that, on top of the cost of the old Hassy gear, the Nikon 9000 scanner, glass holders, and Silverfast with two targets, I am already in this up to my neck--and now another jolt of near $700??</p>

<p>Before I even begin to think about ordering from Aztek to try wet mounting, I have to know: will it give near drum scanner quality, as Ellis Vener says? I trust Ellis, but at these prices I need a second opinion.</p>

<p>--Lannie</p>

<p > </p>

<blockquote>

<p ><a href="../photodb/user?user_id=940372"><strong>Scott Wilson</strong></a><strong> </strong><a href="../member-status-icons"></a><strong>, Mar 19, 2009; 06:16 p.m.</strong></p>

<p > </p>

<p >Let's see, a frame of 35mm being the same as having a 40 MP camera? So you should be able to show some 5400 ppi scans that are as sharp as the pixels level as my DSLR?</p>

</blockquote>

<p >Shot on Kodachrome25 26 years ago with Canon A-1 and Canon FD 35-105mm f3.5 zoom, handheld.</p>

<p > </p>

<p ><img src="http://www.lexharris.net/documents/02-11-E4small.jpg" alt="" width="640" height="434" /></p>

<p > Straight off scanner (Imacon 646, 6300dpi), no sharpening, no post processing.

<p >Full size image (8588 x 5826) here:</p>

<p ><a href="http://www.lexharris.net/documents/02-11-E4c.jpg">http://www.lexharris.net/documents/02-11-E4c.jpg</a></p>

<p > </p>

<p > </p>

<p > </p>

<blockquote>

<p ><a href="../photodb/user?user_id=940372"><strong>Scott Wilson</strong></a><strong> </strong><a href="../member-status-icons"></a><strong>, Mar 20, 2009; 03:05 a.m.</strong></p>

<p >Oh and if I had it to do over I would not have shot anything by kodachrome.</p>

</blockquote>

<p > See above. Actually, if I had to do it over again I would have shot MORE K25 :-)</p>

<p > </p>

<p > </p>

<p > </p>

<p > </p>

<blockquote>

<p ><a href="../photodb/user?user_id=940372"><strong>Scott Wilson</strong></a><strong> </strong><a href="../member-status-icons"></a><strong>, Mar 20, 2009; 03:14 a.m.</strong></p>

<p >I see 50MP for 6x6 and the alarm bells go off, sounds like you are scanning at 3200ppi, which sure sounds like a flatbed scanner to me. Go luck getting those 50MP to be sharp.</p>

</blockquote>

<p> It sounds like you are not familiar with Imacon scanners. Imacon scanners can scan all 120 film sizes at 3200ppi and 645 portrait at 4000ppi. In my experience they do not suffer from focus problems.</p>

</p>

<blockquote>

<p ><a href="http://www.photo.net/photodb/user?user_id=1787762">Steve Smith</a> <a href="http://www.photo.net/member-status-icons"><img title="Frequent poster" src="http://static.photo.net/v3graphics/member-status-icons/1roll.gif" alt="" /></a>, Mar 20, 2009; 04:05 a.m.</p>

 

<p>I know I'm going to get a barrage of counter arguments for suggesting this but shouldn't we take scanning out of the equation when comparing film and digital prints?<br>

Film was always supposed to be printed optically. Therefore a fair test would be a totally optically produced print vs. a digitally produced print.<br>

The scanner is usually the very weak link in the process and is a modification/compromise of the film process.<br>

I am aware that the original intention of this thread was to discuss scanning quality.</p>

 

</blockquote>

<p>Over the years I have heard this a lot, and the solution is so easy. If the print really had more detail then a scan then just make a large enough print so that a fllatbed scanner scan capture all the detail in the print, this is very easy to do.</p>

<p>

<p>This may be antidotal but in going back to old negatives, where I had optical prints done originally, I can get far better print by scanning the negatives and printing from the image file. In most cases the optical print was simply not made well, but unless you do your own prints this seems to happen more time then not. In this link you can see the scanned image compared to what the print looked like, the print is the blurry insert in the photo.</p>

 

 

<a href="http://www.pbase.com/konascott/image/36102779/original">http://www.pbase.com/konascott/image/36102779/original</a>

<p>That is an extreme case for sure. But the only time the optical print has been sharper then a print make from a scan has been when the negative died over the years.</p>

<p>As I said this is antidotal, but I would really like to see some example where the optical print had more detail and looked sharper then the film scan.</p>

</p>

<blockquote>

<p>As I said this is antidotal, but I would really like to see some example where the optical print had more detail and looked sharper then the film scan.</p>

 

</blockquote>

<p>In an ideal world they would be the same. The negative (or transparency) holds all of the detail and either system should be able to bring most of it out.</p>

<p>I would expect a professionally made optical print to be superior to anything made in a high street lab just as I would expect an expertly processed drum scan to be superior to anything made at home on a domestic scanner.</p>

 

<blockquote>

<p>Over the years I have heard this a lot, and the solution is so easy. If the print really had more detail then a scan then just make a large enough print so that a fllatbed scanner scan capture all the detail in the print, this is very easy to do.</p>

</blockquote>

<p>Why? Surely if you have a print already, you don't need a scan. I thought the purpose of photography was to produce prints. I don't understand the obsession with getting everything onto a computer. Especially the regular "I have 3000 + slides which I need to digitise" threads which we get.</p>

<p>To me, it doesn't matter how good your scans are until you output them to something. If your output is just a computer monitor, TV or digital projector then high resolution and dynamic range are wasted. It's only at the printing stage that these parameters start to become important.</p>

 

<p>Lex Harris,<br>

 

<p>You are not trying to say that you think your scan is even to be as sharp at the pixel level as a DSLR are you? You put up your image where I was asking for a 5400 ppi scan that was as sharp at the pixels level as my DSLR. So ok your scan is a 6400ppi, but it is a world away from being sharp at the pixels level, kind of bigfoot photo blurry, I have attached a comparison to a 100% crop from by DSLR.</p>

<p>Try this with your image, down size to 50% and then back up, do you see any loss at all in detail? That scan does not even have good detail when down sample to 50%</p>

</p><div></div>