Jump to content

Poor scans - May explain why some switch to Digital


Recommended Posts

  • Replies 799
  • Created
  • Last Reply

Top Posters In This Topic

<p>Thank you Rishi.</p>

<p>It confuses me when people argue that film (35mm and 6x7) does not outresolve a particular DSLR when I post the entire set of results for them to see.</p>

<p>The responses go from "look at this link, this guy gets much worse results than you (ha ha)" or "the differences can't be seen on print" or "ok, but I don't care". Very strange.</p>

<p> </p>

Link to comment
Share on other sites

<p>And for those of you complaining about noise in the original MF scan of the maps in the original post, here, take a look again at the comparison after some noise removal using Neat Image (and if you think digital looks so clean from the start, that's because you didn't work on the evolved noise reduction algorithms in CMOS sensors -- read up on them in the white papers):</p>

<p><img src="http://staff.washington.edu/rjsanyal/Photography/Film_vs_Digital-Map.jpg" alt="" /> <br /> <a href="http://staff.washington.edu/rjsanyal/Photography/Film_vs_Digital-Map.jpg">Link to Full-Size Image</a></p>

<p>And that's sub-optimal noise reduction b/c the software typically needs much more area to analyze but I just used the crop posted in the original post for a quick-n-dirty run. You can do much better by using your own noise-reduction profiles. Or by not using Kodak UC film, which I hate due to its very offensive grain structure in comparison to Ektar 100 or Velvia slide.</p>

<p>-Rishi</p>

Link to comment
Share on other sites

<p>OMG, look at the crops. I just realized that the blue river starting near Port Harcourt becomes black in the Nikon digital capture.</p>

<p>Actually, a bunch of the blue lines just become black in the digital capture.</p>

<p>HA! That's pathetic.<br>

-Rishi</p>

Link to comment
Share on other sites

<blockquote>

<p>Below is the D2X full frame coverage. Again, the RZ67 coverage way overlapped the map giving the D2X added advantage covering less vertical space as in the link that C.Sharon provided.</p>

</blockquote>

<p>Um... no it didn't. The D2X just has a different aspect ratio. In fact, the D2X covered more horizontally than the RZ67, whereas the RZ67 covered more vertically.</p>

<p>David Littleboy was totally appropriate in calling you out, Les.<br>

-Rishi</p>

Link to comment
Share on other sites

<p>Scott Wilson: your images are ridiculous... as in ridiculous eye candy. I have to say '<strong><em>wow!</em> </strong> ' to every single image you link to :)<br /> <br /> My only guess as to how you do it is by using some combination of multi-image technology... multiple image stitching and/or super-resolution via PhotoAcute? The latter can do quite an amazing job at interpolating resolution and getting rid of noise even with my 10MP Panasonic Lumix LX3.</p>

<p>Short of that, my guess would be some home-brewed piezo-technology... are you gonna tell us or are you gonna keep us guessing? </p>

<p>:)</p>

<p>-Rishi</p>

Link to comment
Share on other sites

<p>Rishi, the problem with noise reduction software is that often there is texture in the photo that is going to look like noise to the software.</p>

<p>In the case of this map this might not be an issue but for most images there are textures that I would like to see as they really were.</p>

<p>Try removing the noise of the film scan with out removing the texture of the book next to it. Sure you could select just the map part, but if the book had been taken with film there would be no way to see the texture in the binding.</p>

<div>00Smjl-117069684.jpg.df33f6bc730df3664ead764e00d54069.jpg</div>

Link to comment
Share on other sites

<p>Hi <a href="../photodb/user?user_id=940372">Scott Wilson</a> , here something for you, my first post was just a flat scan without any postprocessing except color balancing from a RAW scan, now this is real 4800 dpi scan of a 35 mm 400 ISO Kodak negative with little postprocessing (Actual image size is <strong>6691X4391 pixels</strong> ) :</p><div>00Smk4-117071584.thumb.jpg.4d35d7da3296286e311ada62a9130087.jpg</div>
Link to comment
Share on other sites

<p>Les, what I'm saying is that the Nikon has a larger horizontal FOV, while the MF has larger vertical FOV. Sure, you can hold one constant and let the other vary with different aspect ratio cameras. That's irrelevant.</p>

<p>My point is, you can't say just from looking at the full crop comparisons that the Nikon had an advantage because it's 'zoomed in' or such.</p>

<p>Scott -- you're absolutely right. Which is why now after using 'multi-sampling' and PhotoAcute to get rid of noise with digital images, it's hard for me to go back to my film camera. A similar approach could be applied to film, but only very low grain film, b/c these techniques don't get rid of film grain.</p>

<p>-Rishi</p>

Link to comment
Share on other sites

<p>On grain reduction:</p>

<p>I don't use it often (or at all) for prints since the grain doesn't show on 24" and the texture is richer. But it is definetily an option for people (both film and digital) who want to include it in their workflow as a routine.</p>

Link to comment
Share on other sites

<p>Hello<a href="http://www.photo.net/photodb/user?user_id=2378894"> </a><a href="http://www.photo.net/photodb/user?user_id=2378894">Pankaj Purohit</a></p>

<p>Now those scale to 4800ppi, thanks.</p>

<p>But the 100% crops look very soft. If I scale them down to 50% they start to look sharper, but still with a lot of noise. But scaling them down to 50% means you are left with an image of 7.7MP. Sure your original scan has 4 times that number of pixels, but they are so soft that if you print that at 300ppi it is going to look pretty bad.</p>

<p>For what it is worth I don't belive you are getting all that you could out of film, IMO.</p>

<p>This is the same size image as yours, but I would print this at 300ppi and expect it to be very sharp.</p><div>00Smms-117081584.thumb.jpg.8a0c93807633ea178804be41f4f541b3.jpg</div>

Link to comment
Share on other sites

<p>Scott, your image is really great in details, is it a 35mm negative or a medium format? this means I should downsize all my images to 7 megapixels...! or I need to work more on postprocessing or scanning techniques, what you think about colors? what would you suggest me to do with my negatives and scanner?</p>
Link to comment
Share on other sites

<p> Mauro,</p>

<p>Yes my image is stitched. For those that don't want to do stitching I can understand, but it is a pretty simple way to get large prints that look good.</p>

<p>As for stitching using 35mm, there really would be no reason to, much easier to do with digital. I have seen some people who stitched with film years ago, but now with digital just about anyone who is stitching is doing so with digital.</p>

<p>

<p>MF, it done very well, can also make a very good looking 20x30 inch print, but it is very easy to have it get messed up. Trying to get a good 20x30 inch print from a 6x7 camera and a flat bed scanner would be a challenge. The least expensive scanner that I know of that will do a good job on MF film cost around $2000, way more then I would be willing to pay for a scanner.</p>

<p>

<p>I know a number of people who have been giving up on MF photograph when they got a FF DSLR, not because the DSLR has more resolution but because the difference is small enough that it makes little sense for them to keep going through all the work of scanning film.</p>

<p>There might be a few people who would stick with film a bit longer if they had a better scanning solution, but I think the number is pretty small.</p>

</p>

</p>

Link to comment
Share on other sites

<p>I printed the crop its about 4.5 cm when printed at 360 dpi. Which was my point about viewing 100% crops on a monitor that they don't represent how the smaller finished print will look. Digital user will complain about film grain when they view 100% crops even though there is no grain in the prints. I never stated that the D2x would be better or worse than the 6x7 scan only that viewing 100% crops is not realistic for most people who are not making very large prints<br>

I made two prints one as is and one where I color corrected the D2x image and gave it a bit of a sharpen. If I was a D2x owner I would be very pleased with the results once the colors are corrected and the image is sharpened a bit. The tiny sensor of the D2x holds up rather well at that print size not as good as the 6x7 but then we don't really expect it to be but also I can't honestly say that the d2x looks bad either. I don't think a d2x would be my first choice for a project of reproducing old maps but for more general photography it would be a nice well built camera.</p>

Link to comment
Share on other sites

<p><a href="http://www.photo.net/photodb/user?user_id=2378894">Pankaj Purohit</a>,</p>

<p>Down sampling your images will really only help in saving some disk space, the main point is they will not print large and be really sharp.</p>

<p>My image if from digital and is only there to show what sharp pixels look like, don't worry about that for now, you will not get to that point with 35mm film, unless you do B/W</p>

<p>The main problem you are having is using a flatbed scanner, a film scanner will give far sharpper scans.</p>

<p>You are also going to need to use a very fine grain film to get a decent scan, there have been a number of threads on which film is the best.</p>

<p>For film there is not much you can do with sharpping, so your lens needs to be very sharp if you are going to get a good image. You also need to shoot at the sweet f/number of the lens, often around f/8, depending on the lens. And since the best films are iso 100, or less, a tripod with a cable release is a good idea.</p>

<p>But if you really want a sharp image that will print big and you really want to use film you can go to MF, but the cost is high if you are going to do your own scanning. On the plus side you could get some ok scans with MF and your flatbed scanner, not nearly as good as would coule be had with a film scanner but much better then the flatbed with 35mm film</p>

<p> </p>

Link to comment
Share on other sites

<p>"Great shots (again), Scott. What was the elapsed time between the first shot and last shot on the 54 MP bridge shot? How much time does it take to set up and take down your equipment to get the shots for an image like that?"<br>

--Lannie</p>

<p>It took 1 minute 43 seconds between the first shot and the last. It takes about 1 minute to set up the camera, when every thing is in the bag in good order. It takes far less time to take down the set up, maybe 30 seconds.</p>

<p>If you are thinking about getting in to this let me know as there is a LOT of detail that you will need to know to really make it work and to make it easy. The panoramic head that I mainly use cost a fair bit, around $550 now days I believe, but it makes it much easier to take the photos. For lower res shots, say in the 20-30 MP range you really don’t need a panoramic head. It is also critical that you get good stitching software, I use PTGui Pro, I have been going through the upgrade path on this so it has been not too expensive but it is not cheap software to buy up front.

<p>

<p>There is an evaluation version of PTGui, if you would like to play with some of my images let me know and I will send you the source images along with the project file for stitching them, this will give you a better feel for what all is involved .</p>

</p>

</p>

 

Link to comment
Share on other sites

<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 17, 2009; 05:04 p.m.</p>

 

<p>"Scott Wilson: your images are ridiculous... as in ridiculous eye candy. I have to say '<strong><em>wow!</em> </strong>' to every single image you link to :)<br>

My only guess as to how you do it is by using some combination of multi-image technology... multiple image stitching and/or super-resolution via PhotoAcute? The latter can do quite an amazing job at interpolating resolution and getting rid of noise even with my 10MP Panasonic Lumix LX3."</p>

<p>Hello Rishi,</p>

<p>You are right, multi-images, in this case stitching. </p>

 

 

<p>In the last few years stitching software as made huge improvements and has gotten far easier to get really good result from.</p>

 

<p>As I have said to others, I know some will want to stitch to MF and that is fine, I just like to point out that there are more options out there for those that like to print big.</p>

 

Link to comment
Share on other sites

<p>Rishi, thanks for posting the comparison with the noise reduction applied to the scan. That helps make it easier to put the shots by Les Sarile (that Mauro used to start this thread) into better perspective.</p>

<p>Those were darned fine shots, Les.</p>

<p>Mauro, you've hit another home run with this thread, in my opinion. There is a lot to learn, but some of us can get a big boost upward by trying to understand you guys.</p>

<p>--Lannie</p>

Link to comment
Share on other sites

<p>Scott,</p>

<p>Personal preference, but, I refuse to stitch shots, much like Mauro. Believe it or not, being a landscape guy, nature in fact <em>does not</em> provide you with limitless time during a sunrise or sunset, for example, to take shots over a period of 1.5 or 2 minutes. During sunrise on a mountaintop, the alpenglow can change in a matter of 15-30 seconds, which is sometimes the shutter speed of my shots when shooting ISO 40 film with small apertures and a polarizer. To capture that alpenglow, before the whole mountain is bathed in less saturated light, I wouldn't have the time to move the camera up and down & around to capture images to stitch.</p>

<p>I will, however, sometimes use the ability of PhotoAcute. At least with this solution, I don't have to move the camera around. I set up my composition, and I click away a number of shots. This in effect multi-samples the scene, and sensor noise is eliminated. With enough vibrational or translational movement from shot to shot, super-resolution is pretty darn amazing at finding detail that may not be there (or has a lower MTF) on any one of the shots to begin with. The resulting image is quite stunning, without requiring any movement of my camera.</p>

<p>Here's a comparison of one image out of a set of three consecutive shots VS. the super-resolution image (made using PhotoAcute), downsized to the original 10MP resolution of the Panasonic Lumix LX3 camera [one round of sharpening applied to both images in Lightroom, but PhotoAcute applies its own limited sharpening]:</p>

<p><img src="http://staff.washington.edu/rjsanyal/Photography/PanasonicLX3_PhotoAcute-2.jpg" alt="" width="800" /><br>

<a href="http://staff.washington.edu/rjsanyal/Photography/PanasonicLX3_PhotoAcute-2.jpg">Link to Full-Size Image</a></p>

<p>And another crop from a different portion of the frame:<br>

<img src="http://staff.washington.edu/rjsanyal/Photography/PanasonicLX3_PhotoAcute.jpg" alt="" width="800" /><br>

<a href="http://staff.washington.edu/rjsanyal/Photography/PanasonicLX3_PhotoAcute.jpg">Link to Full-Size Image</a></p>

<p>As you can see, even at base ISO (80), the LX3 has a fair amount of noise. And, as Scott would say, the pixels just aren't 'sharp' at 1:1 viewing. But they sure are sharp once you take the super-resolution file (2x resolution) and downsize it (bicubic). I found the results hard to believe myself, but, there it is.</p>

<p>Cheers,<br>

Rishi</p>

Link to comment
Share on other sites

<p ><em><strong>Poor scans - May explain why some switch to Digital.</strong></em></p>

<p > May be, but it’s partially…</p>

<p > </p>

<p >The main reasons seems to me is lack of adequate information. Can anybody imagine that such test may appear on Nikon Imaging website or in Popular Photography magazine? Obviously not. Seems like our so called “free public media” sits under tight financial control of digital manufacturers. And how many of us have seen that Nikon is promoting its scanners? Any ads, commercials, articles, announcements… Apparently Nikon already ceased the production of Coolscan V ED. And the reason is simple: it’s affordable, it delivers good performance, it’s popular and in demand. Yes, of cause, it contributes in film photography popularization and makes negative impact on digital camera sales. </p>

<p > </p>

<p >From my personal experience, 4 years ago I did not believe that scanning my produce good result. The only I’ve read and seen in our magazines it’s how a cheap digital camera can easily overpass scanned film in quality, gray range, res etc. and I was thinking quite serious about buying a DSLR. Until I took a look at Les Sarile Film 2 posting. This posting had drastically changed my mind. Since that I purchased my Coolscan V and never look back at digital again.</p>

<p > </p>

<p >Seems to me that Les and Mauro do for film photography popularization much more than the leading film manufacturers do. They did lots for all of us on this forum to keep our spirit alive. Guys, I guess the digital camera makers must hate you….</p>

Link to comment
Share on other sites

<p>Look at the effects of PhotoAcute on high frequency detail... left is one of the original images, right is the super-resolution file downsized to 10MP from just 3 images combined (remember: no manual movement between shots) using PhotoAcute:<br>

<img src="http://staff.washington.edu/rjsanyal/Photography/PanasonicLX3_PhotoAcute-HighFreqDetail.jpg" alt="" width="800" /><br>

<a href="http://staff.washington.edu/rjsanyal/Photography/PanasonicLX3_PhotoAcute-HighFreqDetail.jpg">Link to Full-Size Image</a></p>

<p>Sorry I may be going a bit OT now, Mauro, but I figured some of you guys might be interested :)</p>

<p>In fact, I wonder if this method would make current dSLR really compete with MF...</p>

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

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share


×
×
  • Create New...