First B&W film processing in many years

[rwbowman]

<p>I found a box in the attic of some of my old darkroom equipment and decided to shoot some B&W film with my old Canon F-1. Here's an example from my second roll, Ilford HP5 Plus, Ilford DD-X, Ilfostop, Ilford Rapid Fixer. Scanned as a color transparency with SilverFast AI on a Konica/Minolta Dimage Multi-Pro scanner, converted to positive and spotted in Photoshop, slight crop and "burning" in Adobe Lightroom. Whew, a lot of effort for one muddy image of my dog, but also a lot of fun. Sure would appreciate a comment or two.</p>

<div></div>

[bgelfand]

<p>A nice sharp image with good tones. I would say you remembered how to do it.</p>

<p>For a better critique, you might post in the Photo Critique Forum.</p>

[JDMvW]

<p>Makes me miss my dog.<br>

I'd say it looks just fine photographically.<br>

A lot better than my own earlier efforts at relearning how to load a 35mm tank reel.</p><div></div>

[lex_jenkins]

<p>Looks great. Most folks starting out - or restarting - tend to underexpose and overdevelop. Your photo appears to have a good tonal range.</p>

[peterbcarter]

<p>Your skills seem intact and your model is stunning!</p>

[Member69643]

<p>That's a very nice shot! The windblown fur and the eye peering out, a very nice composition.</p>

<p>When I scan B&W I always scan as a B&W negative in Vuescan. I am curious why you scanned as color transparency and then converted to "positive" - is that a Silverfast thing?</p>

[sasvata__shash__chatterjee]

<p>Robert, I actually like the tones in your image, nice job!<br>

I also read up a bunch on the scanning, and for now I came to the conclusion that scanning as a RGB gets you nothing but a bunch more redundant bits and therefore a larger file, and the extra inversion step. Once you do some experimentation to show which of the three channels gives you the sharpest scan, you an just stick to one.</p>

[rwbowman]

I don't know, I read somewhere that scanning as a color image gives more detail in the image, especially in the shadows.

I'll do some comparisons--it would be nice to simplify the workflow. Thanks all for the encouragement and suggestions!

[dave_wilson1]

Very nice indeed! Beautiful dog 8-)))

[rwbowman]

<p>Just to follow up with your suggestion, Shash, you wrote, "Once you do some experimentation to show which of the three channels gives you the sharpest scan, you an just stick to one." Does this mean that of the RGB channels, one (say green) might be sharpest? But how can I just use the green channel without scanning first as a color image? (I've mostly been scanning color transparencies and these are my early efforts with a digital B&W process, so if I'm missing something VERY basic, don't be afraid to let me know.) Thanks again for the helpful post.</p>

[CosteaM]

<p>Robert, why would you scan BW negative film as color transparency then conversion back to BW when you can simply scan directly under gray scale BW. Try this next time. Nevertheless your picture turned out very well, congrats.</p>

[peterbcarter]

<p>It's a lot of extra work for no real gain.</p>

[lenny_eiger]

<p>No, it isn't. This is a very specific issue which has a consensus among scanning pros. Whether you are doing a conversion in the scanning software or in PhotoShop, the programs use a specific formula that utilizes about 69% or 70% of the Red channel. They do this because they aren't photographers. If you open an RGB (color or b&w) image in PhotoShop at 100%, and look at the individual channels, you will see that the Red channel has much more noise. Sometimes its a small difference, sometimes its considerable. This definitely affects image quality, depending on how you print, how large, and what kind of look you are after.</p>

<p>Most pros will choose the Green channel, using the channel mixer set to monochrome, setting the Red and Blue to 0% and the Green to 100. I look at the channels and usually do 100 green but sometimes choose 80-90 Green and 10-20 Blue. It's a very small difference. Once that's done it converts all the channels to the Green channel so PhotoShop can do anything it wants in the conversion and you still get what you want.</p>

<p> </p>

[peterbcarter]

<p>I have yet to see any difference. Where is your stats on "most pros". Any pro that I have had contact with usually do a terrible to moderate job, and the ones that do use big scanners that they don't have a clue on how to adjust (drum scanners). Those tend to leave it on auto pilot. Pros are pros because they get paid for it, not because they are the best at it.</p>

<p>I agree that photography scanning did suck 20 years ago but I would seriously contest your points today. Vueue Scan is but one piece of software that has brought the art to perfection and its ability to give you the control to take it further. It does take some time to use it well. I don't like Epson Scan, but I have seen results that will not question it's ability.</p>

<p>Lenny, I feel your points fall into "just because". Until I can actually see any proof, it is only a feel-good-to-do-it-this-way thing. If that's what you want, keep doing it. I just dislike spreading the myth.</p>

<p>Here is the challenge.....<br>

1) Post a RAW scan of some BW film and include some of the unexposed frame (for fog calibration).<br>

2) I will post a pure vuescan result.<br>

3) You post a result using your colour magic method.</p>

<p>This matter can be then put to rest.</p>

 

[sasvata__shash__chatterjee]

<p>Hi Robert, I use VueScan. I am not sure what the equivalent option is in SilverFast, but there should be one. In VueScan, when I set the Input to 16-bit Gray, there is an option to let the software make the grayscale image from specifically one of the Red, Blue and Green channels, or leave it at Auto (I am not sure what algorithm it uses to pick one in that case). So, you could scan once each using 16-bit Gray, but varying the Red, Green, Blue channels and then comparing the images to pick which channel you like best. From then on you'd let the scanner always make the grayscale image from that color channel. If you save the image in RGB, then manually choose one of the Red, Blue or Green channels you are essentially manually doing what you could let the scanning software do automatically in a single step. </p>

<p>If you were converting an original color image to monochrome, you could mix the R, G, B channels in endless proportions to get different effects. But, if the original image is grayscale to begin with, then the R, G, B channels would only show differences due to the difference in the three HW analog-to-digital channels themselves, not anything to do with the original image. More than likely the differences would be the same from scan to scan, and that is why you can scan to grayscale using the same channel once you determine which is best on your particular scanner.</p>

[lex_jenkins]

<p>Lenny has already established his bona fides. <a href="http://www.eigerphoto.com/"><strong>Here's his website</strong></a>, where you can find testimonials from satisfied clients.</p>

<p>As with the tiresome raw vs. JPEG debate, I'll say the same thing about scanning b&w negatives: I've often regretting scanning at less than the maximum possible resolution and quality. I've often regretted throwing data away based on bad advice I read years ago. In the long run I've never regretted spending a little more time to get the best possible data in the first place. I need to rescan hundreds of negatives because I listened to bad advice about scanning grayscale only during the late 1990s-early 2000s. It seemed like I was saving time back then but in the long run it costs more time and effort. And we won't see many new, affordable and top quality film scanners. That era is gone and isn't coming back.</p>

<p>Scan for full data now. You may not get an opportunity later to do it right the first time. Later, as your editing skills and technology progress you'll appreciate having more data to support the best possible output.</p>

<p>And I'll reprise a comment I made earlier today on the digital darkroom forum:</p>

<p>The problem here is that the "invitation" to support opinions with examples sound less like an friendly invitation and more like a prickly "put up or shut up" dare.</p>

<p>And that is precisely the problem with photo.net.<br>

<br /> <br />I took a few months off after April to reevaluate my position and to gain some distance and, hopefully, objectivity. I spent less time discussing the nuts and bolts of photography and more time on photography itself, particularly documentary.</p>

<p>When I revisited photo.net last week I'd hope it was with a refreshed and more positive outlook. But what I read in this thread only serves to reinforce my earlier impression. Photo.net feels unwelcoming to newcomers, discouraging to experienced photographers who occasionally venture an opinion or suggestion, and just generally toxic.</p>

<p>This entire site is in need of an intervention, a makeover and a serious attitude adjustment.</p>

[sasvata__shash__chatterjee]

<p>Robert, I missed answering one part of your question.</p>

<blockquote>

<p>Does this mean that of the RGB channels, one (say green) might be sharpest? But how can I just use the green channel without scanning first as a color image? </p>

</blockquote>

<p>Whether you are scanning B&W or color, the scanner HW is doing essentially the exact same process. There are many different techniques and scanner types, but let's stick to the common denominator basics. In a nutshell, ignoring the infra-red channel, the scanner is essentially illuminating a bunch of discrete points, pixels, at each scan position. Each of those pixels results in a beam of light that follows some optical path eventually through a lens (tiny prism, really) that splits the beam into three component beams, then those go through a red, green, blue filter respectively to hit three different analog sensors, which are then individually digitized to get the R, G, B channel value for each pixel. In an ideal world, from a truly grayscale image, each R, G, B value for a given pixel would be identical. But, in the real world, due to non-linearity of the lens, or the differences in the filters in the amount of light it lets pass, or differences in the sensors, as well as inherent noise in the electronics, each of the R, G, B channels is slightly different. This is why one of the R, G, B channels in your particular scanner will be better/worse than the other channels.</p>

<p>So, the scanner <em>hardware</em> is always going to scan into three R, G, B channels. However, you can set the scanning <em>software</em> to grayscale and let it keep data from one of the R, G, B channels after the scan is complete, or you can let it keep the values from all three channels. </p>

[rwbowman]

<p>Interesting comments and give a lot to chew on!<br>

Here's another version, same process, but green channel only. See any difference? It may be the power of suggestion, but this one seems smoother, less noisy.</p><div></div>

[rwbowman]

<p>Comparing the two images side-by-side at 1-1 in Lightroom, the second image is definitely smoother and less noisy. </p>