Negative Editing in photoshop

[manojit_pusty1]

<p>Here is an image that I have developed at home with Universal Developer manufactured in India.....shot on Kodak Tri-X........developed for 7 mins at 24 degree C...............one is edited in photoshop (sharpened only) and the other untouched..........the film was scanned at the local lab..........I wud like to know why there is a whitish haze in the unedited one.......all the images in the film have come out to be that way.......is there any problem with the developer??</p><div></div>

[manojit_pusty1]

<p>the edited one</p><div></div>

[making_time]

<p>Looks like the first one is just a 'flat' scan, ie as much information on the negative(black and white point settings) is captured as possible for editing purposes later. My epson scans sometimes look like that before final editing in PS.</p>

<p>edit:if you've only sharpened the second one with no curve adjustments, then i'm not sure..</p>

[manojit_pusty1]

<p>the lab guy told me that he has given me a big size image and to get it processed in photoshop by my own.............I sharpened it in this way<br>

cntrl+j<br>

cntrl+shift<br>

'layer' palatte is changed to 'soft light'<br>

cntrl+j<br>

filter>sharpen>unsharp mask (proper values are taken)</p>

<p>these steps are done sequentially in photoshop CS4 to sharpen.........got these steps in the net</p>

[thirteenthumbs]

<p>This image is under exposed. If the whole roll is like this then you need to increase your exposure by setting your ISO dial to a lower number than 400. Your meter is not perfect, your shutter is not perfect and your lens aperture may not be perfect. When compensating for exposure with the meter ISO dial set it to 1/3 stop more exposure to start with. This will be 320 for ISO 400 film (the dot before 400 on most cameras).<br>

The other possibility is that the meter was fooled by the brightly lit bicycle and the girls light colored dress causing it to recommend a too fast shutter speed or if the camera was on auto mis expose.</p>

<p>The attached is the levels adjustment from photoshop. This is the first adjustment that one should do. The top levels is the unedited image you posted as it first comes up. Note that the blacks are way up the scale, they should be closer to 0. The whites are closer to 255 than the blacks are to 0 suggesting that the development was OK but that flat line has no detail. The lower levels window is a suggested adjustment. Set the Black or White slider to the edge of the histogram where the histogram begins and goes up from the base line. Setting further into the histogram after the rise from base line clips out information detail from the image. Set the other slider to the edge of the histogram. Adjust the middle slider to get the mid tones to suit you. </p><div></div>

[lachaine]

<p>I don't know. I mean, there are so many variables in film development and scanning that it's impossible to say. All I know is that I've personally scanned hundreds of negatives over the years on both film and flatbed scanners, and if the negative is right in the first place, the scan with auto settings comes out just about perfectly. You either have a low contrast subject and exposure, a low contrast negative due to development, or a low contrast scan.</p>

[JDMvW]

<p>By the time something is developed, scanned, and posted here, it is hard to tell exactly where the problem(s) may be.</p>

<p>Incorrect exposure, outdated or damaged film, and exhausted developer or other chemicals are possible factors here.</p>

<p>It's amazing what can be done in Photoshop to fix such a problem, but it's really better to get it right in the camera and on the film itself.</p>

[Tim_Lookingbill]

If you look at the histogram of the flat undedited version you'll note that the majority of the data/detail (and there is quite a bit even for a downsized jpeg especially in the background foliage) has been compressed toward dead center. This makes for very difficult editing with Photoshop's tools (curves/levels) in a gamma encoded working space as is the case in Photoshop. What you should do to bring out all that detail is redistribute it within Photoshop's processing engine environment as opposed to the lab's scanner which is what you've been doing. Here are the steps...

<p>

First convert to grayscale and then into AdobeRGB. Choose Auto Contrast. Then find or build a 1.0 gamma color space and assign it to the image. If you can't find one you can build one by loading AdobeRGB as the RGB Working Space in Color Settings...scroll up to CustomRGB...turn AdobeRGB's 2.2 gamma into 1.0...give it an appropriate custom name...click OK...then scroll up to Save As RGB which will save it as an ICC profile you can assign.

<p>

Once you assign this 1.0 gamma profile convert back to AdobeRGB and go into curves and you'll be able to pull and redistribute a lot detail uniformly. I tried these steps and came up with an S-curve with the highlight sampled in the woman's dress on the left. Sorry for the drawn out instructions but you've got an image encoded in the scanner's space that Photoshop knows nothing about.

<p>

Can't post my corrected version because I'm not able to upload images on the old system I'm on.

[thirteenthumbs]

<p>A negative that prints well at paper black on any enlarger scans on average with the deepest black between 0 and 25 with the highlights between 230 and 255. Blacks develop fully quicker than highlights. The highlights flat line begins at 236 suggesting the film was close to fully developed. Weak chemicals would most likely resulted in a more compressed tonal range.<br>

You may need to increase development time by 5% if the chemicals are relatively fresh.</p>

<blockquote>

<p> there are so many variables in film development and scanning</p>

</blockquote>

<p>That's why some of us state the most likely first then work toward the least likely as more information becomes available or testing proves it not to be the cause.</p>

[manojit_pusty1]

<p>The developer stock solution was prepared 4 months back and currently occupying half the bottle..........heard oxygen in air degrades the developer..........can that be the case then........also I'm uploading a properly exposed image.........</p><div></div>

[manojit_pusty1]

<p>isn't there a faint pinking or magenta cast??</p>

[thirteenthumbs]

<p>This one, the second image, the blacks start at 23 and pick up detail at 30 and the highlights are at 236, better than the previous image. There is a fair amount of adjustment room with basic tools and techniques.</p>

<p>It is, IMO, normal for scans to be flat off a scanner set to capture maximum detail from the negative.</p>

<p>Oxygen causes the developer to oxidize. You can take a breath, hold it, then exhale into the developer bottle and immediately cap the bottle to cut down on the amount of oxygen in the air space in the bottle.</p>

<p>The pink cast is residual sensitizing or anti halation dyes not fully washed out during processing. They are not a problem. After adjusting the levels in photoshop use the auto color tool in adjustments, the color cast should be removed completely. There are several other ways to remove the color cast also.</p><div></div>

[manojit_pusty1]

<p>thanks fr the responses so far................here i wud like to know that is it natural to edit scanned negatives in photoshop or only those require which have problems in processing(keeping under exposed aside)</p>

[thirteenthumbs]

<p>It is normal to edit all scans in photoshop. A good scan only captures all the information possible from the negative. A properly exposed/developed negative correctly scanned will be easy to edit and have a lot of adjustment range without the highlights burning out or the shadows going to solid or flat black. A bad exposure will be difficult to edit into a good image even with a good scan. A poor scan will be short on detail. Careful examination of the negative with a loupe will tell you if its the scan or the negative. Detail that can be seen in a negative that is not in the scanned image is a bad scan.</p>

[dennis_oconnor6]

<p>Tim Lookingbill. That was a really interesting reply. Just goes to show what a brilliant site this is.</p>

<p>Dennis</p>

[j_ri_l]

<p>The photos are exposed and scanned well enough. I use Vuescan to scan a low contrast image in order to get all the information on the negative. Then I use "Levels" adjustment in Photoshop to increase contrast as I like. This seems to be the perfect workflow so far.</p>

[manojit_pusty1]

<p>Thanks again to all for clarifying the doubts :)</p>

[john_shriver]

<p>The first key is to properly adjust the black and white levels in the scanning application, before it gets converted to a gamma-curve encoding. This is absolutely normal, this is why every negative doesn't get the same exposure in the enlarger, and why there are different contrast grades for printing paper.</p>

 

[JRCrowe]

<p>A weak fixer solution can cause this also. Do a Search here on Photonet on "weak fixer solution" and you will find advice. You can actually run the negatives back through a fixer bath and remove the haze.</p>

[sebastianmoran]

<p>Manojit, scans are generally flat, as others have commented. It's a tradeoff; if the scan is not flat then an automated process has adjusted contrast and tonal range to give the image punch. </p>

<p>Here's the tradeoff: With automation, you might have less work, but you lose control and the automation might be "off" for some shots. Scans I do myself with my Coolscan usually need adjustment, and I don't mind this. In the US, the only automated scanning adjustments I trust are by North Coast Photographic (San Diego) and Precision Camera (Austin) who standout from the rest.</p>

[Tim_Lookingbill]

It's black and white. You have 256 density levels on your display. Your film scanner is sensitive enough to distinguish the slightest change to density it detects in a piece of film that's maybe an inch from the lens with plenty of light. If a DSLR can do this with a lens shooting landscape detail 100's of feet away, so can a scanner an inch away from film. The film either has these density variations we call detail or it doesn't. I've gotten this level of density variation sensititivy scanning prints even on a mid-range flatbed scanner (Agfa Arcus II back in '98) so a film scanner should be able to pull a lot of these density variations in a negative if it's there.

<p>

It boils down to whether the scanner compresses these density variations to the point you can't distinguish them as in the flat, foggy look of the OP's scan or renders it perceptually as the eyes would perceive the scene. Just because you can't see the density variations doesn't mean they're not there. You can't edit what you can't see, so a uniform 1.0 gamma assigned to the image brightens all levels so you can see these density variations.

<p>

Photoshop layer tricks aren't going to do these types of images justice. You must carefully and gradually redistribute these density variations with a custom curve. Levels doesn't cut it because it uses algorithms assuming a properly exposed capture within a 2.2 gamma encoded editing environment which this image isn't.

[sebastianmoran]

<p>Tim... I've appreciated your advice in the past. And, I agree with you. If the values are compressed into the middle of the range, nothing good will result.</p>

<p>What does one do to get "a properly exposed capture with 2.2 gamma encoded" from our scanners?</p>

<p>And, for what it's worth, I'll still take a somewhat "flat" scan... Better that than a scan with either end blown out.</p>

[Tim_Lookingbill]

There's nothing wrong with a flat scan, Sebastian. All the data is still captured, but it makes it very difficult to see mapping all those tones into a perceptual rendering the way our eyes perceive the scene. It's also very difficult using Photoshop's tools. Try fixing the original image just with curves which is the most straight forward, simple tool to map 256 levels of tone. It's darn near impossible and a PITA at least it was for me.

<p>

I just hope Manojit is scanning in 16 bit output.

[manojit_pusty1]

<p>thanks again fr the responses.....Tim.....from this discussion I have understood that<br>

I shud <br>

1. try to scan with max detail<br>

2. Use levels in PShop<br>

3. Use auto color<br>

Please let me know in simple terms how to get more natural and correct black and white image.......excluding in the in camera adjustments....</p>

[zensphoto]

<p>Hi Manojit,</p>

<p>I have had scans done by a lab before and some of the best negative I have, have come back looking very flat as in your photograph. Over the years I have found better scanners and have done all the development and scanning here at home. I have had a few flat b&w negatives come out from time to time. Most of my home b&w development comes out perfect every time now. Photoshop to the rescue is always a good thing.</p>

<p>The haze from the lab scan usually means a dirty lab scanner. I have seen that before as well and it looks much like this.</p>

<p>I also read you developed Tri-x at 24C for 7 minutes. My time table says 6.5 minutes in 68F/20C and at 24C it says 3.5 minutes. Perhaps, you developed it to long? </p>