Dont know whats wrong with my photos?

<p>For one thing, I think the question of what "speed" you choose to "rate" a particular film at, and how you choose to meter and expose the film for particular subjects are two separate issues.</p>

<p>The ASA value at which you rate a film is usually based on your previous expereince with that film. If your previous roll of Tri-X (for example) appeared underexposed when rated at ASA 400, then try rating it at ASA 320 or 200 next time. </p>

<p>The reason many people will meter in the Zone 3 (for example) is in order to make sure that they end up with adequate detail in that zone on the negative. The typical zone system approach is to give less development to scenes where there is a greater difference in light intensity, so as to control (not to blow) the highlights. But you shoot 35mm or 120, then your whole roll is getting the same development, of course.</p>

<p>Everything has a tolerance, your camera's shutter, the lens aperture, the film production, the developer production. Add all the differences from "perfect" and that is the shift in exposure/development that one has do to compensate for to get good results. Setting your meter to something other than box speed is a convenient way to add or subtract that little extra or little less exposure necessary to get a good image. Film speed testing/bracketing is a controlled method to find out what that exposure variance is. A small amount of exposure will have a greater effect on the end product than a small change in development time will. A exposure increase of 1/2 stop over box speed will not put the highlights of a normal lit scene over the limit of the film.</p>

<p> </p>

<p>I just do not understand how that works. Slightly overexposing the film and then compensating for it when developing will just counter act it. The opposite, underexposing and then developing for longer = pushing.</p>

<p>I really cant see the difference. Again - why not just set to 400, expose accordingly and develop for highlights. It is exactly the same idea, but just with the bonus of added speed.</p>

<p>Faster speed, be it the shutter or the ISO dial on your meter, gives <strong><em>less light</em></strong> reaching the film.<br>

Less light reaching the film gives whites that scan/print as middle light gray, light shadows that scan/print as very dark gray to black.</p>

<p>A little more light at exposure results in higher densities across the board making whites scan/print as light gray and light to medium shadows scan/print as medium to dark gray.</p>

<p>Go shoot some film and prove it to yourself.</p>

<p>I think it is more helpful to think of all of the 'numbers' --like the ASA ratings on film boxes, the developing times in the developer data sheets, etc.-- not as immutable, absolute values which you 'must' adhere to, but rather as guidelines. If exposing at "box speed" and developing according to the data sheet instructions gives you great-looking negatives, then keep doing it. But if you are not getting the results you want, then you can make adjustments with the variables.</p>

<p>Reading "The Negative" by A. Adams a few years ago helped me better understand some of the relationships.</p>

<p> </p>

<p>Thanks everyone. I understand it all now, just need to put it in practice effectively!</p>

<p>I ordered the negative to have a read of, hopefully that should shed some light on it :)</p>

<p>I just added four pictures i developed earlier today to my photostream on flickr ( )</p>

<p>Im not sure if theyre better or what. All i know is that the negatives look like they have all the detail on them. When i scan them its all there on the screen, but obviously very flat at that point. It must be my lack of decent editing skills that is the main problem (and the lack of decent light on the days these photos were taken)</p>

<p>One thing id like you to notice is the highlights, how noisy/grainy they are. Is this just my rubbish scanner? Also is it worth scanning at the full resolution or half enough? Its sharpness i want and obv more res will not help there.</p>

<p>Thanks!</p>

<p>The photos you posted today definitely look better than those posted earlier. </p>

<p>I scan at the manufactures stated optical limit then resize in photoshop. Many will say that the manufacturers specs are exaggerated but instruments can detect differences that the eye cannot and micro etching in test targets can expose strengths and weaknesses of the scanner.</p>

<p>Post a unedited scan in line at the fourms limits that can be copied to other computers so that I or others can play with to better evaluate the exposure/scan.</p>

<p>Thanks - still not happy though. Its my scanning/photoshop work i think. But as i say the light has been pretty pants.</p>

<p>Ill try scanning at the highest res then. Still never seem as sharp as other peoples scans from the 4490 but maybe thats the ID-11? About resizing - i just save as a smaller JPEG is that okay? I dont need to actually resize prior to that to maintain some quality or something?</p>

<p>Image....</p><div></div>

<p>Scanning a strip of negatives is fine for determining which images are worthy of further work. Once you determine which negative(s) you want to work with use the scan window crop tool to crop out the edge of the film so that you have only the image then prescan for that cropped area. Scanner default settings should be adequate. Remember you are only trying to capture as much information from the negative as possible, not make it look good.<br>

Next save the image as a Tiff, not jpeg. Tiff is loss less, jpeg looses a little information each time you save the image.</p>

<p>The image you posted with the black band cut off in PS7 using the standard rectangular crop tool and the adjustment in levels box 2. First levels box is the resulting levels with the band removed. Second levels box the adjustments I made. Note blacks at 17 not 0, not bad, exposure OK; Whites or bright area a thin spike at 255 but insignificant with no information until 210 suggest underdevelopment. Your thermometer may be off a degree or two causing inaccurate development time. Suggestion, increase development time by 15% <br>

An increase in development time is known as contrast increase, a reduction in development time is known as contrast compression. How much extra/less development time depends on the scene and some careful testing.</p>

<p>Use a good loupe to check the negative for edge sharpness and detail then compare to the scanned image. If its sharp on the negative and not in the scan then it is the film holder or the scanner.</p><div></div>

<p>Hi</p>

<p>I always save as TIFF, i just sent that as a jpeg for obvious reasons. On that note though - was wondering if there is any benefit of editing TIFF in ACR instead of PS in terms of quality? </p>

<p>I usually scan as a strip and then scan individually for the pics i want, but just chopped that out of the strip to send. I assume you are saying that the black between and around the frames influences the histogram during prescan. I'll make a note of that for future. </p>

<p>Underdevelopment is the problem then! Cant be the thermometer as i use two! I also start developing while it is a little over 20 to compensate for the loss of temp during. It seems like all my negs come out the same though, and its confusing because i use the times given by the massive dev chart and the FDC charts. Surely they cant all be off by the same amount? It must be something else that i am doing?!</p>

<p>Ill try developing for longer on the next roll. I am waiting for a big order of tri-x because at the moment im using various individual rolls and that could add to confusion. </p>

<p>I imagine my holder probably isnt too special. I have been looking for a place to get a better holder (tried betterscanning but found nothing in stock or available) or some of the anti newton glass to lay in my epson holder but i cant find anything anywhere.</p>

<p>Here are the settings etc i use to scan - please tell me if anything seems wrong...</p>

<p>I am not familiar with ACR. One editing program should not be better than another quality wise. </p>

<p>I have three thermometers, one is off by 10+ degrees and the other two disagree by 2 degrees and I don't know which is correct. I use one and adjust my development time to get the highlights correct.</p>

<p>Box film speed+manufacturers published development times are not guaranteed optimum results. They should produce good results. Your results are in the good category.<br>

Looking at the histogram on your scan settings suggest under exposure as well as under development. You really need to do some bracketing/testing with your equipment. <br>

Your camera's shutter is not perfect, its lens aperture is not perfect, and the meter is not perfect but all are with tolerance. Those tolerances add up to exposure error.<br>

When developing try keeping the developing tank in a tempering bath between agitation cycles that is the same temperature that you are processing at. Monitor and maintain the temperature of the bath.</p>

Ah so my thermometers might be off. Maybe i should get a digital one?

 

I'm going to try some tests now. Got my chemicals in the bath warming.

 

I'm going to just take 1 stop bracketed shots of various things indoors. Some text etc for sharpness. Any tips

before I shoot the roll?

<p>Use a tripod, cable release, and lock the mirror up if possible.</p>

Yeah. Any advice in subject. Presume just areas with a variety of tones?

<p>Variety of tones and lighting.</p>

<p>And if you get a digital thermometer what guarantee do you have that it will be 100% accurate? <br>

Don't assume. Take good notes when testing.</p>

<p>I think it takes a lot of trial and error to get an optimum scan, and I am not sure I am doing it right. I have the Epson 3200 (using the 4770 driver because the 3200 software won't work with Windows 7). My settings screen looks just like yours.</p>

<p>You're not scanning a whole strip of 6 frames and then cropping individual frames out of that image file, are you? I don't think that will lead to good results, because the automated exposure adjustments, histogram settings, etc. would be averaging everything together, including the spaces between and around frames.</p>

<p>With black and white, I find I get better results when I put the right-hand output slider at somewhere between 210 to 230, instead of at the maximum, and then adjust the highlights with the input slider. With the output slider at maximum (255), I find it blows out any detail the highlight areas of the negative. Also, I find what works best for me is scanning 35mm at 1200dpi, and setting target size to 150 or 200 percent. That's what works for me as far as getting a raw TIFF scan that I can easily edit in PS. I am probably breaking all sorts of rules for best practices, but that's my routine. </p>

<p>Thanks John...</p>

<p>Your technique sounds interesting, but seems to go against almost everything i have been told and read! Would be daft of me to start doing something totally different at this delicate learning stage i think! Plus i am sure the input/output is more complex than how you use it? Perhaps someone could explain?</p>

<p>You mention scanning at a lower dpi but changing the size - can someone explain the benefit of changing the size? I would have thought higher dpi was better all round for editing.</p>

<p>Charles - i have the negatives and looking at them as a strip with no adjustments made, the overexposed one looks a LOT better than previous scans! Can you run me through what to look for for correct exposure, because obviously i could make them all look better exposed if i wanted through scanning. Presume it is how the histogram looks when each frame is cropped in prescan?</p>

<p>Thanks!</p>

<p>I have scanned three examples of the same shot taken at different exposures 1 stop apart. Each picture shows the histogram within the scanning software and the result when exported by pinching the histogram with black and white points. I cannot tell which is best as i have made them pretty much the same by exporting them.</p>

<p> </p><div></div>

<p>....</p><div></div>

<p>finally...</p><div></div>

<p>Well once again when clicking the spell check button a blocked popup message dumped all that I had typed when I clicked allow. To keep in step with the thread going to the toilet I went to the garage and got my 5 pound sledge hammer and beat the sheet out of the computer and monitor. It's working better for now. :D</p>

<p>1 st. normal exposure: some detail in the floor, whites a light medium gray, and has a tonal range of 137 units on the scanner.<br>

2 nd. +1 stop exposure: More detail in the floor, whites overall a little lighter but still too gray, has a tonal range of 148 units on the scanner.<br>

3 rd. -1 stop exposure: almost no detail in the floor, whites about the same as other exposures but the lighting is not as even, tonal range 120 units on the scanner.<br>

My older scanner does not have the histogram in the prescan but using the default settings and unaltered output in PS comes out blacks around 20 to 25 and highlights around 240. The wider the tonal spread the better. </p>

<p>Now with regard to your exposures, both the normal and +1 show good detail in the dark areas. +1 is a little too much and Normal slightly low to OK so setting your meter to +1/3 more exposure, 320 for 400, 80 for 100, ect... should be all that is needed exposure wise.<br>

The whites or highlights are dull across the board. It does not matter if your thermometer is accurate or off 1⁰C to 2⁰C you are definitely under developing. Increase your development time by 15%. I looked at the time charts on Ilford's web site and it lists 14 minutes for Delta 400 @400 in ID11 1:1 at 20⁰C. A 15% increase is 2 minutes 6 seconds or 16 minutes as 6 seconds for that long a development time is not going to have any noticeable change in density. There is no time listed forDelta 400 @ 400 in ID11 1:3, if other sites show times then it is someone else's testing which may or may not work for you. <br>

Using the same thermometer and being consistent in your procedure is more important than getting starting point times to work or buying new thermometers and hoping they are correct. Keeping the development temperature constant during the developing time is also important as 2 minutes is the equivalent of a 2 degree temperature change.</p>

<p>As far as scanning at a lower resolution and scaling to a larger percent only the direct print size will be affected and overall quality and detail may be less. The Epson US A web site shows the 4990 to have a maximum optical resolution of 4800dpi. I would scan at 4800 at 100% then downscale in editing software it the file size is too large to work on. I scan my B&W in RGB mode not grayscale.</p>

<p>Ha - sounds like your technical level is not quite as high with computers as with black and white film chemistry! Glad you fixed it though :)</p>

<p>Ill start exposing at 1/3 stop slower speed then. Will try some tests outside today. </p>

<p>I am still waiting for my delivery of tri-x and have been shooting with rollei film i found super cheap. Ill apply the 15% increase to this until i get the tri-x. (big shortage of tri-x in the UK - apparently thousands on back order!)</p>

<p>Disappointed about my thermometer, its a paterson one with a guarantee of temperature accuracy. Bad form by paterson! I kept the tank in the bath between agitations with the test roll yesterday so i'll do that in future too. I agitate by inversion 4 times every minute, or for 10 seconds. This is right??</p>

<p>About scanning, my software allows me to scan to 12800??? What happens if i do that but it cant handle it? I'll switch to 4800 - this should get rid of some of the noise when sharpening i expect, or make it smaller at least. The file wont be too large to work on, so ill just work on it at 4800 then save it as a jpeg for the internet when needed. </p>

<p>Why RGB not grayscale? 24-bit or 48-bit? I presume i still tell epson scan that it is b/w negative film?</p>

<p>Received the negative from amazon this morning :)</p>

<blockquote>

<p>Ha - sounds like your technical level is not quite as high with computers as with black and white film chemistry! Glad you fixed it though :)</p>

</blockquote>

<p>Actually I'm a better electronic technician than a photographer but photography is overtaking the electronics. </p>

<blockquote>

<p>Disappointed about my thermometer, its a paterson one with a guarantee of temperature accuracy. Bad form by paterson! I kept the tank in the bath between agitations with the test roll yesterday so i'll do that in future too. I agitate by inversion 4 times every minute, or for 10 seconds. This is right??</p>

</blockquote>

<p>I do not know that your thermometer is bad. I do not know for sure that a 15% increase in development time is perfect either but it should be within 3% of acceptable. I was using the temperature shift from Ilford for ID11 at 1:1 as a reference only. If your development is even then your agitation is fine. Your development is even.</p>

<blockquote>

<p>About scanning, my software allows me to scan to 12800???</p>

</blockquote>

<p>Anything above manufacturers stated optical limit at 100% is software interpolation. The scanner software adds information based on what the surrounding area is. </p>

<blockquote>

<p>Why RGB not grayscale? 24-bit or 48-bit?</p>

</blockquote>

<p>RGB is Red,Green,Blue the three primary colors. Add the three in different amounts to get any other color, equal parts of each mixed together = white. Each color channel is capable of 16 bits of information per pixel, 3 channels = 48 bits. 8 bit per channel X 3 channels =24 bits.<br>

The 4990 uses a white cold cathode fluorescent light source. Some newer scanners use 3 lcds, one red, one blue, one green. The different channels will have a slight tonal difference. Some will save one channel only as their grayscale others will convert to grayscale and others leave it as rgb. The means you use to reduce the image will also affect the quality. Try it, first duplicate the file then scale down the duplicate using one means and save it, make another duplicate and scale it down by another means then compare the two or three or 4 or??? <br>

Leave the setting at B&W Negative when scanning B&W negatives otherwise you will get a color cast that you will have to remove.</p>

<p>Experiment, keep accurate record (make good notes), and confirm/refute findings or observations.</p>

<p>Hi</p>

<p>Just following up with some more photos i recently developed. The first five on my photostream are new - http://www.flickr.com/photos/tompatrick/</p>

<p>I have been using tri-x and applying the techniques mentioned by the helpful posters of this forum and they seem to be making better results. Still a lot to learn though!</p>

<p>I also switched the rodinal.</p>