Need help doing a lens test.

<p>I have a Polaroid 100 that's been refurbished by landcameras.net. It works well and the photos using FP-3000b look OK.<br>

However, I need a bit more then that as I need to do a lens test to prove the camera is in focus. I require a Polaroid 100 camera to be in manufacture focus for for a project I am doing. According to the Polaroid manual this is <br />40 lines/mm. Since collimators don't exist any longer it needs another type of lens test. I do not have the skill or familiarity to do the ground glass method. I tried using a Korean 2003 lens chart printed at 25 cm using f/8.8 and FP-3000b film (in the shade) at a distance of 5.7 meters. Supposedly, I should be able to see about 20 vertical lines on the chart at that distance. No luck. I can't tell if the rangefinder is off or something with the optics. Quite possibly it's me and I am doing something incorrectly. Since this is for a project and not for me to specifically learn photography I have come to the conclusion I am a complete dunce and need someone more skilled then me to test the camera. <br />Therefore, I am looking for someone in the Chicago area that would be kind enough to assist me. I can also mail out the camera and happy to pay a reasonable fee. I have been told there are other reliable focusing methods and tests, but require someone more skilled them me. <br />Thank you.<br /><br /></p>

<p>Do you need it to be in focus or do you need it to resolve 40 l/mm? The camera is probably capable of that resolution, but I'm pretty sure there's no film currently available for it that would come anywhere close. I think you'd be lucky to get 10 l/mm on a print.</p>

<p>If you are just trying to see if the camera focuses, the way to do that is put out a yardstick on the floor about 8 - 10 feet away from you, running away from you and focus on 18". If the camera is working right, then 18" will be in best focus compared with what's nearer or farther.</p>

<p>Fuji says 20lp/mm for FP3000b. </p>

<p>http://www.fujifilm.com/products/instant_photo/pdf/fp_3000b_datasheet.pdf</p>

<p>Now, you could put a sheet of some ordinary sheet film in a used film pack (in the dark), and load it into the camera (in the dark), and shoot it. I am not so sure why you would do that with a polaroid camera, but you could. </p>

<p>Otherwise, to get that resolution, you will need to focus really carefully, which depends on how well the rangefinder is calibrated. That is why you want the ground glass.</p>

<p>You should be able to glue a piece of ground glass into the frame from a used film pack, then hold it in the camera, with the back open. </p>

<p>Can you tell us what you are really trying to do? This doesn't sound like a normal use for a polaroid camera.</p>

<p>Just dragging this off on a tangent, here, does 20 l/mm sound credible to you for a diffusion transfer process when 15 l/mm is usually cited for silver prints? I think Fuji is engaging in wishful thinking here, perhaps based on some idealized lab setup. :-) The Fuji prints from my Hasselblad look nothing like sharp, and I'm pretty sure it's not the Blad's fault.</p>

<p>I completely agree with Glen---we need more info. I hope this isn't some job where the allure of Instant is going to result in inevitable failure because instant technology isn't up to the job.</p>

<p>Correct this is not a normal test for the normal use of the camera. I am trying to do blur tests. In reality my photo consultant is attempting to get these tests done. However, he is out of the country and unable to do them so he relys on me. He is trying to determine focus errors and aberrations, which the f/8.8 setting is more sensitive too. Other tests will be done at the f/42 setting.<br>

We are trying to apply deblurring methods to a old Polaroid photo. It is difficult to predict the accuracy of the results because they are depend on how each algorithm reacts to grandulation, nonlinearties and other image features and flaws. Then the best way to test accuracy is by means of focus shots. We have the film specifications for the old Type 107 film including the MTF curve, but FujiFilm never did the this test for FP-3000b (confirmed by FujiFilm). There is a difference between the two films, but he has that information. Since this is over my pay grade I don't always understand what he tells me. I just attempt to do what he asks. It is interesting that I might only get 10 lines/mm as that's near what I was getting. I was told with this Koren 2003 test that I should be able to get a resolution of around 20 lp/mm. So perhaps this expectation is in error and needs to be explained.<br>

At this point all I can do is email my consultant and ask him to explain exactly what he wants done and why which could take up to a week to get a response. I really don't know how he came up with 20 lp/mm except from the FP-3000b specifications. I maybe in error in what I am saying. <br>

If you would be patient and stay tuned I'll get back to you.</p>

 

<p>I wonder if it wouldn't be easier to shoot a flat target at a slight angle, or a stepped target of parallel targets at graduated differences, then select the best focused spot to do your tests on. That way you'd eliminate the focus error problem completely. </p>

<p>Thanks for your responses. What my photo analyst really wants is a ground lens test which is beyond me. He did inform me that I missed the part were he said the 20 lp/mm could be less. As I said I would be happy to pay someone to do this test as he is not confident the Koren 2003 test is valid. The whole point is to make sure the camera is in focus. Even if the lens is good the rangefinder could be off. The photos look decent, but that's not good enough.<br>

Michael you gave the suggestion of a flat target at a slight angle. Sorry, but in the area of cameras I am a dummy. However, I am not sure this would give the results he is after nor can I get a good visual on what your suggesting. As is my eyesight isn't that good. If someone can do the ground lens test that's what we really need and willing to pay someone to do this. <br>

This probably won't help, but going to attach my last test photo using FujiFilm FP-3000b, Polaroid 100 set at f/8.8, distance 5.7 meters. The chart used<br>

<a href="http://www.normankoren.com/Tutorials/Lenstarg_50_5906p_15g_0is.png">http://www.normankoren.com/Tutorials/Lenstarg_50_5906p_15g_0is.png</a> printed at 600 dpi exactly 25 cm long.<br>

The photo was scanned in grayscale at 1200 dpi. <br>

Because of the high speed film at the f/8.8 setting I had to do this on a cloudy day as the shutter couldn't keep up and I got a dark vertical area on the left when sunny. At least that's what I was told why I got a dark area by Landcameras.com. The camera can't resolve more then around 20 lp/mm, but should I be getting better then I have. The only way to really prove it is a ground lens test. The lens maybe fine, but the rangefinder off. How that is done with a Polaroid 100 I have zero idea and I really don't need to know. It just needs to be done and my photo analyst really won't except anything else unless a better test if there is one. Thank you and I hope someone can help me out here. </p>

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

<p>Sorry, I should have added I had to reduce the file size to be under 100k. The scan itself was a 35 MB tiff file.</p>

<p>Ms. Castner: <br>

<br>

The Fuji specs here: <br>

<br>

<a href="https://www.fujifilmusa.com/shared/bin/fp3000b-ds.pdf">https://www.fujifilmusa.com/shared/bin/fp3000b-ds.pdf</a> <br>

<br>

say the resolving power for 3000b is 20 lines/mm or 10 lp/mm. A young “perfect” human eye resolves about 7.5 lp/mm, depending on who you believe. Some WWII pilots were tested at over 10lp/mm which gave them an advantage spotting their opposition. <br>

<br>

The three-element lens in the 100 should get you about 25-30lp/mm on a good film that will resolve twice that much, but only in the center. The Polaroid 360 had a good glass four-element lens that could probably go 50 lp/mm +or- in the center, down to below 20 at the edges. Back when we had much more film choices I did a lot of film/lens tests with a 1951 USAF chart. <br>

<br>

You might want to do you focus testing and adjusting on higher resolving sheet film then do your “blur” calculations. <br>

<br>

Are you aware that the film plane in the plastic Fuji FP series pack films sits at a different rear focus than the Polaroid pack films in their steel cases? Are you also aware that Polaroid changed the materials of both the negative film and print papers many times over the history of roll films and pack films? Your old Polaroid film may have very different grain and blur patterns than the Fuji FP 3000b. <br>

<br>

Also the various lenses used by Polaroid on their pack film and roll film cameras will certainly give different blur patterns along with giving different patterns as the adjustment point on the print moves away from the center. <br>

<br>

Is your project feasible without really being able to duplicate the original equipment both as to film and camera? <br>

<br>

A. T. Burke <br>

<br>

P.S. Mr. Darnton has a good suggestion about using a scale on an angle. A yard stick or meter stick placed at an angle with the center at the scaled distance from the film plane is a way to see if you are on or not. Also when calculating where the focus is, remember to adjust the distance reading using the triangle formulae, A squared plus B squared equals C squared. </p>

<blockquote>

<p>The Fuji specs here: <br /> <br /> <a href="https://www.fujifilmusa.com/shared/bin/fp3000b-ds.pdf" rel="nofollow" target="_blank">https://www.fujifilmusa.com/shared/bin/fp3000b-ds.pdf</a> <br /> <br /> say the resolving power for 3000b is 20 lines/mm or 10 lp/mm. A young “perfect” human eye resolves about 7.5 lp/mm, depending on who you believe. Some WWII pilots were tested at over 10lp/mm which gave them an advantage spotting their opposition.</p>

</blockquote>

<p>Usually lines/mm means the same things as lp/mm. Usually. Digital images tend to be rated in dots/mm, which is twice the lp/mm. Not that no-one ever gets it wrong.</p>

<p>Having the whole MTF graph, instead of just one points, is better. Especially, with just one number, you don't know what the response (%) is at that spatial frequency. It could be 10% or 50% or something else.</p>

<p>Note, for example: http://www.kodak.com/global/en/professional/support/techPubs/f9/f9.pdf</p>

<p>Tri-X is 70 cycles/mm at 30%. The graph doesn't go down to 10%.</p>

<p>Note that the MTF for film is for the film. That is, not including lens, other optical parts, or focus. It might be made with a mask, that is, contact print, maybe metal with etched holes. You have to also include the MTF of the optical system for actual results.</p>

 

<blockquote>

<p>He is trying to determine focus errors and aberrations, which the f/8.8 setting is more sensitive too. Other tests will be done at the f/42 setting.<br /> We are trying to apply deblurring methods to a old Polaroid photo.</p>

 

</blockquote>

<p>Interesting. I wonder if you have Jansson's "Deconvolutoin of Images and Spectra"?</p>

<p>http://store.doverpublications.com/0486453251.html</p>

<p>I bought it when the hardback one came out, but it is a little expensive. Then again, there might be much research done since them. Still, it is a good reference.</p>

<p>If you want to do deconvolution on actual pictures, do you want to include focus error?</p>

<p> </p>

<p>Michael is suggesting putting many targets in the image, at different distances, so you can choose the one that is best in focus. That is, has the highest MTF. Though you also want them at different places, such as near center, near edge, and near corner. You should be able to get enough in.</p>

<p>If you are careful, you can use that to calibrate the rangefinder.</p>

<p>You could put non-instant film in a film pack, to more accurately measure the optics MTF. </p>

<p>Thank you A.T. and Glen for your responses. First, I thought lp/mm and lines/mm were one and the same, but a quick check seems to indicate lp/mm means line pairs per mm and I take it lines/mm is just one line? The Koren chart is based on line pairs/mm. My photo analyst, Andres, emailed it it was around 20 lp/mm or less on the Koren 2003 chart that I should see in the test photo. I assumed that counting each black vertical line counts as a pair with the white line next to it. I should have said 10 lp/mm not 10 lines/mm earlier to be clear. The specs on Fujifilm says the resolving power is 20 lines/mm. How do I read this? As I read lines/mm was an old standard for TV resolution. "All MTF charts and <strong><em>most</em></strong> resolution charts display spatial frequency in <strong><em>cycles</em></strong> or <strong><em>line pairs </em>per unit length</strong> (mm or inch). " http://www.normankoren.com/Tutorials/MTF.html So when the FujiFilm states resolving power do they really mean 20 lp/mm or do they mean 20 lines/mm? If that statement really means 20 lines/mm then that would translate to 10 lp/mm if I am counting correctly. A.T. your statement that the 100 should get about 25-30 lp/mm is that for FujiFlim or the old Type 107? The Polaroid 100 series at the time I am dealing with only had the Type 107 film so there wasn't really any other choices except color and that's not my interest. <br>

Yes, we tried to get the FujiFilm MTF graph, but according to FuijFilm's corporate headquarters in Japan it was never done. Technical rep for FujiFilm in US call them for us. Andres says he can do work arounds without this focus test, but he really want's a ground lens test. We did get the specs for Type 107 including the MTF graph. Somewhere lost among my documents is the 107 specs, however, I can't find it, but I sent to Andres.<br>

Yes, I know the old film had a steel backed case. It's lack can cause a problem with the fp3000-b with the extra two prints in pulling the film out. You may or may not have to adjust the spring on the door. After I got a camera from Landcameras.com I no longer had that problem. I didn't realize the fp3000-b film sets at a different focus because of the plastic back, but wouldn't FujiFilm compensate for the lens focus? I do know the chemical composition of the two films are a bit different for the dye diffusion process . Supposedly the fp3000-b is finer grained. I think there is another one or two things, but can't recall off hand.<br>

I am still back to finding someone who is willing to do a ground glass test. Is this asking too much? It can't be done? Nobody knows how to do it with a Polaroid or it's too complicated?</p>

<p>Best <br>

Mary</p>

 

From another post I once made: ANSI PH3.50 (1970s) says, "Line pairs per millimeter" may also be expressed as "Lines per millimeter."

<p>MS. Castner. <br>

"I didn't realize the fp3000-b film sets at a different focus because of the plastic back, but wouldn't Fujifilm compensate for the lens focus?"<br>

Yes, but only in the cameras they made for themselves. <br>

lp/mm is twice l/mm. Mathematics does not change. Some resolution reporters do not get the distinction and have been known to use the wrong label. <br>

Both Fuji and Kodak made materials for Polaroid Instant Film, Roll, Pack, SX-70 680style and Spectra. If you are lucky, your original might be made of the same material. <br>

The lens, its iris, the light source and placement, film, paper, and processing will all determine the characteristics of the out of focus parts of the exposure. You may have insufficient resources to properly complete this project, but not for want of trying. <br>

Mr. C. <br>

Could there have been a clarification when ANSI PH-3.50 was superseded by ANSI FP-3.609?<br>

A. T. Burke </p>

 

<blockquote>

<p>From another post I once made: ANSI PH3.50 (1970s) says, "Line pairs per millimeter" may also be expressed as "Lines per millimeter."</p>

</blockquote>

<p>I believe this is right, though that doesn't mean that no-one ever gets it wrong.</p>

<p>Line pairs are pairs of black/white lines. Lines are the black lines, not counting the white, so the same. This should be true for all analog (non-sampled) systems.</p>

<p>The complication comes with sampled systems, such as digital cameras, where we count pixels. It takes two pixels in whichever direction to make a line pair, or a line different from the background color.</p>

<p>Some people, used to counting pixels, want to count black and white lines separately. As you note, that doesn't follow the standard, but does happen.</p>

<p>Thanks for your responses. I contacted FijiFilm's Sensitive Product Hotline and learned a few things. I am also interested in the difference between fp-3000b vs Polaroid t-107 (old B&W). http://www.everythingretek.com/PRDLibrary/LIBRARY8/LIBRARY/FILM/107FDS.PDF The old film has 14-17 lp/mm resolution while fp-3000b has 20 lines/mm. Convert to lp/mm so you have 28-34 lines/mm for the old film compared to 20 lines/mm for fp-3000b. This would give the appearance the fp-3000b had poor resolving power, but not the case.<br>

1) Fp-3000b has less grain and is more smooth.<br>

2) Resolving power of fp-3000b is 20 lines/mm, but Rob told me FujiFilm is quite conservative with its estimates and not only will the lines/mm vary with lot but with the randomize of the coupling of the silver halide. The resolving power of the film can be higher and 25-30 lines/mm or higher would not be unusual and similar to T-107 (28-34 lines/mm)<br>

3) Apparent resolution with the fp-3000b over the T-107 is higher because of finer grain. So the fp-3000b photo will look smoother then the T-107 film.<br>

Once the difference between lines/mm and lp/mm became clear I check my photo test again. The Koren 2003 is a lp/mm test. I counted 9 lp/mm and a friend 10 lp/mm. Convert and you have 18 lines/mm and 20 lines/mm respectively. Therefore if we read the chart correctly my results are within the parameters of the fp-3000b resolving power. I also noticed my print chart looks fainter then when viewing the chart on my monitor. I suspect if I had printed out with semigloss or luster paper and used an inkjet printer my results might have been better. </p>

<p>The sheet film idea is possibly one tool to document resolution and focus. A related idea would be to first use a piece of drafting Mylar or even a small piece of ground glass at the film plave to focus with before loading film.</p>

<p>I would like to hear more about the project driving this too.</p>

<p> </p>