john_luke

Digital files from a high end drum scanner like a Heidelberg Tango, if scanned probably, may not need any unsharp masking at all (USM). It's files created from CCD's/CMOS's that need proper USM. (Weather it�s a camera or scanner, makes no difference.) outbackphoto.com has an essay about how these chips interpolate RGB color from what is a monochromatic sensor. Most use what is known as Bayer Pattern Interpolation which needs to soften the image to get the color interpolation to work. The user then needs to use USM to mask out that softness.

 

http://www.outbackphoto.com/handbook/rawfileprocessing.html#5

 

http://www.outbackphoto.com/workshop/photoshop_corner/essay_15/essay.html

 

John Luke

APA/ASMP

I thought I heard that using a polarizer on a digital camera can be dicey because it is basically a monochromatic chip and the color is interpolated by an array with microscopic R, G, & B filters on top of the chip and use of a polarizer can sort of wreck that interpolation.

The most stable process is what you have mentioned, Crystal Archive paper on a light jet printer. Reed Photo (reedphoto.com) in Denver uses Ilford Gallerie BW in their light jet printer for BW prints.

 

If you want the feel of a water color stock, a pigment ink from either MIS or Media Street (inksuppply.com, mediastreet.com) on an acid free stock such as Hahnemuhle paper (digitalartsupplies.com) is the best inkjet technology can deliver. Avoid glossy inkjet stocks. They are the least stable.

 

Some people are even making large digital negs for contact printing using the platinum palladium process.

When desktop printers hit the scene, common output sizes were based on standard sizes of brochures/magazines in the graphic arts/printing industry. One page was 8.5x11, a doubletruck ( double page) was 11x17. Paper was manufactured to those specifications so graphic designers could make their comps and proofs at full size. Photographic print sizes, 8x10, 11x14, 16x20, never correlated to those of the graphic design field. Those photo dimensions go way back to some ancient rule of proportion.

Many pro-sumer devices operate in sRGB. Try assigning that one first. (just a hunch).

 

White balance is color temp. Go to digitaldog. Andrew has a neat printer test file that is a free download. It has flesh tone, and grey scales on it. You could also use this to see if D50, D60 or D65 looks good on your monitor. One will look good, while the others will look either too warm or too cool.

 

Only "Raw Processing" software has a colortemp slider. If I have a file that looks good in density, but has a unpleasing color cast, I go to curves in PS. Click in the middle eyedropper, and look for an area in your shot that you think should look nuetral, like a shadow area. Click the dropper on that area and that will "neutralize the entire shot. If it moves it to a color cast you don't like, undo and click on another area. You may have to try several areas to find the one that does the job. (Could be tough for shots that have no neutral areas, like a sunset.) Other than that method, play with the curves in the various colors, or use the sliders with the adjust color balance sliders, paying notice to casts in shadows and highlights

If you are viewing the image in the wrong colorspace/gamut, it may render over/under satutrated, too ligh/dark, or off color. It will basically look lousy. The cameras grey or white balance and the monitors white point are 2 different things.

 

The cameras balance is basically either a preset or a custom function that will render neutral across R, G, & B. . If it is a simple camera, it will have either selections as Daylight, Tungsten, Flourescent or something similar. If it is a high end camera that allows a custom balance, do not grey balance on a grey card. it's too dark. Use a MacBeth chart and use the 2nd or 3rd patch from white on the MacBeth chart. Look for something around the 190 on the PS densitometer. Once balanced, those patches will read sameRGB values.

 

Monitor white point is the color temp or how "blue" the white gets. I use D50(5000K). Some people think that is too warm. Most of my work goes to a printer for brochures, so D50 has been that standard for quite some time. People doing web only graphics will want to use D65. Inkjet print guys may ues D50, D60 or any of the above. A monitor calibration package will allow an infinate amount of control, where you monitor set up in the OS may only give you a choice of D50 or D65. What OS are you in? I am proficient only in the MAC OS , so I could only tell you how to find it there. Under the Apple, go to monitoers, then to color. There will be a step by step process to do a simple calibration. Or if you have Adobe Gamma in your system open that and go through that process. Once properly set, compare a sheet of proofing stock or your favorite paper under your 5000K print viewing booth (got one- right?) Its whitness should look similar to the whiteness you have selected as you monitors white point.

Never select the monitor profile as a working space, nor assign it as the embedded profile.

 

That being said, try "assigning" the various RGB profiles, as you have been trying. Find the one that looks closest to your level of acceptance. If GenericRGB looks closest, assign it. The less moves you have to make with levels, curve, color, etc, the better. If you overwork those edits, you are losing image data and risk banding, posterizing and a whole host of image degradation issues. This can be observed by checking the histogram under levels. It should be smooth. An overworked file will have a spikey, or choppy histogram. If you assign a profile that gets it close, you will not degrade the image. If you want to set up a consistent work flow, and others may be opening your files, you could then "convert" that finished image to AdobeRGB. (converting will not change the way the image looks, but rather changes the name of the profile.)

Consider the Rodenstock 45mm. It's wider, maximun aperature is f4.5 (every little bit helps when focusing in dim light) and has an image circle of 131 as compared to the 47mm with its 125. I've used the Rodenstock on my Arca Swiss 6x9 for years and have never been disappointed

Depends on the degree of perfection they want. If the museum has an attractive façade, suggest the car be placed outside, and photograph either on an overcast day or when the sun is at the horizon. Hosing down the pavement gives a nice look also. (keeps it from looking washed out and helps mask oil and tire stains.)

I am on a Mac. I was working in ColorMatchRGB and had my monitor calibration system set to a gamma of 1.8. This is the Macs native gamma, and ColorMatch was designed

to render at a gamma of 1.8. Everything was peachy. Now, most of the professional digital cameras have AdobeRGB as a selectable colorspace( no ColormatchRGB available for

digital cameras), and the current US Prepress Defaults in Photoshop are AdobeRGB. I am switching over to AdobeRGB to get in sync with the rest of the world. In an article by

Rich Adams from Graphic Arts Technical Foundation, AdobeRGB was designed for a gamma of 2.2. To be true to AdobeRGB, do I recalibrate at 2.2 and disregard the native

Mac gamma?

Bob- this is easier said than done. Due to the design of the Arca GG frame, the actual GG sits on metal "pads or feet" and is slightly larger than the fresnel. If I remove the fresnel, the GG still sits a top these 4 "feet" at the same plane as it did before removing the fresnel. To bring the GG closer to the lens means having Steve Grimes shave/machine down these 4 "feet" the desired amount.

This thread was discussed 2 weeks ago! It is in the archives also! I replied 2 weeks ago and jokingly stated that it comes up every 2 weeks, and here it is right on schedule!

Sandy-

 

This could open up a can of worms here as my GG is placed to accommodate the factory fresnel which is 90 thousandths of an inch thick If I remove it or replace it with another one of different thickness, I will most likely have to machine the metal frame to allow the GG to be reset in acccordance with the 1/3 rule. Drat those bloody interior fresnels. I guess thats why Linhoff switched their design to an exterior fresnel.

Sandy- I spoke with the Arca dealer, distributor and authorized repair technician. They told me that the Arca factory supplied fresnel is supposed to work with all lenses. I've heard what you are suggesting before, but have not had success in trying to find out specifics form anybody familiar with the Arca F Classic. I suspect you are onto something, so if you have any advice or sources you can direct me to, that would be appreciated.

 

John Luke

James-

 

A pixel is the smallest discreet piece of information you can have. It is a specific size, and a specific color. How many dots the printer lays down to make that pixel is what the output resolution is all about.

 

Assume an original file that is 1� x 2� @ 360 ppi for all examples. If your file size is 360 ppi, the image has 360 �pixels� for each inch in one dimension, lets say height for example. The pixel size and print dimensions stay constant for each example.

 

If you set the printer output size at 360dpi, each incoming pixel will be printed by 1 dot. (By the time the printer get to the next pixel, there will be some gaps between the pixels, true continous tone does not exist, the tone looks choppy)

 

If you set the printer output size at 720 dpi, each incoming pixel will be printed by 2 dots

 

If you set the printer output size at 1440 dpi, each incoming pixel will be printed by 4 dots. (At this point, by the time the printer gets to the next pixel, it has pretty much filled in all the space from the last pixel it printed with tightly laid down dots, creating the illusion of "continuos tone")

 

If you set the printer output size at 2880, each incoming pixel will be printed by 8 dots

 

From what the people at TSS Photo say, and this has been confirmed at a numbr of other forums, the Epson driver disregards anything above 360 ppi going into it. It is simply too much data for any purpose. Settle on 360ppi or 240ppi as your �standard� file size, and control your quality of print by using the printer resolution setting. Matte papers for instance, usually won't resolve anything above 1440 because the dots spread slightly on the matte surface and the image looks as good as the paper will allow it to look. High gloss paper may allow you to see a difference between 1440 and 2880, but it really, really hard to see with the unaided eye.

File size:

 

Send your files to the Epson at either 360 or 240. Anything higher than 360 will choke your system as you have experienced and has no increase in quality. The Epson print drivers algorithms round better in multiples of 720 as mentioned earlier.

 

 

Printer resolution:

 

I personally see no difference in output settings of 2880 comnpared to 1440 other than more ink consumption and slower output times. People who make large prints on the 40" wide machines often output at 720 as these huge prints will most likely be viewed far away.

 

This ppi to dpi stuff makes us all crazy.

Kelly, I hope you are charging the attorney "by the hour" for all your fixes in post (in 1/2 hour increments, after a retainer of $500 up front, of course) :-)

If you don't anticipate adding long rails for really long lenses doing close -up work, you will have better success with the Horseman. I've used the Horseman, the Sinar F, and the Arca F Classic. At least with the Horseman, you get gear driven standards. The others, with their free falling standards can be a bit of a pain for catalog work while you are composing under a darkcloth. The Horseman is really beefy, durable, and if you ever upgrade to a Sinar P, your lens boards will fit. The Arca F is great for architecture or location guys when you can take the time to carefully set-up and work, but can be a bit delicate in the fast paced rough and tumble studio catalog environment (The Arca is pretty pricey as well.)

Thank you Julio-

 

I had so many issues trying to get my 47MMXL images sharply focused. My detents were ever so slightly out of perfect parrallelness, I was getting funny tilted planes of focus going through my images. I had the detents reset to factory specs, and I also use the Zig Align each time I mount my 47XL to insure perfect parrallelness. I also discovered the entire GG & frame assembly was defective, it was off spec, so I bought a new one. These errors "compounded" together made focusing at infinity with my 47XL show up as focus at 5' on the film. Too much for hyperfocus to "take care of". It took months of research on this forum, lots of dealing with camera technicians, and a lot of agony to trouble shoot and eventually solve the problem. I looked at everything. I tried various loupes, had my eyes checked, measured my film holders with a micrometer, checked my lens board, and was eventually ablte to isolate the variables that were causing trouble.

 

The Arca, with its interior fresnel is an achilles heel as I found out for those who use ultra wide lenses. According to the Wisner artical from the link posted earlier, it was mentioned the when using inside fresnels with ultra wides, the focus at the corners of the GG will not be accurate as the light is so off axis it comes through with huge focus shift even when the GG and fresnel are properly placed. When using my ultra wides, I only focus in the center of the GG, or directly over the rear lens element if using a shift or rise. Unfortunately, the way the Arca is designed, if I wanted to abandon my interior fresnel and replace it with one of the many high quality outside fresnels, I would have to have my GG assemblies machined by someone like Steve Grimes to accomplish this for both my 6x9 and 4x5 backs. I'll just live with these shortcommings now that I know how to deal with them. I would advise anyone who desires to do landscapes with anything wider than a 75 to not buy a camera with an interior fresnel, period.

 

I view these forums as invaluable and thank you to everyone for this free exchange of ideas!

"And I don't recall ever making an out-of-focus negative."

 

Bruce, as discussed very thourougly above, focus shift occurs only when placing a fresnel between the GG and the lens. Your placement will result in no shift, and hence no GG compensation measures needed. For those who do mount fresnels between the GG and lens, the focus shift often goes un-noticed unless using a super wide lens, where the depth of focus is about nil compared to a medium or long lens.

I bought one, but don't use it. It's 6X, so I end up seeing too much texture on the GG. My 3.5X Toyo works much better. Also, I just read an artical on the Wisner site about fresnels, and if you have a fresnel on the inside, such as an Arca like me, it can be dicey when using ultra wides such as the 47mmXL. According to the artical, if you try to loupe the corners of the GG, which is what the tilting loupe is for, with an inside fresnel in place, you may be in for a surprise. Apparently, due to the extremely close proximity of the lens element to the fresnel, and the fact that at the corners, it is extremely off axis, the fresnel no longer gives an accurate indication of true focus. When I use my 47XL, I zero the camera movements, I focus by using my Toyo loupe placed flat in the middle of the GG, (or if I have used a shift, or rise, I place the loupe flat and centered over the lens element) and I rely on hyperfocus or depth of field to take care of me.

AC-

 

There was a posting earlier on the Rob Galbraith Forums about someone like yourself who was using ColorEyes to profile his camera, but now that he has bought ACR, it does not recognize that. Apparently for now, its either ACR or your custom made camera profile, but not both. :-(

 

Regarding making edits in post, as long as you are still in 16 bit, you still have the ability to make these moves, as any loss in small amounts of data usually aren't missed when you flip to 8 bit. I said "usually". It would appear to make the best use of ACR by doing it all there. Do it both ways, flip to 8 bit and take a look at your histograms for combing, or the "fingers of death"

Both is possible! Of course with the consideration that focus shift can happen if the fresnel is placed between lens and GG.

 

Rick- not "can" happen, but rather "will" happen. Again, back to my Arca with the factory fresnel on the lens side- if I take a plunge micrometer and get a measurement in thousandthes of an inch from a fixed point to a sheet of film in a holder, and take another measurement from that same fixed point to the side fresnel that faces the lens, the micrometer tells me that the fresnel is actuall closer to the lens than the sheet of film by 60 thousandths of an inch. If I remove the fresnel, leaving the GG in its original position, the micrometer tells me that the GG is further away from the lens than the film is by 30 thousandths of an inch. Neither reside at the same plane of the film. (The fresnel is 90 thousandths inch thick, so 1/3 of that is 30 thousandths, which is the compensation that I measured.) The compensated placement of the actual GG is further AWAY from the lens by exactly 30 thousandths of an inch. That is 1/32", or the thichness of 3 sheets of film. With all the pains we take to insure correct and flat placement of film, some people taping film in the middle so it doesn't bow out, some people tapping the film holder to get it to settle, etc, etc, this is a serious amount to be off. Don't neglect it.

I have had no problems affixing my Lee resins to the outside of a screw in polarizer. A side note if you use a circular polarizer in a series adapter ring (ie, threadless): There is a specific orientation to a circular polarizer that must be observed.

Sandy and Doug-

 

Yes, the my Arca has the fresnel on the lens side , as it is designed that way. Linhoff used to be that way also, but changed to an externally placed fresnel. The exact position of the Arca GG and fresnel are designed with 'fresnel focus shift' compensation taken into account. If I decide to do away with the factory fresnel from my Arca, I will have to machine or mill the GG frame to allow me to set the GG in closer to the lens, not where the fresnel used to be, not where the GG used to be, but at a unique point based reversing out that 1/3 compensation factor. A job best left to someone like Steve Grimes. If you take an ordinary GG (no fresnel) and add a fresnel to the lens side, you will find that the focus has shifted, causing you to plant proper focus at a different point than where the film sees proper focus. When fresnels are added to the inside, the usual compensation is 1/3 of the thichness of the fresnel. This may seem like a slight amount, often disregarded, and if not done, focus error often goes undetected if using standard or long lenses, but if using super wide lenses, such as a 47mmXL, any incorrectly placed GGs will show up here in the form of out-of-focus images. Threre are many threads in the archives about this, as this topic comes up a couple of times a month.

 

Please be careful!