alan_schietzsch

I've use these units for literally hundreds of thousands of catalog photos, and they are more reliable than the very expensive Elinchrom units we also

had in house. I still have two packs and six heads or so. Sometimes you can buy entire used heads for about the same as a flashtube alone, so I

wouldn't be too concerned. The flashtubes last a VERY long time, and are usually only replaced if damaged by impact and clumsiness, chnaging

softboxes or rings too hastily (oops- how do I know this? shoots that lasted til 4am.) They are very reliable. The modelling light bulbs have delicate

filaments, but unlike flashtubes, they are regular tungsten halogens that are cheap and plentiful.

Assuming it's a good scan (or a digital body) there is nothing preventing a

very good quality image in using Photoshop perspective control.

(Remembering with a grain of salt that digital originals do retain quality when

severely interpolated somewhat better than scanned images in practice.)

Rather than simply increasing the width at one end, or decreasing the width at

the other, it's important (if you want the proportions to be acurate) to do BOTH,

so that the middle of the image does NOT change in width. This method

reproduces what a view camera properly does (pivoting on the axis).

It all depends how severe the cast is. If it's moderate, a Color Balance

adjustment layer is a non-detructive solution you can easily change to suit

your judgement.

If it's truly a horrible cast, going into LAB mode will enable much larger color

shifts than are practical otherwise; you can recover photos that you'd think are

lost forever. Just work on a COPY - it takes "some getting used to" the radical

nature of even very tiny changes, when in LAB color space.

Canon sets the native resolution of their DSLRs at 180 dpi, which despite

being lower than the much touted 300dpi standard, works exceedingly well.

The 300 figure comes from using a 150 line screen in the printing industry,

and having a 2x oversampling rate for safety. It does NOT directly relate to

direct digital imagery, and certainly not to inkjet prints which use stochastic

patterns rather than conventional screens. More than one inkjet dot mix to

reproduce one pixel, so the printer res. does not particularly matter.

Certainly having the print driver do less math is convenient - for the print

driver. But we buy computers so that THEY do the work...let them. Modern

drivers are perfectly capable of doing the math, so all that really matters is

image quality. In a nutshell, 180 is good for general use, 360 is the max

EVER needed. Don't believe it? Try it and be pleasantly surprised. What really

matters for printed image quality is sharpening and level/curves gradation.

Resolution just isn't all that important.

I'd concur with Wayne's advice because each scan and image size is

different; but yes, ALL digital images should be sharpened. And I used to

operate a drum scanner using OPTICAL unsharp masking, so sharpening of

scans IS needed, for those who don't believe that. It was so 15 years ago, it

still is now.

I tend to change the order just a touch; the principle is the same:

First set image at the final size and resolution, and view at 100%, preview ON.

Set Amount to max (500%), Threshold 0, and adjust Radius so halo is

BARELY visible. Increase Threshold until objectionable noise disappears,

probablt set to 3 or so. Then reduce Amount until it looks slightly "over"sharp,

not garish.

Note that the image will always be somewhat softer in ink or as a print than on

screen, since inks and dyes will diffuse, and dot patterns "dither" to the eye.

So what's slightly "too sharp" on screen may be correct on paper.

There's more on Sharpening at the forums on interpolatethis.com.

This works, assuming your scanner has sufficient dynamic range. One trick is

to scan only the centre part of a frame and then median/blur it to a consistent

frame where every pixel is the same colour. This eliminates any grain

contribution from the mask frame.

Unless you shoot wider than a 20mm on film (which you can do

with a 14mm on a D60 or 10D), does it REALLY matter if it's full

frame?

Consider that the smaller sensor uses the "sweet" central

imaging area of the glass, has telephoto advantages, AND costs

less. Plus, it's available now.

If the histogram is centred, the scene is low contrast. You can

increase the exposure until the histogram ALMOST touches the

right side. If it touches (or goes past) the right side, you are

overexposed and will have "blown out" highlights.

I can confirm that the D60 is indeed discontinued, but Canon

refuses to say what the replacement is, and will not confirm (or

deny) the rumour of 2 models replacing it. They have said that

the new model(s) will be announced March 1 in Las Vegas.

James Harvey (wunch_of_bankers@hotmail.com) writes:

"My hands were *shaking* because of the amount of money I

was spending....the prices really are Scary for those of us in the

real world..."

My sympathies (genuine!), but I've got to ask for yours, too.

Think of this: you had a choice to spend the money or not. We

have a lesser choice - this is our JOB, so if we can't get the shot

because we didn't want to spend the money, we're FIRED by the

client. So we're scared if we spend and scared if we don't. This

isn't play, it's survival. We ARE in the real world. Very much so.

You're right in saying that some lens-body combos have issues,

that's "Does Lens 'A' work OK on Body 'C'?", a totally legit (wise,

even!) question.

"Lens A vs. B" is a different, non-body dependent question, that's

what bugged me.

What about taking a class and buying the educ. version...not to

save money (tuition + edu. = full price), but to learn how to fully

use PS as well as to own it? After all, it's not PS that does the

image processing, it's the skill of the operator. That way you win

2 ways.

I used a 1015 for a number of years; I used oil mounting rarely,

only if newton rings were visible.

When the drum is absolutely clean and the transparency is not

"oily" (sometimes E-6 processing leaves a very slight "oily"

finish) there is rarely (<10% of the time) any problem.

If it is working OK, don't worry about oil. When you need oil, it will

be very visible.

I had a Screen 1015AI for a few years, and it was great; real

optical USM, not just software, and bulletproof construction,

totally reliable. For a small volume of 4x5's it's ideal.

I echo the suggestions; start with a new light source and a good

clean drum, get a quick lesson from any scanner operator

(printing shops can network you with someone) and you're good

to go.

Is it just me being too picky after a long day? I keep seeing (what

to me is) the same question:

- has anyone tried Sigma 24-70/2.8 EX DG DF on an EOS 50E?

- Canon 24mm f/2.8 vSigma EX DG 24mm f/1.8 on an EOS 50E

- Use of Tamron/Sigma Lenses on EOS 300

Do people think the lens' characteristics or performance

somehow changes when used on a different body? Does the

glass magically grow somehow? Yikes.

If I compare a 200/1.8L and a Promaster 28-200/5.6-8, will the

200 be suddenly softer than the cheap zoom on an EOS 1v and

then sharper when on an Elan? What has the body got to do with

this?

"I have a 1/2 inch socket wrench and currently own a 1/4, a 3/16,

and a 5/8" socket. I'm planning to drive my Ferrari to Chicago.

Which size socket should I buy next?"

Maybe I'm a cranky 'ol photographer who needs a Tylenol. (-;

Hope I didn't offend anyone, just blowing off steam at something

that seems annoyingly repetitive to me right now. Cheers, and

hope I made you laugh.

I've done this experiment, and have found that while it certainly does not

produce a true profile (I'm agreeing with Ellis here) it *IS* useful in terms of

hardware calibration, which you can then fine-tune much more effectively with

colorsync and the monitor calibration control panel.

I have a Sun/Sony 20" Trinitron, and a 17" Viewsonic on the side that I use

for palettes and such. To make the two match as closely as possible BEFORE

software calibration, I did what you're proposing, and it worked far better

than my attempts using the software utility alone.

My method was to make a 21 step greyscale file in Photoshop. Because it

was generated from the numbers I typed in, I knew that the color should be

exactly neutral; any deviation was due to my monitor.

Then I measured each patch and tweaked the monitor's R, G, and B guns until

there was no magenta-green or warm-cold crossover from highlight to

shadow. It took a few iterations until the curve was neutral, but this was

WAY better than "out of the box."

So, NO, it doesn't generate a true profile, but it DOES help; with some

monitors, it helps a lot:

I tried this on a brand new NEC monitor being installed at our local RCMP

forensics lab, and it was incapable of being adjusted enough to be neutral.

With the controls maxed out, there was considerable warnth in highlights if

shadows were neutral, or we got blue shadows if highlights were adjusted to

be neutral. We replaced it with another brand, which was within 3 CC points

of perfect, right out of the box!

I used a Gossen Colormaster; I imagine a Minolta would be no different.

I should have also mentioned - since you say the original is in color - you can

convert to L*A*B Mode, guassian blur the A and B channels, leave (or even

sharpen the L channel) and convert back to RGB/CMYK.

Perhaps the problem is in terminology; what you have posted looks like a

digital photo (or scan) of a PRINT of a house. Did you not post a digitally

photographed laser print, newspaper, or other IMAGE of a house?

There is moire visible in the siding; that's because the frequency of dots in

the PRINT is close to the frequency of lines in the siding.

You can either print with a different dot/inch setting, change the dot screen

angle, or blur the image.

I use the B&W for exposure testing, too. It gives a much better idea of

contrast and density than the colour stuff.

I find colour Polaroid materials are VERY sensitive to development

temperature and time; I find they're often a bit cyan when cold, and I like toi

develop them longer than the info sheet suggests.

Try a few variations and see what works for your circumstance.

Wow - John, Wayne, Johan, and the rest - you've made some really good

points, and given me a lot to think about. Which is exactly what I'd hoped for

in this discussion - what a great bunch you are!

I'm in a location where it's not easy to rent 4x5 if I should need it; then again,

I've only had one request for 4x5 work in nearly a year, which is what made

us start to think about selling it. So the demand doesn't seem likely.

Of course the minute I don't own it, it'll be in great demand, right? (-;

Interestingly, we *DID* own a drum scanner until recently; scanning business

has dropped off dramatically, just as digital demand has risen, and mid-small

"semi-pro" scanners have become more common.

We figured we should sell the drum while it was still worth something. It's

easy to get scans done at a number of places nearby.

I'm beginning to think that the smart thing to do would be to sell the Sinar,

even though you can tell I really like the machine. The proceeds would likely

go to an ultrawide lens for digital use (maybe a 14mm) which would be very

handy for interiors; the last architectural job I just did, the architect asked

specifically for digital! I have to admit that perspective correction and

removing signposts and wires is easier than ever.

I figure I will earn more from having a 14mm digital than from the remote

possibility that more 4x5 work comes in...and if the larger digital backs

become cheaper, it sound like I can always get a used Sinar again on eBay.

That's how it looks to me - what do you think? Have I missed anything?

We've had ours for over a month; it may take a little phoning around to find a

dealer with inventory coming in soon, but I certainly wouldn't say they're not

available. I *am* glad we sought it out, as we've had several digital jobs

come along lately that it has been well suited for.

Hopefully it will earn enough to make the 1Ds a reality sometime! Wow,

things are changing faster and faster.

Alan Schietzsch

ARS Technical Images

Edmonton, Canada

Our studio has a Sinar P 4x5 monorail that we're not using all that much, and

we're trying to decide if we should sell it.

It's a truly superb camera, but we have to consider that we're a business

rather than doing it only for the pleasure and art. More of our clients are

asking for digital than 4x5 chromes. Yet I love the camera...solid, smooth, it

feels like you're doing something tangible when creating a photo with it.

How active is the market for this sort of equipment these days, would *you*

keep or sell it?<div></div>

The only real difference between a "home" and a "professional"

studio is your budget, and that's subject to debate.... My studio is

at my home, yet I've been a professional shooter since '85. So

the real question is: what do you want to shoot? And do you have

a style of working that you prefer? Once you've decided that (not

that it's an easy question, that's for sure!) the rest of it will kinda

answer itself. If you want to do group portraits, you'll need to set

up a larger area, with broader fill light, for example. Let everyone

know what you want to do with this studio, then we can say

something more helpful. Be specific, then we can be, too.

Robert's comments are right on the mark. In the book

Professional Photoshop by Dan Margulis, the author

demonstrates the use of Photoshop's info pallettes by setting a

completely colourblind person against a normal-vision scanner

operator. He shows how paying attention to the numbers is

important in scanning, especially in critical pre-press work, and I

recommend it wholeheartedly. If you understand the whole book

(which may take a while) you'll be able to get a second job in the

printing industry! (-;

I have to agree with the other posters - they are all sharp. Let me

Illustrate: while working in a Hassy-based studio one of our C/M

bodies packed up, so I brought my own Bronica GS-1 in so we

could stay in production. When printing the negs, there was NO

difference in colour pack or density, and the sharpness was

visually identical. This is with the same lights, subject,

background, and photographer, so the camera/lens is the only

variable. The flyer was completed on time and no difference (we

were corncerned BEFORE we tried it) could be seen. Maybe

instruments could see some difference, but 2 pros, our

manager, and the client's eyes couldn't see any. What else

matters?