mab

And, for the sake of completeness, full size (4288x2848) jpgs of the

above:

<p>

<a href="http://www.crypto.com/private/d2x-sample1.jpg">

http://www.crypto.com/private/d2x-sample1.jpg</a>

<p>

<a href="http://www.crypto.com/private/d2x-sample2.jpg">

http://www.crypto.com/private/d2x-sample2.jpg</a>

<p>

These are about 2.5MB each.

<p>

-matt

And here's a non-compressed NEF (about 20MB). Again, not art, just

intended for testing.

<p>

<a href="http://www.crypto.com/private/d2x-sample2.NEF">

http://www.crypto.com/private/d2x-sample2.NEF</a>

<p>

Note that although I have a fair amount of server bandwidth (I've survived several slasdottings without serious injury), a zillion people downloading 20MB files all at once may expose otherwise untested limitations in my system. If that happens, I'll have to take the links down.

<p>

-matt

Here's a quickie sample I just snapped (didn't have any D2X raw files on my laptop handy, but I did have a D2x, so it was easy enough to make one. It's not art, but it should do for your purposes). This is compressed NEF (about 10MB). Uncompressed to follow...

<p>

<a href="http://www.crypto.com/private/d2x-sample1.NEF">

http://www.crypto.com/private/d2x-sample1.NEF</a>

<p>

-matt

Who claimed that glass filters were never useful with digital? ND filters (and especailly ND grads), polarizers, and filters that adjust light outside the range of the sensor (such as IR filters) are very useful. But color correction filters are not; their intended functions can be completely subsumed by arithmetic correction of the RGB data.

 

I don't mean to be harsh here, but while many questions in photography reduce to subjective matters of taste, others can be answered objectively based on science and engineering. The behavior and limits of (ideal) optical and digital filters is one of them.

As I wrote above, at edges of the historgram, there may be some benefit to using optical color filters on digital capture, but this does not occur under normal lighting conditions or with the very modest color correction filters (e.g., 81A) being discussed here. All a color filter does is act as a neutral density filter on each of the RGB channels unequally. That it, the filter shifts the histogram of each of the color channels to the left or the right a bit. If there is room on one side or the other of the historgram, as there should be under normal exposures of ordinarily-lit subjects, using a perfect optical filter and doing the same operation digitally will have exactly the same effect. (And if there isn't room on the edges of the histogram, you're probably using specialized color measuring equipment or using special techniques like IR or UV photography). Naturally, shooting in raw mode and using flexible software is required to take full advantage of this.

 

The only reasons to use optical color filters with digital capture are if you're doing very unusual stuff with color shifts that are beyond the range of the sensors (in which case you aren't using an 81A filter), if you need finished jpg pictures right out of the camera, if you've not got software that can do proper filtering, or if you just want to get more use out of the expensive filter collection you've built over the years.

 

This is one of those things that manages to be true not only in practice but in theory, too. That is, if you've got a counterexample, it would be very significant and would attract considerable interest. You'd be proving wrong not just me, but also some of the basic foundations on which (digital and analog) photography is built.

John,

 

Thanks! Their web site's design was very effective at hiding that from me!

 

It doesn't sound like the differences in the "pro" version are especially significant in the steady state if conversions are done as images are imported in, although it might be somewhat more convenient for converting a large existing base of files.

 

-matt

Why would optical versions of those filters would be preferable to doing digital transformation on the capture that ave the same effects (but without the degradation caused by imperfections in the filter, increased flare, etc).

I've got a Sinar monorail (on which lives a 6x9 back) and a GX680. I'd say I use them about equally for the 1/3 or so of my that I can't easily do with my D2x. The monorail is certainly more flexible (both literally and figuratively), but if I can comfortably get away with the GX680, I much prefer to use that. It's very quick to set up, the lenses are well optimized for MF (sharper within the MF frame + movements than any of the 4x5 lenses I have), and it has many of the conveniences of a regular MF SLR like interchangable finders and electronic shutter release (I especially like being able to trigger the shutter with my pocket wizards). But it is limited in the kinds of movements it can do).

 

If I had to pick one, it would have to be the monorail, but I do like the GX680 very much when it works. A basic GX680 system is very inexpensive on the used market these days (as is a basic monorail system).

 

matt

Forest: what made you get Pro insted of LE? Presumably someone told you what the difference was; can you let me in on the secret?

 

Thanks

 

-matt

Right, that same address is pointed to as the demo for both LE and Pro.

 

-matt

Tom,

 

I've so far been unsuccessful in locating this chart (and not for lack of trying, I assure you). Are you using LE or Pro?

 

Thanks

 

-matt

Also, for what it's worth, I've got PS CS2 (with ACR 3.1) and Nikon Capture. I've have mostly switched to ACR 3.1 for the D2{hx} files, but would like to consider the Phase One software, given how many people seem to like it here.

Can anyone explain the difference between the "LE" and "Pro" versions

of the Capture One software? I've read many rave reviews of this

software, and I've navigated the Phase One web site many times, but

for the life of me I can't figure out exactly what the difference is

between the two packages (other than the price). They have a trial

version of the LE package, but not the Pro package.

 

Right now my main sources of raw bits are Nikon D2x and D2h cameras

(plus a Minolta Multiscan Pro film scanner from which I get TIFFs).

 

Thanks!

 

-matt

Just to amplify the above response: The only on-lens filters you need for a digital camera under normal shooting conditions are polarizers, IR filters and ND-grads (and, perhaps, if you're shooting in very bright light or need less DoF than you can otherwise obtain, straight NDs). Simple color correction filters are unnecessary.

 

Why such categorical advice? Because any color correction you can do with a single color over the entire frame you can do with less image degradation in digital post processing. Any optical filter will degrade lens performance; color manipulations, except at the edge of the histogram, generally do not (and shooting in high bit raw mode gives you a lot of room to play with even near the edges).

An aside: It's worth understanding just what the Sto-fen dome does, and what it *doesn't* do. What it does do is convert your directional, narrow-angle flash into a "bare bulb" (at the cost of a stop or two of light). As you've seen, this effectively widens the spread of light (e.g., to cover more of your angle of view with a wider angle lens than the flash head is designed for), and for this the device is useful in almost any shooting environment. But it's not a "soft box". Any "softening" of the shadows is due to the bare bulb effect of stray light bouncing off the environment (light colored walls and ceiling), not from any "soft" (large relative to the subject) light being emitted from the flash itself. The light source remains just as small as it was before, and so in the absence of large surfaces bouncing the stray light around, your lighting remains hard and specular, with the associated pronounced and usually ugly shadows.

 

What this means is that while these devices work well to soften the lighting in many indoor environments, they are usually next to useless for this purpose outdoors or in large, cavernous spaces (except for widening the angle of the light spread).

 

-matt

It's above your weight spec, but I'd seriously consider a 12 inch Apple powerbook as a traveling laptop. Very nice screen in terms of color and consistency, although not especially high resolution (1024x768). Current model has a 100GB drive and a DVD burner. And it runs Mac OSX, which most people find a welcome relief from Windows.

 

I use my 15 inch powerbook for photo editing regularly.

 

Make sure you take into account the weight of all the accessories (especially the power supply and perhaps an extra battery pack).

Every DSLR has this problem - interchangeable lenses expose the mirror box to dust, and the shutter exposes the sensor to the mirror box...

 

You can slow down the collection of dust by taking care to minimize exposure when changing lenses, but any DSLR will collect some sensor dust with time. I end up cleaning my somewhere between every week and every couple of months, depending on use.

 

Since it can be cleaned off, it's not a huge problem. (And my DSLR sensors seem to collect less dust than the glass plates in the carriers of my film scanner, so at least in that respect digital capture has some advantage over film).

 

-matt

A digitial camera (or an MF/view camera with a Polaroid back plus a warehouse full of Polaroid film) is an ENORMOUS aid to learning studio lighting.

 

I don't know what exact Speedotron system you're thinking of, but those are good units. 4 flash heads is enough for almost any common setup, especially aided by an arsenal of reflectors and gobos. (In fact, you could profitably start with just two and add more only after you feel you've mastered one and two light setups, which can be very complex by themselves).

 

You'll want a stack of white and black foamcore boards to reflect and control light, some umbrellas and/or softboxes that fit your flash heads, and heavier light stands that you think you'll need (the ones included in "kits" are never sturdy enough). Seamless paper backgrounds are nice. You'll also want a flash meter.

 

"Light Science and Magic" by Hunter and Fuqua is an excellent textbook for learning principles.

No, You want your working space to be something standard like Adobe RGB. You also want Photoshop to know what your monitor profile is and automatically convert from the working profile to your monitor when it renders on your screen (in Mac OSX, you do that in the operating system simply by loading in in your display preferences from the system preferences menu; no idea how you do this in Windows, but there's a similar operation somewhere). Before you send to an output device, you convert the file to the output devices's profile.

As others have pointed out, you don't match your monitor to the printer, but rather convert your files to the printer color space before printing.

 

However, you *do* need an accurate profile of your monitor if you want to make any meaningful adjustments to the image with your computer, and you simply can't do this properly without a hardware device (such as, e.g., the Eye-One). If you're doing any kind of digital editing, especially where you care about color, this isn't an optional luxury; there's no alternative. Either borrow one from someone local, or take the plunge and buy one (ideally, you want to re-profile your monitor every month or so).

A 50mm lens on a standard 35mm camera has about a 45 degree angle of

view (in the longer axis). 45 degrees (give or take) is roughly the breadth of normal human visual perception.

 

For portraits, the conventional wisdom is that people look good from a perspective that's at least a couple of body lengths or so away, often more. Otherwise,

facial features that are closer to the camera seem disproportionally large (e.g., the nose seems to grow). You can use your "normal" 45 degree lens at this distance, but the subject will be small in the frame, which will require more enlargement and cropping to produce a tight composition. To fill the frame with the head or body at that distance without cropping, you need a narrower view than the the "normal" lens gives (e.g., 30 degrees or less, depending on how tight you want to be). Filling the frame rather than cropping also allows a higher reproduction ratio between the sensor and the subject, which gives narrower depth of field at a given apperature, helping to separate the subject from the background.

 

While it's true that 50mm lenses are often among the sharpest and best corrected made by a given manufacturer (and also usually among the least expensive), with a 35mm frame that improved quality would almost always be outweiged by the degradation from increased enlargement and cropping that you'd need to do for portraits taken to give a natural perspective.

 

(One of the nice things about small DSLR sensors is that your high-quality, cheap 50mm lens suddenly starts to become a reasonable portrait lens, especially for wider-than-headshot compositions).

Part of the reason for confusion here is that in informal, colloquial English, "perspective" means simply "point of view," and implies both the position of the viewer as well as the breadth of the view.

 

In photography (and art and graphics generally), however, perspective is a precise technical term referring only to the relationship between objects as viewed from a particular position. It depends only on the viewer's (camera's) position, not on its angle of view. No matter what lens you use or sensor size you have, you always get the same perspective from a given position/sensor plane/film plane.

 

The lens focal length and the sensor size together determine the angle of view that you *record* from the perspective that your camera position gives you, and if you want to preseve the angle of view when you change to a larger or smaller format, you have to change the lens focal length. But you get the same perspective.

 

Note there are other factors that come into play that affect how an image looks or that might seem to change the perspective without physically moving the camera. The first is if the film/sensor plane is moved -- tilted up or to the side -- which changes the point of view and hence the perspective. (View cameras are designed to make it easy to do this). Another is depth of field, which depends on the film/sensor size as well as the apperature.

 

The bottom line is that a camera with a smaller sensor/film size requires a shorter focal length lens to record the same angle of view as a larger sensor/film size at a given perspective (camera position).

 

-matt

Raymond,

 

For whatever reason Lexar charges for the software that they bundle on the 80x cards; it's not available as a download on their web site,

so if the original poster wants it, he has to grab it off of the card.

 

-matt

Especially in 16 bit mode, Adobe RGB is a perfectly good color space to work in - it's plenty wide, and 16 bit color allows you to shift colors all over the place without creating big gaps in your historgram.

 

But you need to CONVERT the image to the lab's color space (sRGB or

the profile they give you -- and make sure to tell them which you converted to when you send them the file if they give you a choice of either) before you send it to them. Otherwise, the you'll get the strange level shifts you saw when you proofed.

 

-matt

I'm curious about these horror stories you've heard about the D2h. As far as I can tell (about a year of moderate use), it's pretty much what it claims to be, and I've not heard inordinate reports of surprising problems with it. I'm very happy with mine, for what its worth.

 

Anyway, I've not shot any 35mm since I got my first DSLR (a D1x and a D100, now a D2h and D2x), although I do shoot a great deal of 6x8 and 6x9 film. You won't look back with the D70. Other than a few accessories, the D70 is, as I understand it, an improvement over the D100 in most respects.

 

Your 85mm 1.4 becomes a terrfic (and tight) portrait lens. You might consider something on the wide side; the f/4 12-24DX zoom is not bad at all (and outperforms the Nikon primes in that range in some respects), and the 10.5mm fisheye can be a lot of fun.