mab

For the record, you may find <a href="http://www.crypto.com/chargers/">http://www.crypto.com/chargers/</a>

helpful, if a bit dated and incomplete.

<p>

-matt

Shun,

 

It looks like your results are consistent with mine. I saw a larger difference (at 1:1 pixels) in contrast and sharpness, but that's easily explained by the difference in resolution between the D100 and D2x.

Thanks, everyone.

 

Kevin: what did the repair cost?

 

-matt

Shun,

 

My point was that comparing the optical quality of these long lenses is without being influenced by these factors is harder than it may seem, and especially if you're interested in what happens other than wide open under bright light. Even 1/250 of a second was slow enough to reveal vibrations (on the full resolution of the D2x, with MLU and a relatively heavy tripod) at 300mm with the stock collar.

 

If you don't have a better collar, I'd suggest you light your target with a remotely controlled flash when doing your tests, if you want to be sure you're isolating the optical quality of the lenses themselves.

It's worth pointing out that both the 80-200mm AF-S and the 300mm AF-S

lenses are shipped with especially crappy tripod mounting collars that make it very difficult to eliminate the effects of vibration (from either the camera or the environment). My very first experiments with these lenses (with their original collars) suggested that they get softer as they are stopped down -- the reason, of course, was that the slower shutter speeds used at smaller aperatures amplified whatever vibrations were present in the environment (even with mirror lockup, and even at shutter speeds of 1/500-1/125). I was only able to get good performance out of these lenses after I switched to the Kirk collars (RRS also makes them).

 

I believe the VR lens comes with a better collar (although I've never used the lens myself, so I don't know for sure), and, of course, if you turn the VR feature on that will help dampen vibrations as well.

That could explain better peformance of the 70-200 + TC14E vs. the 300mm, even if the 300mm may actually be sharper when properly stabilized.

I've never reverse-engineered Wein's circuit, so I have no idea how,

exactly, that particular unit is designed.

 

A circuit to do the isolation is relatively simple, but not as

simple as adding a resistor. The correct design of such a circuit is

a bit too involved to describe here, but any introduction to

electronics will get you started. (And, frankly, if you think just

adding a resistor will do it, you NEED a closer introduction to electronics than you have before you can safely design such a circuit). Horowitz and Hill is the standard college text, although it has a lot more detail than you need.

 

That said, even if you design such a circuit, a lot of the practical cost of making it work is in the connectors and housings. I doubt, for example, that you'd be able to get a male and femail hotshoe and/or PC connector and a case for your circuit, in one-off quantities, for much less than Wein charges for their entire completed device.

Brian and Harvey, thanks; I found a few threads mentioning similar problems on Nikonias and dpreview.

 

Anyone know the magic words to get just-out-of-warrantee service for this problem.

 

Thanks

One year and two days after I bought it (and, therefore, two days

after its warrantee expired), my D2h appears to have developed a

problem with its meter. It always indicates an exposure of

1/8 at f/2.8 at ISO 200 (or equivalent exposure at other shutter

speed, aperature, or ISO), and the meter mode is always matrix

no matter where the switch on the prism is set. This remains true

no matter what the ambient light, no matter what lens is attached,

and no matter what other controls are set. Everything else about

the camera appears to function properly.

 

A quick search of the archives does not suggest that this an

especially common problem, although perhaps I didn't search in the

right places. I've tried obvious fixes like removing the battery

(and the button

cell) for a while to let the CPU reset, etc, to no avail.

 

Has anyone experienced a similar problem? What was involved in

getting it fixed? Has anyone had a meter problem repaired out of

warrantee by Nikon USA service? What was the cost? Any chance this

is a known defect that Nikon will fix even with a slightly expired

warrantee?

 

Thanks

Unfortunately, "digital" is not a standardized term when used to describe flash functionality.

 

Obviously, digital photography is, at its root, still just a kind of photography and digital-compatible light is, at its root, still just light. There's no FUNDAMENTAL special requirement associated with flash for digital, although some digital cameras and flash systems do have special quirks.

 

I've seen the term "digital" used in discussing flash no less than six different ways:

 

1) To refer to a digital setting and/or numeric readout of, e.g., power output. My Photogenic monolights are "digital" because they have a remote control and an display of the power setting.

 

2) To refer to a high degree of output repeatability, e.g., for easy use with multi-pass digital capture systems or for color management,

 

3) To refer to a compatibility with a dedicated camera exposure control system (e.g., Nikon iTTL); this usage almost always applies only to small hotshoe flashes, not AC studio units.

 

4) To refer to a slave system that won't be triggered by the pre-flashes used my many dedicated digital camera flash systems (such as Nikon iTTL). For example, Wein has a line of "digital" optical slaves that are not supposed to trigger until the main flash goes off.

 

5) To refer to a low trigger voltage for compatability with sensitive and perhaps inadequately protected digital (and other electronically controlled) cameras.

 

6) As a meaningless comfort phrase to reassure the clueless consumer that the light produced can be recordered by digital sensors and not just on film.

 

Which way is the term being used? You usually can only tell from context, but sometimes not even that way.

I own a 1227 and a 1348 (similar to the 1325). I hardly ever use the 1227. Neither are suitable for "ultralight backpacking," and once I've committed to schleping a tripod, the 1325 isn't really that much worse. I appreciate the extra stability (I shoot with a DSLR with up to a 300mm lens, a Fuji 680, and a Sinar P). With tripods bigger and heaftier is almost always better, right up to the point that it's so big and heavy that you don't take it with you in the first place.

 

But you didn't say what kind of photography you're doing. Backpack camping in the wilderness has different constraints than, say, architectural photography with assistants.

I also posted in the other, now deleted thread.

<p>

I think the key is restraint; it's very tempting to create images that have such wide dynamic range that, when rendered to include all the

shadow and highlight detail, are flat and lifeless or that just look freakishly wrong.

<p>

I've tried the feature exactly once so far, on a cliche sunset near Big Sur, CA. I'm no nature landscape photographer, and this sort of photo

really isn't my thing, but sunsets seem like an obvious choice for playing with this feature. Anyway, I measured 12 stops, and tried to

get all of them (using a D2x, an 85mm/1.4 lens, and a tripod).

<p>

After much fumbling around with curves at the 32 bit conversion stage

and in the 16 bit photo, the result I like best so far (and that

captures detail across 12 stops is at <a href="http://www.crypto.com/private/hdr-sunset-cliche.jpg">http://www.crypto.com/private/hdr-sunset-cliche.jpg [link]</a>.

<p>

One thing I noticed was that it is very hard to avoid moving the camera when changing the settings; auto bracketing is really a better way to do this.

<p>

Anyway, this image, for all its cliched dullness and uninteresting composition, is considerably more detailed in the shadows and highlights than I was able to get with any of the single captures that I merged into it, and probably better than I could get with an ND-grad. But I've still got a long learning curve to climb with this feature before I can tell whether I'll be using it routinely.

Wow, that's another one for the "I wish <em>I</em> had thought of that!" category.

<p>

Thanks for posting it.

<blockquote><em>

Matt, I had a similar thought when I first started to play around with some D2X NEF files that had been uploaded to various websites. Unlike some previous cameras (which I think use an internal number apparently unrelated to the one on the outside of the camera), the D2X serial number EXIF tag seems to contain a number in the known D2X serial number range. Can someone with a D2X and a copy of (e.g.) exiftool confirm that the 'Serial number' tag contains the number that's actually stamped on the camera? If so, D2X users might want to think carefully about the implications of broadcasting their serial number to the world before uploading files to public sites...

</em></blockquote>

 

Yes, it's the actual serial number (same as that engraved on the back of the camera. (I'm not sure how much worse that is than any other persistent and un-changable identifier would be, though).

It just occured to me that there may be privacy implications in the way Nikon implemented the WB "encryption" scheme (in quotes because it stretches the usual

technical definition and application of encryption).

 

Because the scheme is keyed to the camera serial number and shot count, it becomes non-trivial to strip these fields out of raw files before sending them to others (e.g., with exiftool). (It may be possible to do so by decoding the white balance and re-encoding it under a fixed "fake" serial number and shot count, but I'm not sure enough is understood about the scheme well enough to actually implement this properly).

 

This is clearly not an issue for most people -- most photographers don't deliver raw files to clients and most of those who do probably don't care about exposing these fields -- but it could be a significant issue for some users and applications. For example, the shot count reveals the relative order that images were taken, which make be commercially or otherwise sensitive, and the serial number presumably would closely correlate images to particular photographers.

 

While this may not be an issue for everyone, I can certainly imagine circumstances where it could be a deal-breaker.

 

Ick.

Brian,

<p>

Sure, I know Ed; he arrived at Princeton just after I left there, and we cross paths regularly. In fact, you can read my declaration in Felten vs. RIAA (one of the first DMCA cases, although it never went to a full trial) <a href="http://www.crypto.com/papers/mab-feltendecl.txt">here (at http://www.crypto.com/papers/mab-feltendecl.txt)</a>.

<p>

Best,

<p>

-matt

Bernard: have you tried ACR3.1? I've found that gets the "as recorded" color temperature correct just about every time (except with auto, where it does its own calculation). It doesn't report what the name of the setting was, but it gets the temperature, etc. correct.

 

That said, there are two issues here. First, there's the practical one of how to deal (today) with D2x raw files. Fortunately, this doesn't seem to be nearly as dire as it could be, and even Adobe's software seems to manage reasonably well under the circumstances (although not as well as it should be able to, had Nikon opened the NEF spec properly).

 

The second -- more abstract but equally important -- issue is how the photographic community should respond to Nikon's use of a proprietary format for the output of their cameras, with the implied threat of a lawsuit (fueled by Adobe's apparently eager captiulation) against those who build tools to process the data in "unapproved" ways. I don't think this is a road we want to permit the vendors of our cameras to follow.

 

-matt

In practice, ACR 3.1 (which works with PSCS2) guesses the "as shot" white balance. I've found it has always been correct if I select one of the preset settings. If you use auto WB, ACR does its own auto WB, based on the image data (but not, presumably, based on the camera's ambient light sensor). It works well as a starting point, but it's not as good as Nikon Capture in my experience.

 

So it's not as bad as it could be, although who knows what the next firmware will bring.

 

No idea whether there's a special Canadian version. My D2x (and software) is from the US market.

Disclaimer: I am not a lawyer, however I do lead a cryptology and infosec research group whose work often approaches the edges of the US DMCA, and I've testified as an expert witness in several DMCA cases as well as on the Hill on this issue.

 

Many provisions of the (US) DMCA constitute bad law with a corrosive effect on inovation, to be sure. There are many reasons for anyone who creates or works with electronic intellectual property to dislike and oppose these provisions and to fear their reach into what would otherwise be intuitively legal and entirely ethical activities. No argument from me about that, and I'm on the record all over the place saying so. I'm also a Nikon digital (and Adobe software) user, and I think the use of a propritary, closed data format for RAW captures stinks. There are many reasons for users to decry this hostile and wrong-headed scheme, and I sincerly hope the market punishes Nikon for it quickly enough that they adopt a new, open format sooner rather than later.

 

However, I'm not sure what provisions (civil or criminal) of the DMCA would be violated by "reverse engineering" the existing "encrypted" white balance data. I can imagine contract claims arising from violating some anti-reverse-engineering clause of the Nikon software license, but that's got nothing to do with the DMCA (and wouldn't apply to reverse-engineering that wasn't based on examining the software). In fact, I don't believe Nikon ever actually raised the DMCA here. As far as I know the DMCA was first invoked here by Adobe, in explaining why the were reluctant to incorporate the obfuscated WB fields into their software.

 

To say the least, there is reason to be skeptical of Adobe's interpretation of the DMCA here. Adobe produces digital rights management software that, in part, relies on the DMCA for its "security." They have been the plaintiff and/or criminal complainant in several DMCA cases in an agressive effort to enforce this. A broad interpretation of the DMCA is very much in Adobe's interests.

 

I'm not saying the DMCA doesn't apply here, but only that it is not obvious that it would. And those who are most loudly advancing the position that it does have a vested interest in doing so.

 

-matt

Yes, of couse. But while it may be true that the 50/1.4 lens doesn't perform well wide open (I don't have one anymore so I can't test), the 28/1.4 and 85/1.4 are quite remarkable in this regard. So poor wide performance is not a fundamental property of fast lenses.

 

Focusing is, of course, very, very critical at f/1.4 - much more so than most people think.

Well, there's always the 10.5mm DX fisheye... (A fun lens, with limited applications).

 

I don't understand the comment that fast lenses wouldn't work well on a DX sensor. I get excellent results on my D2x wide open with the (full frame) 28/1.4 (now that Nikon fixed it) and the (also full frame) 85/1.4.

 

What I want is a wide angle DX shift lens -- anything wider (and sharper) than the full-frame 28mm PC lens they offer now. Please?

I meant to write "I'll try the calculations control" in the last

sentence above. Too early in the day, I guess.

I note that the few negatives written about in this review were pretty much the same things I found annoying when I first got my D2x, particularly the mis-designed burst mode (and which I posted <a href="http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=00BmSR">to this thread here</a>). My comments, for whatever reason, provoked some hostility at the time here, I guess for "complaining" about a rather expensive product. But I stand by my observations.

<p>

After two months of use I remain mostly delighted with this camera.

Another option you might want to play with, especially if you find positioning a 7 inch grid to cumbersome, is one of those fresnel lenses used for "long distance" flash (usually in nature photography). They reduce the angle of coverage significantly when used close up, although not as sharply as a snoot or grid. I believe Kirk Enterprises sells one.

Thanks for all the responses. Again, I ordinarily do an adjustment layer using either the color mixer, or, when I want a quick-and-dirty conversion, the desaturate command or the saturation control, and frankly, simple desaturation normally looks pretty close to what I had wanted in the first place. This is the first time desaturate failed me so dramatically while conversion to gray did what I wanted, and it's aways disconcerting when things stop working as you expect them to.

 

I'll try the channel mixer technique, too.

 

Best

 

-matt

Denis and Byron,

 

Thanks for the comments; I ordinarily use the channel mixer if I'm trying to actually create a serious B&W image as the finished product, but I'm still curious as to what's going on here. I find the desaturate and grayscale conversion functions to be convenient for getting a quick preview of what B&W might look like, and sometimes the results are quite satisfactory as a starting point (as, I think, the grayscale image was here).

 

So what's going on?

 

Best

 

-matt