david_r._edan

<p>Answer to Mike and anyone else who's interested.<br>

First, people must understand that I simply described my method. It does not appear in a "How-to, step-by-step manual" format. Therefore it should not be treated as such. It actually requires a fair amount of technical proficiency to execute the procedure with an adequate degree of precision. There are many variables that could go wrong. Just one would be enough to render the images unusable. Nevertheless, the technique can be modified in a number of ways that would produce equally reliable results. To be able to do that, one must truly understand what's going on at any given stage. Regardless of your own procedure, the test results can be of use to you. However, do not publish your findings, claiming that they rely on scientifically obtained data, unless of course, you are 100% sure that you know what you're doing.<br>

Now, Mike, I know what you're trying to do. A couple of things... First of all: a rear-curtain sync? There are several ways to do it but I think that you just want to use your speedlight on-camera. Your camera's CPU and the speedlight need to be communicating in a proper manner to be able to do that, which is one of the two: through your camera's hotshoe, or a TTL-cord. (I am quite aware that there's equipment out there that can do this via a different approach). Using the cord in an off-camera setup should work fine. I know I don't have to say that the speedlight must be in the manual mode. A speedlight on the camera could work too but glare would certainly render areas of the image unreadable. I only tried this once (a very long time ago). I printed out a test chart on matte paper. Can't remember the brand but it was made by Epson. It wasn't 'photo' paper, just very dense, very smooth, high-res paper with no kind of coating whatsoever. The d-max wasn't great but I figured it would be good enough. In the test I was actually testing the chart itself. When I got the slides back from the lab, I didn't even bother scanning them. After checking them out on the light table I decided to abandon the 'on-camera' approach for good. The paper reflected the light back in a way that bugged the hell out of me... and that was the least reflective type of paper I could use for the tests. Your results may differ.<br>

Now the rear-curtain approach. Ever heard of the 'afterglow'? Any light source (including strobes) will still be emitting light for a period of time *after* you have cut the power. This is especially true when the strobe is fired at maximum or near its maximum capacity. Relatively to the duration the capacitors actually discharge, the length of the afterglow can be very significant. So 'picture' this. Your speedlight had just fired. Your camera's CPU waits a predetermined amount of time and then begins to close the curtain. However, there's still enough light to go around. So the shutter is moving, shaking the camera body while that shake is being registered by the sensor. You could argue that it's insignificant but I must remind you that *my* method is about removing the shake out of the equation *completely*. Not some of it, not most of it but *all* of it. I would never actually use this approach because IMO the shutter travel during the afterglow is enough to offset your data. We're talking D800/E's fitted with 300mm primes @ f-11: enough resolving power and magnification to register you coughing in your hand in the other end of the room... Just kidding...<br>

Anyway... you should wait a sufficient amount of time before closing the shutter. That would probably be just a few seconds.</p>

<p>This discussion hasn't been about Moire for quite a few posts. I believe it is me who was first to point out that vibration is a very important factor in the long equation which dictates the sharpness of any given image. People are talking about tripods and testing lenses / sensors, which is all relevant. Here's how I used to test my 'imaging systems'. You will see that my method takes all kinds of vibration *completely* out of the equation. Properly running these tests will tell you *exactly* just how sharp your lens / sensor is at any given setting. The next time you'll be looking at a not-so-sharp picture, instead of guessing, you'll be able to tell who was at fault.<br /> You will need a room / studio which can be made completely or 'almost completely' dark. You will also need at least one good strobe. Two is better. Depending on the apertures or and ISO's you'll be dialing in, you might be able to pull this off with a single dedicated speedlight (it would have to be off-camera).<br /> Put your test chart/s in place, measure the distance, place the camera on the tripod - where it needs to be, make sure it's parallel to the chart. Take your flash readings (or estimate) and adjust. Dial in the aperture you're testing, set the camera on bulb, focus. Turn off the lights. Release the shutter. Wait a few seconds. Manually fire your strobes. 'Close the shutter'.<br /> *Use a flashlight or your phone to help you navigate in the darkness.<br /> *The waiting time from the moment you release the shutter until you fire the strobe/s should be at least 4-5 seconds. I'd suggest waiting at least 10 with lenses of 300mm and longer. <br /> *If your camera doesn't have a 'bulb' setting or you do not have a cable release, you can estimate how long you will need to hold the shutter open for and set the camera to that exposure time. If you set it to let's say 30 seconds but were done in 15, you can always finish exposing prior to the pre-set time. Put the lens cap in place, or switch the camera off.<br /> *Depending on the material your charts are made of, glare can be a factor. Ideally, I'd suggest using modifiers such as softboxes fitted on two monolights, placed opposite of each other, on either side of the chart. Since you're photographing something 2-dimentional, the angle doesn't really matter, so long it doesn't produce unwanted glare. You should also be aware how much direct light from the light sources reaches your lens and deal with that if necessary. You can block it off with flags. I use custom-made lens hoods for my studio work to deal with this kind of issues. A kicker, for example, can ruin the shot if you let it hit the front element.<br /> *One last thing which may not be very obvious. You don't need to sync the camera with the strobes in any way. If you're using two of them, you can actually fire each of them separately. Be careful not to move or shake anything as you're walking from one to the other in the darkness.<br /> PS: Now that I think of it, this can also be done with continuous sources. You would have to manually control the burn-time and, obviously, you can't be fast and accurate enough with high ISO's at wide apertures. You would have to deal with other issues as well. I just wouldn't go there. Nevertheless, the produced test results could be very useful, at least to you.</p>

<p>If that's a 1:1 crop, I don't think it's very sharp, especially given that the crop is taken from the central area of the frame. Certainly, there's nothing amazing about the quality of this photo. If the lens was stopped down to f/8 or f/11, I'd say that camera shake must have made its contribution as opposed to poor optical performance at a wider aperture. A VR lens on a monopod? Aren't you supposed to switch off the VR when the camera isn't hand-held? I do not want to get into an argument regarding the equipment and/or the circumstances as I do not need to be convinced that the D800 is a marvelous piece of technology. I think (and others will agree) that the photo is a poor demonstrator of the camera's true capabilities.</p>

<p>Just contributing something here... I believe that 3 major factors are at play here. 1 - Shutter speed, 2 - The resolving power of the sensor. The third *major* contributor is the 'magnification' which dictates how fast / how much the projected image is moving on the image plane.<br>

Those ultra dense sensors certainly change the rules of the game. One could always downscale a 36mp image to let's say 16, to 'hide the blur' but then why would you need a 36mp camera in the first place?</p>

<p>Just throwing something out there... The enormous resolving power of modern 30+MP and razor sharp prime lenses considerably increases the chances of registering camera shake at pixel level. There - your D800E's miniscule advantage over the D800 just flew out the window, at least under certain circumstances. I'm a 95% 'tripod shooter' and most of my work is actually produced in my studio (strobe lighting) but listen to this...<br>

The paradigm is shifting and camera shake becomes more and more of an issue. This post isn't about it, however, we live in a real world, photographing real subjects, not just all kinds of charts in a lab. Camera shake is present in *every* photo you take, regardless of how stable you think your setup is, so it is definitely a factor in real-life scenarios. I happened to take 'blurry' photos in a studio with no ambient light being a factor whatsoever. Given a handheld camera, a long telephoto lens and a slow-burning strobe it can happen to you too. No need to start arguing, just illustrating a point here: statistically speaking, with the resolving power of the D800 or the D800E you are more likely to end up with photos that exhibit motion blur at pixel level than with a lower resolution sensor of the same dimensions. Of course in the latter, *all* your photos would not be as sharp as the D800/E's.</p>

<p>Visually identifying Moire patterns through an optical TTL viewfinder? And those being registered by the sensor in a similar manner? The idea itself seems to be bizarre, especially when looking at the projected image with real human eyes. Maybe you're onto something here, Richard. Can someone else comment on this? If I can train my eyes to spot those bastards before taking the picture I definitely need to know how!<br>

PS: Nice chart.. Takes me back. Kinda miss testing my lenses on Provia with my old F5. No friggin' Moire back in those days...</p>

<p>I read all the articles people have posted here. We can end this discussion. I'm getting the D800. It appears that the sharpness increase in the images produced by the D800E is trivial at best, totally not worth having to deal with the extra chance of Moire. I was also under the impression that the D800E was in some way suited better for my workflow, being 'optimized for RAW'. That turned out to be completely false.<br>

Now I don't even need to start asking people around about dealing with Moire in LR. I'm buying the Nikon D800 because it's a good body and there's no Moire-free digital camera. And as for removing the patterns in Lightroom - I'll cross that bridge when I come to it.<br>

Thank you all again. It was fun.<br>

PS: Hey Peter, ever seen the flick "<em>Frankenstein</em>"? That's what those Kodak DCS's were...</p>

<p>Thanks to all for the input!<br /> I really need to read those articles. Kinda busy right now, will do it tonight.<br /> The thing about Moire is that it catches you by surprise. You may try to deliberately 'create it' in hundreds of photos with no success whatsoever but the next day you may be running into it in every second photo, when you least need it. It is not really a question of whether it does or does not happen, or how often. It is more about what to do about it when it *does* take you by surprise.<br /> The responses here do sound reassuring and it is obvious that the D800 is nowhere near of being Moire-free. I guess I needed to hear that it wasn't a real issue with the D800E. I've had some bad experience with Moire with other bodies, so I suppose I just wanted to make sure I wasn't making a big mistake of buying a 'headache' of a camera.<br /> Another thing is that I'm pretty much a perfectionist. I cannot simply ignore any kind of artifacts in my photos only because they appear in 'insignificant areas'. I really need to hear from someone who had to deal with Moire in Lightroom 4.1 (photos from Nikon D800E). It looks like I'll have to pose this question in the Digital Darkroom forum.<br /> To most of you (or at least some of you) I may appear like a paranoid nutcase. In my defense I'll state that Moire patterns have ruined a considerable number of my photos and even caused me to lose money. Perhaps the most memorable occurrence was when I was hired to produce several corporate headshots of one dark-skinned fella. Wrinkles on his forehead produced some really nasty blue and purple patterns which when printed out, looked like bruises. I was on a very tight schedule and I didn't notice those until I made the delivery the next day. The man pointed out to me that he "did not remember having been beaten up recently". Obviously he refused to pay and naturally, I did not insist. Never heard from him again. There have been plenty of other cases, albeit less dramatic.</p>

<p>Hello.<br>

Trying to decide between the D800 and D800E.<br>

I shoot only in RAW. I process my images in LR and I put a lot of work into every one of them. Naturally, if there are any visible moire patterns - I have to take care of them. LR4 offers a remedy via the Adjustment Brush or the Graduated Filter but how good is it?<br>

My main concern, however, is the frequency with which I would encounter this kind of problem with the D800E. Almost every 'scene' has the potential of yielding images with nasty moire patterns. There's no way of foreseeing or avoiding it, well, not a practical one anyway. I'd like to get some input from people who actually own this body and use it on a regular basis.<br>

How often do you get moire patterns? How affective is LR4 in removing them? Let's say that a very important area of the image has moire all over it, can the tools available in LR make it disappear completely? What about the artifacts? Again, I'm only asking the people who shoot with a D800E and use LR4.1<br>

Many thanks!</p>