Neutral Density

<p>Hello, I have been taking an interest in long exposures, and only recently it got to a temperature and time of year where staying out and still that long is practical, but I decided: I first became interested in long exposures for the purpose of sunset, but the problem is that the light is too bright. What ND filter would I want so that I could make a picture that I can currently hand hold with a 20/2.8 or a 50/1.7 into a picture that will take somewhere from 15 seconds to two minutes to expose? Yes, I realize that the sun will have a motion blur, but<br>

A: I want to overexpose those shots so that the brightest spot will be washed out almost to white, but still with color, but don't worry yourself with the exposure stuff involved there. I want to find out for myself, but I have to use my digital camera because nobody has responded to "Diagnostic?" in classic manual cameras. Could one of you? I do have more than a glasses repair kit now.</p>

<p>B: The sunset is far from my only intent.</p>

<p>Thank you. Especially if you answer "Diagnostic?" You will probably have to search the forum, because it's been so long.</p>

<p>The sun isn't the only thing that will have blur. You can't hand-hold that long without blurring everything just from the hand-holding.</p>

<p>As for which ND or whether to use an ND, you'll need to determine the exposure data for the effect you want and then work back from there to an exposure time you think you can hold. If the correctly exposed combination is f16 at 100ISO for 1/60th and you want to blow out the sun to what it would look like at f/8, then without blur you're looking at 2 stops density at a 60th.</p>

<p>So if you want 15 seconds, you'll need one helluva filter, nearly black!<br /><br />But, since you're going to get lots of blur with just a 20th of a second anyway, you might as well not sweat the long hand held exposure and simplify your calculations. Put the rig on a tripod and let the natural action of the situation create the blur at a 20th or 10th with a 2x or 3x filter. </p>

<p>Finally, read the warnings with your digital sensor about exposing it too long into direct sunlight. I know my 7D has them for video, so the warning may also apply to still.</p><div></div>

<p>Arg I hate typos IS) 100, overexposed 2/3rd stop</p>

<p>Arg I hate typos IS) 100, overexposed 2/3rd stop</p>

<p>For most other things, the goal is blur. The sun is the only one that could pose a problem.</p>

<p>Blur with 1/20 second? I seldom get blur at 1/4 second. How can I get blur at 1/20? I did a night exposure of the ocean. It's pretty bad. Severely underexposed to say the least, but it shows what I want to achieve, generally in better light (i.e., so my eyes don't need a "focus assist lamp")</p><div></div>

<p>Well, do your blur studies - we all do somewhere along the way for the type of ecosystem we specialize in. Draw up a little chart, tuck it in your camera bag and use it for a while until you have a sense of how the factors interact. I'm not certain you can purchase a ND that's more than 4 stops, add your polarizer and you might get 5. After that, you might have to haunt the theatrical light gel store if you still can't get the effect you're after.</p>

<p>Lotta noise there. Try dropping your ISO to as low as the camera will go and see what kind of speed you need to get blur. And get the camera on a tripod. My sample shot was ISO 100 and has no noise in the original.</p>

<p>I could stack filters... I just realized what I typed. I could use "Startrails" and stack exposures to give the same effect in some cases.<br>

PS what can I do to get blur at 1/20? What I have done in the past is gone really far away, gotten my 500/8, setting it to f/32, loaded ISO 50 Velvia, over exposed a full stop, underprocessed half a stop, and corrected the rest in post processing or underprocessed it to a greater extent. that produced no blur at 1/15 with a really turbulent sea.</p>

<p>I could stack filters... I just realized what I typed. I could use "Startrails" and stack exposures to give the same effect in some cases.<br>

PS what can I do to get blur at 1/20? What I have done in the past is gone really far away, gotten my 500/8, setting it to f/32, loaded ISO 50 Velvia, over exposed a full stop, underprocessed half a stop, and corrected the rest in post processing or underprocessed it to a greater extent. that produced no blur at 1/15 with a really turbulent sea.</p>

<p>Well, check out this link:</p>

<p>http://www.luminous-landscape.com/essays/impressions.shtml</p>

<p>I'm assuming you're shooting from a tripod if you are looking for 15 second to 2 minute exposures.</p>

<p>You're going to likely need in excess of 10 stops of ND filtration, best accomplished with a Cokin or Lee rectangular filter holder. Both Lee and Hitech make a 10 stop filter as well as 4 stop and under filters. Here's a shot I did with the Lee 10 stop filter, taken right as the sun set. It's a 10 minute exposure at ISO 400 and f/11. I had to go to ISO 400, otherwise the metered exposure would have been 40 minutes, but with fading light into evening it would have taken all night. The color/tone are just something different I did in post processing. The rest is motion blur from the heavy wave action.</p>

<p><img src="http://farm5.static.flickr.com/4141/4749470158_f68edff85c_o.jpg" alt="" width="670" height="1000" /></p>

<p>Interesting photos, Sheldon. What did it look like in color?</p>

<p>What you are asking falls in the sphere of filter factors.<br>

<br />Most filters come with a slip of paper announcing their filter factor (FF). Say you meter a scene and the exposure is f/11 @ 1/60 second. You mount a filter with an FF 4. To solve for the revised shutter speed we multiply. Thus 1/60 x 4 = 4/60 = 1/15 second. Maybe you do not remember how to handle multiplication or division by fractions, that is OK. The point is a FF is a multiplier.</p>

<p>You ask, what ND to mount to bump up the exposure time to 15 seconds. OK, suppose you meter the scene and find the exposure time to be 1/60 second without filter. What value ND must be mounted to lengthen the exposure to 15 seconds?</p>

<p>We divide 15 seconds by 1/60 seconds to derive the FF. Best to convert the 1/60 to a decimal = 0.0167.</p>

<p>Now the math is 15 ÷ 0.0167 = 900. In other words; 15 seconds is 900 times longer than 1/60 second. We need to mount an ND with a filter factor of 900.<br>

The problem is; ND fitters are labeled in density units and not labeled in FF. Density units are the traditional notation used because all this stuff came about before calculators and computers. Photo scientist used this notation to simplify their life. This traditional (logarithms) method allowed multiplication and addition to be performed by substituting additional and division. Do not worry your head; all you need is a table. Below is density to FF.<br>

0.05 = 1.12<br>

0.10 = 1.25<br>

0.20 = 1.6<br>

0.30 = 2<br>

0.60 = 4<br>

0.90 = 8<br>

1.20 = 16<br>

1.50 = 32<br>

1.80 = 64<br>

2.10 = 128<br>

2.40 = 256<br>

2.70 = 512<br>

3.00 = 1024<br>

If you mount a 3.00 ND, the multiplier is 1024. So 1/60 second X 1024 = 0.0167 x 1024 = 17 seconds.</p>

<p>One caveat: When we lengthen exposure time severely we enter into the realm of reciprocity failure. The light sensitive goodies (silver salts) operate under a kind of chemical inertia, somewhat like a desire to stay unexposed. When the exposure is super elongated, we must apply additional light energy to overcome what is called reciprocity failure. Better add 15% more time thus try a 3.00 ND filter using an exposure time of 17 x 1.15 = 20 seconds.</p>

<p>All this stuff falls under the heading of gobbledygook, nobody said its easy. </p>

 

<p>Thanks Emily, the color was just a little warmer/more magenta.The light got cooler after the sun went down, so I just went with it in post processing.</p>

<p>Alan brings up a good point about reciprocity failure and shooting film. I remember from my days of 4x5 film shooting that Tri-X had a terrible problem with reciprocity failure. The curve was something on the order of a power of 1.4. So, if you had a metered exposure of 15 seconds, the actual exposure was 45 seconds (15^1.4). If you had a metered exposure of 2 minutes it would be almost 15 minutes, and 10 minutes would be over 2 hours. I ended up switching to Acros film for specifically that reason, it had almost no reciprocity failure.</p>

<p>Here's another 10 stop shot, on Acros 4x5 film (ISO 100). Middle of the day, cloudy conditions, 2 min at f/22.</p>

<p><img src="http://farm3.static.flickr.com/2150/2067276037_3ae5133aef_o.jpg" alt="" width="1000" height="793" /></p>

 

<p>I couldn't have used a lower ISO. That was shot with ISO 800 color print film in a Vivitar v3800N. I only pack what I'll need, lest I use 50 rolls inside of a day. For obvious reasons, I did not know if I had the correct exposure. I metered as ISO 3200 at f/1.7 and then did a calculation to determine what the appropriate speed was 3 stops lower (ISO) and then 7 f stops lower, and exposed for that time by counting seconds. It was on a tripod, but the film was long expired (, by about 8 years.).</p>