Approximating Line Art

This is about approximating line art in Photoshop.

Now credit where credit is due: I didn't invent either one of these methods, I found one of them in a column by Larry Becker in the Dec-11-Jan-12 issue of Photoshop User Magazine (which you get if you are a NAPP member — click here) and the other one in a tutorial that I found via the forums on the NAPP site (again, — click here!). The tutorial is one of several by Dee Dee Martin (Swampy) who is also a NAPP denizen. Please visit her tutorial site here. Her video tutorials are excellent.

Thank you Larry and thank you Swampy for two great tips on using Photoshop.

Some background: I wanted to do a different treatment for some images I shot the other day. I've been thinking about line art and I remembered reading something about emulating line art in Photoshop (it was the Becker article but I couldn't remember where I saw it) so I posted the question on the forum and I quickly got the two links above. Both methods work, both are brilliant and both are entirely obscure! No way I would have found them on my own! I still don't understand why they are where they are in Photoshop but why look a gift horse in the mouth? As usual there are 80 different ways to skin a cat in Photoshop!

So I'm posting this here for two reasons: (1) to share what I've learned and give my readers some ideas about some new things to explore, and (2) as a convenient place to put these tips where I can find them again! One of the biggest problems with Photoshop is how complex the program is, and how easy it is to forget how to do something if you don't do it regularly. These fall in that category, especially in view of how obscure they are.


So here goes:
This was a somewhat 'nothing' image that I took at the beginning of the party I was shooting, more for testing exposure than anything else. By the way, lighting came from my SB-600 flash mounted high on a light stand, controlled remotely from my D300 in Commander Mode. I had the Gary Fong diffuser on the flash, pointed towards the ceiling mostly (although the nature of the diffuser is that it also acts as a soft direct source). The pop-up flash was set for 1 stop under just for a little fill, and white balance was set to "flash" (some of you like details, I know!)

Nikon D300/Nikkor 70-200mm @ 155mm. 1/180 sec@ f/2.8, ISO 200.

First, the "Becker Method" from the magazine.

How do you do this?

■ Working on a duplicate of the background layer, select Filter→Blur→Smart Blur (note: you have to be in RGB 8-bit mode or it will be greyed out)

■ Change the Mode at the bottom of the popup to "Edge Only"

■ Play with Radius and Threshold until you get the right 'look' you're after

■ It will give you white lines on black. If you want, invert the image with Image→Adjustments→Invert (Ctrl/Cmd-I) to black on white

■ Change the opacity or the blend mode (or both) of this layer to let a little colour through

■ (and this is my change to Larry's method) apply a layer mask then paint on it to let more of the colour from the original through.

That's what I did in the above image. I like the charcoal-like texture of the lines and the subtle pastel colours.


Now the "Swampy Method"

It works like this:

■ Duplicate the background layer. Change the duplicate to a monochrome grey layer with Image→Adjustments→Desaturate (Shift-Ctrl/Cmd-U)

■ Duplicate the new layer and invert it to a negative with Image→Adjustments→Invert (Ctrl/Cmd-I)

■ Set the blend mode of this inverted layer to "Color Dodge"

■ Apply Filter→Other→Minimum and adjust the Radius slider to your liking

Here's what you have at this point:

■ Merge these two new layers together (Select the top one and "Merge Down" or Ctrl/Cmd-E)

You can add some colour at this point by reducing the opacity and letting the original layer leak through. You'll get something like this:

Alternatively, you can change the blend mode to "Luminosity" before merging the layers down

AND play with the opacity, yielding still a different result:

For what it's worth, every time I reduced the opacity, I found that a setting around 60% gave me the best result, but hey, it's up to you!

Two different obscure approaches, both of them work!

---

This PS is added in May 2012. I often find myself going back to this post to remember the keystroke sequence to do these line drawings. In the meantime, I found some other ones so I'm adding this note so I can find them all in the same place. I wish I could remember where I found this...

Create a new layer (Ctrl-J) and change the blend mode to "Divide". Now use a blur tool: you get different results with Gaussian Blur or Box Blur or other ones I haven't discovered yet.

---

Here's another method, added in October 2012: courtesy of Bonnie Glidden. This one is even easier: create a new layer, then select "Filter → Stylize → Find Edges". You're basically done! You can vary the opacity, add a black-and-white adjustment layer, etc. if you want.

---

— 30 —

Comparing HDR Programs

I posted this also on the NAPP forums.  I got interested in the different results you get from different programs. I've done these comparisons before, trying to decide which program I like better but I've concluded that it depends on the image.

That said, up to now Photomatix has been my first choice but these results may change that.

The goal here was to produce the image that I liked best in all three programs. I did not look at the other images while working in any program, with the exception that I knew I really wanted colourful skies when I got to the third one (P'matix). And I was not trying to MATCH the images.

Workflow:
was to select all 5 source images (bracketed burst, 1 stop apart, RAW. I open them in the HDR program, then look through the thumbnail presets for the one closest to what I want, then I modify it with sliders until I had done the best I could. I then saved the files and exported them to CS5 for further tweaking.

Once open in CS5 I created a new layer and ran a hi-pass filter. I adjusted the image with curves, then I ran DFine 2.0, sky preset which I masked back in. Next I ran Viveza 2 to increase the saturation in the sky (except the CS5 one where I had to REDUCE the saturation overall).

Back to Lightroom where I tweaked a little and synced the cropping and metadata. I then exported each one as a 720 px-wide jpg for posting here.
Here are the images:

1. Processed in Photoshop CS5 HDR Pro

2. Processed with Nik HDR Efex Pro

3. Processed with Photomatix HDR Pro 4

Now remember, I was not trying to MATCH the images, just to do the best I could with a pleasant image.

Comments:
1. On first look, CS5 HDR Pro produced a very pleasing image with lots of detail and saturation, with the least amount of effort and fiddling. But the saturation came in way too high and there's very visible chromatic aberration.

2. Nik HDR Efex Pro gave me a really clean image but there was almost no colour in the clouds until i ran Viveza. Again there was some evidence of chromatic aberration but I thought the image was much smoother overall. This program gave me much more range doing details than the other two. I should take out a little colour bias from this image but didn't notice until I was posting this.

3. I maxed out almost all the controls in Photomatix and still couldn't get the kind of detail and sharpness I liked and again, there was no colour in the clouds. A small hue/saturation adjustment in Photoshop gave me all the colour I wanted but really bad artifacting so I backed it out. Microsmoothing helped a lot with the noise in the sky. There was NO evidence of chromatic aberration even when I viewed the image at 200%. I ended up with lots of Halo around the trees, though.

Conclusion:
Until I actually looked at all 3 images together, I didn't know which one I preferred. I'm not going to tell you, though, because that's subjective (well this whole thing is subjective!) and you may have a different preference. CS5 was the easiest one to use. Nik had the broadest range and lots of room left to play. Photomatix was maxed out but gave the cleanest result.

I know this is a silly comparison, using just one image and subjective rather than objective adjustments and inconsistent treatment. So please file it under "For What it's Worth".

— 30 —

ALT-NUMPAD ASCII Character shortcuts

This isn't really about photography, but I find myself using these shortcuts all the time so I thought I'd share them…

I like to use properly formatted extended ASCII characters sometimes (technically they're "ANSI characters"). This post is about giving some of you some keyboard shortcuts.

Why only "some of you"? Well because I think they work on a Mac, but I'm not sure since I don't have one. I haven't found a way to use them on my iPad, and laptops without numeric keypads are a challenge as well.

I'm pretty sure this is not news to a lot of you, but there's a whole new generation out there who don't know what the extended ASCII characters are. If they want to insert a special character, well they're only familiar with the menu in the MS Office programs.

Since I have some typography background, I know, for instance, that an "ellipsis" is not just three periods in a row. I even know what an ellipsis is! If you don't, well Google is your friend! Do you know when to use a hyphen, or an en-dash, or an em-dash? I know I don't use them exactly correctly according to the book, but I find that an em-dash — surrounded by a space on either side — makes text read better than either of the other two, in my eyes anyway.

So without further ado, here's a table of some of the special ASCII characters that I often use in my workflow. Feel free to add to the list. You can look these up here, but it's rather a lengthy list — there are lots more characters that I don't use. My list only includes the ones I use regularly.

These characters are typed by holding down the alt key ('option' on a Mac?) and typing the 4-digit ASCII code (don't forget the extra zero) on your numeric keypad. The regular numbers at the top of your keyboard don't work. On a laptop, you need to turn on the numlock which, on my Lenovo, is done by holding down the fn key and hitting scroll-lock.

(Blogger doesn't want to let me insert a table. So forgive the lack of formatting here. I used "Courier" font to maintain the spacing)

…     alt-0133     Ellipsis
‘     alt-0145     Open single smart quote
’     alt-0146     Close single smart quote
“     alt-0147     Open double smart quote
”     alt-0148     Close double smart quote
•     alt-0149     Round Black Bullet
■     alt-254      Big Square Black Bullet (no extra zero)
–     alt-0150     en-dash
—     alt-0151     em-dash
™     alt-0153     Trademark symbol
¢     alt-0162     Cent symbol
©     alt-0169     Copyright Symbol
®     alt-0174     Registered Symbol
°     alt-0176     degree Symbol
±     alt-0177     plus/minus Symbol
²     alt-0178     superscript "2"
³     alt-0179     superscript "3"
¼     alt-0188     one quarter
½     alt-0189     one half
¾     alt-0190     three quarters

The web page I linked to also suggests using one or more of these characters in a password, which would make it much more difficult to crack. But I'll leave that for the geeks among you. Oh wait, I am one…

Here is a formatted .pdf of the above list for you to keep handy near your computer. Enjoy!

— 30 —

Two Superb Lenses

This week I'm babysitting my friend's lens. We're heading up to Lake Superior for the Gales of November workshop, I'm driving up with a couple of other photographers and Ron is flying up, so I offered to take his extra stuff in the car. He cautiously gave me his "baby", his Nikon 600mm f/4 lens and told me I could 'play' with it if I wanted to.

What he didn't give me was the Wimberley mount and tripod... so I was a bit limited in what I could do. That lens is heavy! And of course, with the 600mm super telephoto focal length, handholding it is a challenge. I've held a lot of rifles that weigh about the same or more (the lens is 10 lbs by itself, without the body attached) but with rifles you have a shoulder stock to help steady it to counteract the muzzle-heaviness. I have read that some people (Moose Peterson comes to mind) regularly handhold this lens. I guess I have to start lifting weights again if I were going to.

So the old "1/focal length = slowest shutter speed" rule goes by the board. I cranked the ISO up enough so I was shooting at 1/1000 sec or faster, which met with some success. Most of the time, I rested the lens on top of a soft suitcase.

Anyway, this is a superb piece of optics. It's not meant for shooting landscapes, it's for birds and for long distance reach-out-and-touch shots of dangerous game. So I looked for a dangerous animal to shoot and sure enough, I found one.

This wild beast had to be 1/4 mile away. This image has been post-processed, I did everything I could to sharpen and enhance it. Then I cropped it slightly just for positioning, so you're not really looking at what I saw exactly through the lens.

This next image hasn't been touched, though. Well not really: I cloned out the head of a calf on the right edge and ran Nik Sharpener (pre-sharpen RAW) but that's all. It's not even cropped. This is exactly what it looked like through the lens. I didn't even level the horizon.

My title says "TWO superb lenses". The other one is my old standby, Nikon 70-200 f/2.8 VR. It was like holding a helium balloon in my hands after the big guy! No weight at all! Anyway, I took a couple of comparison shots of the same scene. Here's one:

Shot at 200mm, 1/500 second at f/8, ISO 400. Nik Sharpener RAW pre-sharpening, clone out a couple of ugly houses in the background and straighten the horizon.

The same tree, shot with the 600mm, 1/1600 second at f/4 handheld, ISO 400. This one has not been sharpened: I ran presharpen on it but didn't see any discernable difference, so I'm showing the original here.

So the 600 shows outstanding sharpness and contrast. You know how we say, "you can't take a bad shot with the 70-200"? That's partly because of the quality and partly because of the beautiful isolation you get with the shallow depth of field and great bokeh. I want to say the same thing about the 600: look how the tree jumps out with the shallow DOF!

Here's a direct comparison:

What you have to understand is that the shot on the left is only 250 px wide after cropping. Incredible sharpness given the huge enlargement. The contrast on the 600mm shot is excellent. And look how it jumps out with the small DOF! As indicated, they're both shot at f/8.

So two wonderful lenses. It was fun comparing them.

Using LensAlign to fine tune your lenses

This is an evaluation of the LensAlign MkII product, designed to help you microalign or microadjust your lens/body alignment on higher end DSLR cameras.

Every manufactured product has tolerances. You might have the most precise milling machine in the world, but whatever you produce, if you're shooting for 1.00000 in, the resulting product will not be exactly that, it will be off by a small amount. So if you're a high end camera manufacturer, the fit of your lenses to your camera body might vary by a few thousands of an inch. That's enough to affect where the camera actually focuses when you press that shutter release and activate your autofocus.

And it can change. Using your camera, swapping lenses, firing that shutter 50,000 times, it's going to change. If you're at all fussy about it, you're going to want to tune the alignment of your lens and camera body to give you the best possible focus. That's what the LensAlign product is for.

I was given a LensAlign MkII package to evaluate. I made it clear that if I did so, my evaluation would be 100% unbiased, I would report on my experience exactly as it happened, good and bad. I have no interest in this company and/or product, and the device will be going back to them after this is complete. I was not paid in any way for this evaluation.

On a scale of 1-10, 10 being the best, I would rate the product like this:

(a) Not adjusting your lenses........... 3
(b) Adjusting your lenses using a cobbled together setup with a tape measure.......... 5
(c) Adjusting your lenses with the LensAlign product......... 7
(d) Sending your camera and lenses in to a professional to do the job............ 9

Now let me explain, then we'll get into the procedure.

Most people don't have to microadjust their lenses (for the sake of saving typing, I'll just call it 'aligning lenses' from now on. It's actually 'microadjusting the alignment of a specific lens with a specific camera body by using the software in a high end DSLR'). Now why do I say that? Well because most people aren't pushing the edge of the envelope. If you want to focus precisely, you do it manually, not with the autofocus. And the only time it really matters is if you're shooting with a shallow depth of field, and you've damped out all vibration that would affect sharpness, and... you get the picture.

But if you do want that expensive lens to give you the best results with that expensive camera body, well, you're going to want to take the trouble to align it. There are 3 ways to do it. Let me skip ahead to option (d) for a second. Sending it in to the factory or a service shop to do the alignment for you is the best way. When you get it back, somebody with fingers smaller than yours and eyes that work better than yours, and the skill to take apart the camera or lens if necessary, and a machine that probably cost more than your car to do the work, will have done a fine job. He'll probably have taken 2 weeks or more to do it and the bill that he hands you is not going to be trivial, but he'll do a great job.

The question is, what is your time worth? It took me about 90 minutes, from start to finish, to align two lenses with my D300 body. It took me a bit longer than that when I did it a couple of years ago (September 2009. Here's the writeup) and it cost me nothing. I put the camera on a tripod, stretched out a tape measure at an angle, and did pretty well the same thing that I did with the LensAlign product. The results were similar if less precise, but this isn't something a professional or fussy enthusiast would do. It works, though.

The LensAlign product is reasonably professional. It's made of thin plastic, though, and I could see that you might have some troubles with it after repeated use. Assembly is very straightforward, even for a numnutz like me (I don't fix things. If I look at them the wrong way, they break. My aging eyes don't work as well as they used to and let's not get into arthritis in my hands...). I wanted to do it right, so I used the instruction sheet and it still took me only 5 minutes. It's designed to be disassembled so you can transport or store it in an 8x10 envelope. Here's what it looks like when you're done:

Edit: This original shot was done at night in lousy lighting. I took the LensAlign outside today before putting it away, and re-shot it. You fit those 7 plastic pieces together and mount it on a tripod.

Here's the thing. You want that target thing on the left to be parallel with your shutter plane. That way, the scale on the right is the right distance — the SAME distance — when you focus on the target. And if your focus sensor is not exactly in the middle, it doesn't matter because everything is parallel and equidistant.

Here's where I had the biggest problem. Getting it parallel. If you blow up the picture, you'll see that there's a hole in that back panel, in the red dot in the centre of the circle. There's also a hole in the centre of the front target. If you align it so that you can see the back hole through the front hole, through the lens, it's parallel. Oh yeah? HOW? I figured out that if you look through the back hole and move the assembly so that you can see the lens of the camera (which is of course mounted on a second tripod), then it's lined up. But the lens is black... so I tried to fool it by taping a piece of masking tape with an "X" on it to a lens cap... I thought that worked but in the end, it was still off by a bit. You can see red through the target, but not the centre of the hole.

Edit: it works MUCH better in the daylight. It helps when you can actually see the camera.

Here's the next problem: nowhere in the package were instructions on how to do the alignment. Now I've done it before, so it wasn't hard to figure out, but why not put an instruction sheet in the box? I went to the website, and tracked down the instructions. Pretty simple: align the target and the lens, open it up all the way (the lens), shoot a picture using autofocus (remember, the target and the "0" point on the ruler are exactly the same distance away if it's parallel). Blow up the image on your LCD (nah. you're kidding, right? Use the computer, but see the paragraph below on "Tethered").  Find "microadjust" in your camera menu (they don't all have it, only the higher end ones. RTFM). Adjust it and shoot another shot. Keep doing that until you get it right.

Each lens has to be done separately. In fact, a zoom lens at different focal lengths will behave differently. For me, the longest focal length was the most critical, so that's what I based the adjustment on. Here's what I got:

 

Nikkor 70-200mm at 200mm f/2.8, unadjusted. You can see that it was forward focusing around 3 units (whatever they are) where the arrow is pointing.

 

Nikkor 70-200mm at 200mm f/2.8 after adjustment. This is set to +12 on a scale of -20 to +20. It's still not perfect, but it's close.

 

Nikkor 24-120 at 120mm, f/5.6, unadjusted. This is harder to see because the DOF is higher because it's a slower lens. The sharpest focus is around +8, backfocusing.

 

Nikkor 24-120 at 120mm, f/5.6, after adjustment. It took much less adjustment to align this one, about -5. I probably should hav dialed in -4 or -3 to be precise.

So once you get everything set up, it's a pretty easy task — just shoot and examine and adjust, shampoo, rinse, repeat until you're done. However if you have to take the card out of the camera, load it in the computer and upload the pictures between each shot, it will take you all day. If you're doing this, you have a high end camera. Probably you can shoot tethered to the computer (here's an argument for getting Lightroom!). That's what I did:

 

No comments from the peanut gallery. Shot with my Blackberry in, like, the dark. I loaned my P&S to Rosa so that's all I had to shoot with. You can see that the camera is tethered to the laptop on the left, the target is under the floodlight that illuminates my artwork on the wall. Not the greatest lighting.

Edit: Here's another shot I did today in daylight with the 200mm, now properly aligned. You can also see that I got the hole in the target properly aligned with the lens so everything is parallel now.

OK. Does the LensAlign MkII work? Yes. Is it worth the $80 list price? Again I'll say Yes, if nothing else than because it's a more professional way of doing things than laying out a tape measure. Should you own one of these? Well if you're fussy and you have a lot of money invested in cameras and lenses, then Yes. Remember I said that usage of your camera, especially if you regularly swap lenses causes things to change. When I did this two years ago, the 70-200/D300 combo needed a setting of +7. Last year after i got it back from Nikon, it went to +15. Now it's at +12. If you're a macro photographer and you're shooting focus stacks with rails, then no. If your thing is landscapes at f/11 on a tripod, then no.

And that's all I have to say about that.

— 30 —

Shooting with a 10-stop Neutral Density Filter

I recently got a 10-stop ND filter and wanted to share some tips about shooting with it. As time goes by and I gain more experience or get more hints from other people, I'll re-open this page and add to it. Check the version number here to see if there's new information you haven't yet read.

 

This is the initial post, version 1.0 on September 27, 2011
Updated: October 3, 2011 (bottom of post)

 

---

 

Background:
I had a variable (up to 10 stops) ND filter that I was really unhappy with. Not only was it extremely unsharp but also it generated a terrible interference pattern when used with my wide angle lens. The concept is interesting, because you can screw it onto the lens and change exposures just by rotating it, so you could focus, meter, compose, etc., then spin it. But the results were just too unacceptable. There are better brands available: Singh-Ray and Fader are well-thought of (the former apparently is clearly better, but pricey). But I realized that there was no real reason to use in-between values, like 6-stops or 7-stops, so I might as well go whole hog and get the 10-stop filter.

 

I bought a B&W brand filter made in Germany, and you can get it at B&H Photo — click here to access the page in their online store. This link takes you to the 77mm single coated version, you can find different sizes from there. The first thing you realize when you pick up this filter is that the build quality is superb. It's heavier than most other filters that I've had, for example.

 

There are a number of tutorials around. One of them is by Scott Kelby and there's a YouTube video here. If you're reading this on an iPad, I'm sorry but it's in Flash so you may not be able to view it. As usual, Scott gets quickly right to the point and his tutorial is great, but if you're new to using these filters, you may want a bit more information, so here goes!

 

There's LOTS to think about!

 

---

 

Here's what you can do with the filter, shooting fast water in the middle of the day:

 

Before:

 

 

This was shot without the filter. The exposure was 1/30 sec at f/11, ISO 100. The lens was the Nikkor 70-200mm f/2.8, set at 70mm.

After:

 

 

Exactly the same setup, except the filter was screwed on the lens. The shutter speed was now 61 seconds (I was trying for 1 minute!). This was in the middle of the day and the sun was shining brightly, by the way!

---

 

One thing you have to realize when you shoot these kind of shots: It's not like your usual digital experience: you know, we live in a world where people complain about how long it takes to heat something up in a microwave oven! This is more like the old days, where you have to consider carefully what you're doing and go through a whole bunch of steps and a mental checklist to capture an image. It will probably take you 5 or 10 minutes just to set up the shot!

 

For those of you who remember the old days: essentially what you're going to do is set it up, compose it, take a few "Polaroids" to make sure you got it right, then finally do your image. If you're shooting long dawn or dusk shots, it could be 2 hours before you're done with just ONE shot — so make sure everything is right so you don't waste your entire time!

 

What you need:

• Your camera, with a fresh battery and suitable memory card
• A sturdy tripod
• A cable release, preferably with a locking button
• Your Neutral Density filter
• Something to cover the viewfinder eyepiece to keep light from leaking in.

That last one is important (well, they all are!). Normally, your eye is blocking the viewfinder and of course exposures are shorter. Scott mentions gaffer's tape, which I don't like because I don't like leaving sticky residue on the camera. I usually just throw an opaque cloth over the camera, folded a few times. Brings back the old days shooting under the hood with a 4x5!

 

Remember also that holding the shutter open takes battery power. Don't run out in the middle, and if you have a Nikon with noise reduction, you'll have to keep it powered up for twice the length of the exposure.

 

If it's windy, even a dangling camera strap can move the camera too much. Weight your tripod if you can. Do it right — you're investing a lot of time to make it right!

 

---

 

Composition is up to you, but you need to think about how much of your image contains moving things (water, clouds in the sky...) and what's not moving (rocks, structures, old shipwrecks). There might be stuff that is moving that you didn't think about, such as trees, branches, leaves, grass, people... I like to try to think about the fixed, motionless things first and remember that anything that is moving is going to be soft and out of focus so it can't be the 'subject' of your image.

 

Set everything up without the filter. Focus carefully, then switch your camera to manual focus — your autofocus won't work through the filter, so the camera will hunt back and forth and maybe never find the right focus setting.

 

Take some test shots at the aperture you will be using to evaluate the depth of field. Use Aperture-Priority. Also find the right exposure for your shot. The metered setting may not be right. Note what the shutter speed is. You will need this for your starting point.

 

Most modern DSLR's are set up to default to 1/3 stop per click. Assuming you haven't changed this, you will need to click 30 times when you put the filter on. That can be any combination of stopping down the aperture, decreasing the ISO and lengthening the time the shutter is open. 30 clicks. Learn to count to 30 and try not to go the wrong way!

 

I find myself trying to calculate all this stuff in my head. Taxing for this tired old brain, even though I'm used to using the left half a lot! To make life easier for you, maybe this little "cheat sheet" will help. I created a .pdf file for you to download and print, it fits on both sides of a 3x5 file card which you can laminate and put in your bag. I did it in Excel, so you can also download the original spreadsheet and modify it for your own preferences.

• Link to the .pdf file.
• Link to the Excel spreadsheet.

 

 

Now carefully spin the ND filter onto your lens. I remove any other filters — skylight, UV, polarizing — the more glass hanging on the front of your lens the more distortion, reflections, etc. can creep in. Once it's on, you can't look through it any more so make sure everything is set the way you want.

 

Checklist:

• Focus (where you want it and on manual mode)
• Exposure mode: MANUAL
• Aperture set to the opening you want to use
• Shutter speed. Use 'bulb' and a timer like a stopwatch if it's over 30 seconds
• VR OFF if you're on the tripod (the new VR-II is supposed to be better, but why introduce hunting if you don't have to?)
• Exposure compensation and bracketing off
• Shutter release: use "mirror-up" or "self-timer" and a cable release to minimize any camera shake
• Block the viewfinder to keep light from leaking in
• Long Exposure Noise Reduction ON.

Now shoot your image. Be prepared to wait for the Noise Reduction process to complete (if you have a Nikon. Not sure how Canon does it). This should take the same amount of time as your initial exposure.

 

 

---

 

Great! I hope you get some outstanding images! You may find that you have to color correct them (long exposures turn things blue) or tweak exposures in LR or PS.

 

You have not used the optimum exposure values for the fixed, non-moving components in your image. You may want to shoot them separately and merge images for the best results. In this picture, I took 5 bracketed shots without the filter, then I merged them to HDR (using Photomatix Pro), tweaked and corrected them, then I imported the long exposure image as a fresh layer, and carefully masked the layers manually.

 

Update: October 3, 2011

I had a disappointing couple of days in Algonquin Park this weekend, focusing on using my ND filter. I learned a few things (for I am but a grasshopper...).

• I said it in my original post but I guess I didn't take it to heart: you can't use the water or any of the moving things in the image as the focal point of the image. They only enhance the composition that is already there.

My friend Dr. Ron put it succinctly and really well: you need to have something in the foreground which is sharp and not moving. The above shot is a great example of that, somewhat by accident, though. Don't just shoot fast flowing water, it's meaningless.

Long exposures and milky, silky water isn't for every shot. For instance, a really fast waterfall probably looks better with much shorter exposures because you wipe out any vestiges of detail with a long exposure. Also the transition between water and fixed portions of the image have to look natural, or else it looks like you just snipped and pasted it together. Here's an example (look where the rock meets the water)

By the way, long ND exposures do interesting things to the sky as well. I copied and flipped the sky and pasted it at low opacity into the water.

• Don't skip the test exposures.

Your meter may or may not give you a proper starting exposure, depending on the settings, the lighting, the effects you're looking for. Shoot without the filter, find your starting exposure and calculate from there.

• Don't forget to switch all the stuff off that you have to on the camera.

More than once, I found that I forgot to switch to manual focus, and once I had 5-shot bracketing selected. Stop and think!

• The "cheat" sheet I created above is invaluable.

Counting clicks sucks. Especially if you also want to change apertures or ISO.

<end of update #1>

Drop me a line if you have any other comments or suggestions (or comment here). I'd love to see your results. If you found this useful and you feel like contributing to my lens fund (!), every little bit helps! I'm trying to spend more time and resources on these blogs and your encouragement will really help.

 

 

 

— 30 —

Color Label sets — an obscure Lightroom feature

Lightroom and Photoshop are such incredibly deep and rich programs that there are many features most people don't know about. Although they're documented, users have to dig deep to find them and sometimes when you dig deeply enough, you find a gem.

Color Label Sets is one of those features. Here's another powerful way to classify, organize and search for specific images or sets of images in LR. Scott Kelby touches on it in his LR3 book but doesn't really say how to use it, nor did he point out the pitfalls!

This one reared its ugly head when I discovered that I couldn't retrieve the images that I had labelled as "Completed Keepers"  in my 40,000 image LR database. The reason? Well, the other day I decided to do a little cleanup, so I happened to go into the Color Label Set menu item (you find it under the Metadata tab in the Library module). I had previously changed the labels associated with the color labels but not knowing any better, never saved them anywhere. Clearly, I had changed them more than once because, as I discovered to my horror, searching for a specific color label under Attribute only finds the ones with the current text title associated with it. I changed, for instance, the text associated with the green color label (you flag a photo green by hitting "8" in Library or at most places in Develop (not when you have a tool active!)). I changed it from "Editing Complete"  to "Ready to Export".  I changed a few others as well.

Then I went into my main image folder — my whole library — and in Attribute, searched for images flagged green. "NO PHOTOS MATCH THIS FILTER". WTF?

LR wasn't searching for "Green" labelled photos, it was searching for "Green (Ready to Export)"  photos, and there weren't any. Don't panic, roll back to last night's saved catalog backup. No change. See this feature is a global LR feature and not tied to a specific catalog. Now panic.

I eventually posted the issue on the NAPP forum and a few hours later, I had a response from Michael Hoffman, one of the forum moderators, with the solution. Under Library Filters, use the Metadata tab, then change one of the columns to "Label"  and the whole list of all the labels you've ever used appears. I could breathe again! Thanks, Mike.

Now I got to thinking: there are lots of ways to categorize images in LR but not really enough for all my purposes. For example, suppose I've completed editing some images, they're "Ready to Export".  Some of them are really good, they're going up on my SmugMug page, or to my Blog, or to be entered in competition, or... and some of them were not that successful — I want to mark that they're done but they're ordinary. I don't want to throw them away... I'd like to mark them as "Edited - Archive".  But I've used up all my color labels. Not so fast! Let's turn this sow's ear into a silk purse!

Create multiple Color Label Sets!

FIRST: So you don't lose the ability to find images you have already labelled, save your current color label set. In Library, go to Metadata then Color Label Sets and Edit.

Select the little arrow to the right in the preset heading

then "save current settings as new preset".

Now you can always get back to those labels and find all your old marked pictures by selecting that preset.

Now create a new set of color labels. Then save them as a new preset, like this example:

So when I import a bunch of images, and I want to flag some burst sequences that I intend to merge into HDR's, I simply change the color set I'm using, then hit "7" to flag them. Later, when I want to find them, I do the same thing — I change color label sets and search for yellow labelled images! Or I use the Metadata tab in the Library filter to find them:

Note: I created this color label set just to illustrate this article. Then I flagged a couple of random pictures just to test it. I'll go back and make some more logical sets later. I'm thinking one set for when I'm importing, one set for post-editing... etc.

BTW, here's a bonus: if you leave a color label name unchanged from one set to another, then you can find ALL of the images with that label. Notice the 930 images labelled "Completed".  They're from both sets.

You could also create color label sets for specific assignments, or jobs, or events... the possibilities are endless.

Just another obscure but powerful feature in Lightroom 3!

— 30 —

Lightroom Workflow

I recently participated in a discussion in which I described my Lightroom workflow in 4000 words or less (LOL). I thought it might be worthwhile repeating here.

This isn't everything I do, nor is it what I do 100% of the time (although I should). I thought it might help the relatively new Lightroom (LR) user to organize his or her procedures. It reads more complicated than it is. Look through it, and please add your thoughts and comments.

The beauty of both LR and PS are that there are 60 different ways to do things, none of them wrong. Whatever works: but the key is to stick with a system.

IMPORTING PROCEDURE

After a 'shoot', I import to LR in a dated subfolder hierarchy (Example: 2011/2011-08/august13) and flip through all the images quickly and rate them as follows:

• First Pass
• Images that deserve a second look get a "P" for 'Pick'.
• Out of focus or just plain bad shots get an "X" and will get deleted from disk.
• Second pass ("P" images only), or sometimes on the first pass
• I rate very good, excellent and outstanding images "3", "4" and "5".
• I rate images I intend to run HDR on "1". If an image stands out, I might also give it a "6" (red) for "EDIT NOW".
• Now I go back and Keyword, if I didn't do it on import (if I have all shots from the same shoot on the import, I do it then. If they're mixed, I do it later).

After deletions, I have the following images in the LR database

• No rating, no stars - junk that I don't have the heart to delete. I'm a packrat.
• "P" rated, no stars - technically acceptable but really not interesting
• "P" rated, 1 star - sequences to be merged into HDR's
• "P" rated, 3-5 stars - the good stuff
• "P" rated, red flagged (likely star rated too) - stuff I want to edit NOW.

I can select any of these groups in Library.

PROCESSING PROCEDURE

HDR Images

• I currently create HDR's in one or more of 3 programs: NIK HDR Efex, Photomatix Pro and Photoshop CS5.
• I generally get .tif files back from the first two, which I immediately reopen in CS5 and convert to .psd files. So I don't really care what they're named before I do that. NIK and Photomatix have different naming conventions, but I rename the files coming out of CS5.
• I try to keep a vestige of the original filename, using the first one of the series of 3-7 images. I try (OK, I'll try harder!) to add HDR to the name - example: FAC_5233HDR.psd.
• I try (!) to put the program used (HDR Efex, Photomatix, CS5HDR) in the keywords for searching purposes.
• I rate with 3-5 stars (or if it's garbage, none)
• I flag it with "7" (yellow) if I'm not finished editing it
• I flag it with "8" (green) if it's ready to export.

Other Images

• I have two kinds of files
• Those I've edited in LR only - NEF files
• Those I've exported to CS5 to edit - PSD files
• I'll have a rare file that came from somewhere else as a JPG.
• I flag it with "7" (yellow) if I'm not finished editing it
• I flag it with "8" (green) if it's ready to export.

EXPORTING PROCEDURE

Exporting to a file
I have half-a-dozen presets in place depending on what purpose or location the export is going. It only takes a keystroke or two to modify them for specific use. They all go to their respective subfolders on the internal HD drive

• Export for Blog: fit to 1280x1024 and add watermark
• Export for DropBox: not resized, not watermarked.
• Export for RHCC competition: 1280x1024, no watermark
• Export to iPad 1024x1024 (the longest side becomes 1024px) no watermark. As part of the discussion, I came to realize that doubling the iPad screen size is a better idea so that the image looks better when zoomed in, so I'm changing that to 2048x2048.
• Export to web 800x800, no watermark.
• Export for NAPP 720x540, watermarked. These are the specs for the NAPP forum. I use the 800x800 format for uploading to my NAPP Portfolio.

Getting the images to the iPad is a pain in the a$$. I wish Apple would do something about this. Generally I transfer them to another folder in Dropbox and pick them up on the laptop (it's enabled as my iTunes machine), then move them into a dedicated subfolder that syncs with iTunes (feh. Ptooi). I could also pick them up individually via DropBox if I'm in a hurry, or if I can't connect the iPad to the computer. Advantage: they can be deleted from the iPad. Disadvantage: they can't be organized in folders.

Publishing to SmugMug

This works pretty well, assuming the folder on SmugMug has already been created. Technically, you can do that on the fly in LR but I've yet to be able to make that work. So I go to SmugMug, create the folder, then go back to LR and sync it and it finds the new folder.

• Select the images you want to publish
• Write captions (I like to caption the images on the gallery)
• Drag them to the appropriate SmugMug publish collection
• Click "Publish Now".

Publishing to Costco for printing

• Select the images
• Use "Export" because Costco doesn't show up under Publishing Services
• Log into your Costco Account
• Creating a folder is a pain in the neck. You always end up having to rename it later. Not important.
• Click "Export".
• Go to the Costco site (usually I have to log in again) and place your order
• Pick up your pictures the next day.

AFTER EXPORTING

• Re-flag exported images with "9" (blue) for "Completed".

By the way, you can rename your color label sets or create a new preset via Library->Metadata->Color Label Sets. However this can create a problem for Lightroom. The program ties the Attribute search to the Color Label Set currently in use. So if you change the name associated with a color label, it may not be able to find it. With a tip from Michael Hoffman on the NAPP forum, He found a way to work around it, but I also turned the problem into an opportunity. Before you do anything else, go into LR and SAVE YOUR CURRENT SETTINGS AS A CUSTOM LABEL SET (same menu item). I'll document this procedure in a separate post (here).

BACKING UP

• All of my images are imported to my external 2Tb drive, and an automatic copy goes on the internal drive, in a folder called "Temporary Image Backups".
• Every day, LR prompts me to do a catalog backup which also goes on the external 2Tb drive in the same folder
• Once a week or so (when I'm being good) I run SyncToy and duplicate the Images folder onto a second external drive
• Once a month I'll delete the 60-day old files from the internal drive.
• Really critical stuff gets backed up elsewhere as well - multiple computers, the occasional thumb drive and, more frequently lately, up in the Dropbox cloud.

I know I need to address this some more. At the very least, another external 2Tb or 4Tb drive.

If I said I follow these procedures 100% of the time, I'd be lying. Sometimes I'll dive into an imported set of photos and start editing. But I'm going to give it 95%. Being able to find things later is one of the powers of Lightroom.

Perspective Cropping in Photoshop

On the NAPP forum, someone was asking for advice because the pictures he was cropping were getting distorted, faces elongated. I suggested that perhaps he had the Perspective box selected when he was cropping. I still don't know if that was the problem, but the following is the answer I wrote. The question I was responding to was, "so what does the perspective box do?"

Some of the tools in Photoshop are mysterious when you haven’t used them regularly, if at all. I think the perspective cropping tool falls in that category.

OK, a picture is worth 1000 words. I just grabbed a shot of my grandson who REALLY, REALLY, REALLY needs a haircut. Here's the original, trimmed down to fit better here.

I decided to crop it to an 8x10 ratio.

Notice that this is without the perspective box selected. If I drag a corner, it keeps the same aspect ratio, just includes more or less of the image.

Now I did it again with the perspective box ticked:

Nothing else changed, except I dragged the top down and the bottom up. Doesn't look like an 8x10, right? But here's what you get when you invoke the crop by double clicking inside the selected area:

Elongated face! The reason is that with the perspective box selected, the cropping tool will take whatever's selected and make it fit in an 8x10 aspect ratio. It will stretch or shrink it as necessary to make that happen.

I'm guessing this is what's going on for you.

The perspective control is a very powerful tool. You can use it to make lines that are not parallel, parallel (couldn't think of a better way to say that. Oh wait: a picture is worth...)

Drag the cropping rectangle (OK, not a rectangle any more, a quadrilateral) until the sides are parallel with the things you want straight up and down or sideways

Magic!

Perspective cropping in Photoshop. That's how it works.

— 30 —

Button, Button

This is a bit of a departure. This is about using Photoshop to create buttons for your website. It has nothing to do with photography. But everyone has a website, right?

One of the basic rules of design is to maintain a consistent look and feel  and one way to do that is to make all your user interface buttons look alike, or at least similar. I set this up some time ago for my First Aid site, and I use the same devices on my other sites.

The idea is to be able to quickly change the text on a button, and possibly to change the colour of it as well, for example if you want it to change colours when you hover over it.

If you're already using Photoshop to create webpages, with sprites and slices and good stuff like that, this is going to seem rather simplistic. But if you're like me, and you only know a little HTML (and perhaps a smattering of Javascript), then this might work for you. Without further ado...

It took me less than 30 seconds to make the second button from the first one. Want to know how? It will take you a few minutes to create a master template then a new button is only a few seconds away.

Create a new document in Photoshop. Keep it larger than you're going to need, you can reduce it in your HTML. Don't worry, it'll be a small file.

Create a type layer. Use whatever font you want. You can scale it, resize it, change it any time in the process.

Create a shape. I used two circles and a rectangle to create an elongated button (I want room for a couple of words). The colours are just so you can see what I'm doing. Recolour them the same and merge the layers together.

Open the Layer Styles dialogue and make it pretty! I used a bevel and emboss, gave it a drop shadow and used a gradient overlay. Note that to keep the colour in place, the blend mode for the gradient has to be "overlay".

Duplicate the layer, open the Layer Style again and add a color overlay. You can change the colour of the overlay in the dialogue.

But we lost the gradient overlay. (OK, that was an "oops". But I have a solution which gets you a bonus too!)

Convert each layer to a Smart Object. That way you can change anything you want with ease. That's the bonus. Go back into the Layer Styles and add your gradient.

By the way, when you create a button, turn off all the unused layers because the drop shadow multiplies. It's outside the filled area.

Here's another hint. If you do a gradient and you make it darker on the top, the button looks concave. If you make it darker on the bottom, it looks convex.

You can edit the type layer just as easily. You can even add styles to it, change the colour, the size... anything! Move it to the top.

Of course, you have saved your work as you went along. In the end, you have a .psd file with a layer of text and as many layers for colours that you want. Normally you would use colours which are coordinated with your website palette.

To create a button, open the PSD file, choose the colour layer that you want, choose the type layer and modify it.

To save your button to use on your website, choose "Save for Web & Devices". If you make it a GIF file and leave Transparency on, the drop shadow will work with your web page background.

Complicated? Not really. Does it make it easy to give your website a consistent look? Can you create buttons in a few seconds? You bet.

If you never really used layers in Photoshop, this might give you a bit more understanding of how they work. Smart objects too.

-- 30 --