tom_mann1

<p>Following up Andrew's suggestion, here is an example of what one can achieve in ACR. I started with a high contrast outdoor photo taken on my iPhone and reduced the contrast and added a color cast. 95% of the work was in ACR, but I took the image into PS to juxtapose the before and after versions next to each other. If this is better than what you have been able to obtain, or you simply are interested in this approach, let me know and I'll post some screen shots of the ACR sliders.</p> <p>Tom M</p> <p>PS - My guess is that the photographer used a scrim on the 2nd shot that you posted (ie, the outdoor one), so the amount of work that had to be done in PP was considerably less.</p> <p>PPS - At the risk of stating the obvious, every starting image will have different white balance, different contrast, etc., so one absolutely must know how to appropriately "tweak the sliders" based on what you see, and not attempt to use a canned action, ACR preset, or anything similar. </p><div></div>

JDM, I understand, greatly reduced my own activity here a couple of years ago, but am more active than ever in photography & on another forum. Best of luck & have barrels of fun! Tom M

Hi Sarah - Sorry about the poorly worded earlier response, but it looks like it was at least somewhat decypherable, LOL. BTW, unfortunately, I only have the latest version of PS installed, so my observations only apply to it. All the best, Tom

A question, David ... Looking though some of your sample images, it appears as if the dynamic range of the image is reduced by stacking. For example, in the shot of the Lincoln Memorial, the bright lights do not have the bright halos around them that I would have expected if I shot a single longer exposure sufficient to brighten the trees to the level in your photo. If my observation is correct, and you are using the "add" mode, I don't see how this can possibly decrease the dynamic range. The same goes for the "brighten" mode unless the image is flickering from exposure to exposure. Can you speak to this a bit? Thanks, Tom M

<div></div>

<p>Hi Sarah - When I added 1% of noise to the 1st 16bpc test image that I tried, I saw just about the same amount of combing / jaggedness in the histogram both before and after I apply a substantial curve (either as an adjustment layer, or directly). </p> <p>However, if I don't add the noise, then I see effects like you describe.<br> <br />What I didn't mention was that my first 16 bpc test image actually started out life as an ordinary 8 bpc file to which I did a simple conversion to 16 bpc. In contrast, if I start out with a "real" 16 bpc image (say from ACR), it acts just like the image after adding 1% noise.</p> <p>My guess if one's starting image really was 8bpc and then was artificially up'ed to 16bpc, the numerical precision of the curves math being done in 16 bpc mode is so good that it effectively retains the 8bpc character of the starting image. In other words, when viewed on a 16 bpc level of resolution on the horizontal axis, a histogram that started out as a picket fence, stays a picket fence, just that the pickets get moved around a bit so that when they are graphed on the 256 bins of the histogram display, some of the moved spikes will line up with a bin, while others will not, and hence you see a comb like display.</p> <p>Just my $0.02,<br> <br />Tom M</p> <p>PS - I hope that was understandable. I wrote it being half asleep, LOL.</p>

<div></div>

If you are not familiar with "HDR photography", you should read through a few of the many articles on the subject. Although the details of the method are slightly different from the method u suggest, the end result sounds exactly like what u are looking for. Tom M

<p>To be honest, I rather liked the original. To my eye, that's pretty close to the way it can look in that area. However, if I was asked to add pizzaz to the image, I would probably do something like this, always giving the end user the option of blending in some of the original if they don't want so much effect. I used a bunch of efx, but probably the most important single effect was Topaz Adjust.<br /> <br />Tom</p> <p>PS - And, yeah, I intentionally left the power lines and power pole in place. I thought they kept the realism of the scene and added a very subtle inner frame.</p><div></div>

<div></div>

<p>In case anyone is interested, the OP asked this question on other websites /forums and got a very nice answer here:<br> http://photo.stackexchange.com/questions/61695/does-getverticalviewangle-give-actual-camera-angle-field-of-viewcafov-value</p> <p>Tom M</p>

Personally, used as an off-line, after-the-fact analysis tool, I have no need for this. I wouldn't use it if it were given to me free. However, I could image that if it were tightly written, low level code for some of the microcontrollers used in cameras, there might be some interest from mfgrs of consumer level cameras. Tom M

<p>... and adding a bit of a green-yellow "cross processing" effect (Google it.)<br> Tom M</p>

<p>If I have an architectural image that has several buildings, each with a sunny side and a shady side, plus a foreground and one or more distant background areas, sure, I might start work by immediately generating a series of fairly sharp edged masks that will isolate the various areas so that I can work on them entirely independently, and then mix them back together at the end of the process. I might do something similar if I have an interior shot in mixed lighting and I need to set different white balances in different areas of the room. In this case, I might do two or more ACR conversions and using soft edged masks select the areas to which they will be applied.<br /><br />However, I must say that with portraits, for me, usually, a simple sequential approach using masks that I generate as I go works just fine, especially if combined with converting the occasional layer to a smart object if I think I'll need to come back and tweak an adjustment made to that layer at some time in the future (eg, I did a Gaussian blur and may want to come back and adjust the radius).<br /><br />The other PS feature that I frequently use is "merge up" ( http://frenchkisstextures.com/tutorials/photoshop/photoshop-keyboard-shortcut-copy-visible-and-paste-in-a-new-layer/ ) , ie, copy all the visible layers and merge them into a new layer (Ctrl-Alt-Shift-E on the PC). <br /><br />Hopefully, you already are familiar with this technique, but the effect is very, very different from a simple duplication of a layer, a technique that you mentioneded in your post at least twice. <br /><br />Doing a "merge up" doesn't affect the appearance on your screen, but, as you observed, duplicating a layer certainly might do so. A "merge up" layer also gives you a snapshot of where you were at a particular point in processing the image. This means that you can always revert to exactly this state, or, if you decide that changes you made after this point are a bit too much for your likeing, just make a copy of the "merged up" layer stick it on top of the layer stack and drop it's opacity to continuously adjust between the results at two points in time. Also, there is no need to do only one "merge up" -- you can change the adjustments involved in the "merge up" and save several different versions.<br /><br />After re-reading your post, I'm starting to think that the fundamental problem you are running into is because you are thinking in terms of a completely parallel approach to processing images, ie, fix a, b, and c separately, and then merge them into one nice image, versus a more serial / sequential / incremental approach to post processing.<br /><br />There's a good chance I didn't understand your problem, and, if so, just tell me, but hopefully the above paragraphs might have given you a few ideas.<br /><br /><br />Cheers,<br /><br />Tom M<br /><br /><br /><br /></p>

<p>FWIW, there is a nice, fairly recent thread on this exact topic <a href="https://www.photoshopgurus.com/forum/general-photoshop-board/53673-request-group-pic-please-help-removing-raindrop.html">here</a>.</p> <p>Tom M</p>

<p>Thanks for the comment / idea, Lex. I can't wait to get back to this, but I've got to get done with some pix for fire dept recruiting posters before I can turn back to this project.<br> Tom</p>

<p>Thanks for your comment, Ian, but I have some very specific requirements that wouldn't be met with a simple diffuser cap. As I said in my opening post:<br> 1) <em>"I would like to create a soft-edged, small-source-size, flat, wide beam ..."</em>; and,<br> 2) <em>"The horizontal spread should be of order +/- 30 degrees, but I would like to keep the vertical beam divergence to be as small as an SB910 (or its older siblings) will give me when the flash is zoomed out to its maximum telephoto position."</em></p> <p>Unfortunately, a diffuser cap will spread the beam in both the horizontal AND vertical directions.<br> <br />Regards,<br> <br />Tom M</p> <p>PS - I had to turn to some other work, so I haven't yet had time to experiment around as suggested above, but I will post in this thread when I try thing out.</p> <p> </p>

It's a pretty simple mechanical device with just about all of its components easily accessible for inspection. If u get a good return policy, you should be ok. I doubt any serious defect will remain hidden. Tom M

<p>@Tim - A good observation. Interesting, as always.</p> <p>Tom M</p>

<p>Hi Wade - a couple of thoughts:</p> <p>1) re: <em>"...looking at the many "Lightroom Wedding Presets" offerings out there, it seems obvious what's prevalent..."</em> <br> I would be a bit hesitant to make such an inference. The goal of the people selling / offering the presets and actions is to sell <strong>more</strong> presets and actions, LOL. The goal of wedding photographers is to make the clients happy. These two goals may be related but definitely are not the same.</p> <p>2) re: <em>"...Is skin smoothing a good thing for most shots, btw?..."</em> <br> One absolutely doesn't want to make the subjects' skin look plastic. However, if you are trying to give the image a soft romantic look, you've got to fight the tendency of digital images (especially after enthusiastic processing) to have very strong microcontrast. You still want the image to be sharp (ie, the width of an edge), but you don't want the change in brightness or color as you go across an edge to be quite so large. One easy way to do this is to put a smoothed version of the skin on top of the sharp version, but set it to 70-90% opacity, so that a little of the underlying sharp image shows through, but the contrast across each edge or dot is reduced. This is what I did.</p> <p>HTH,</p> <p>Tom M</p>

<p>Whenever I think about adding some effect to a wedding photo, I force myself to imagine how the proposed effect will be perceived by the couple and their family 30 years from now. Will they wish that the photographer had just given them a nice straight shot, or one with efx that are well accepted in this genre (eg, romantic soft focus, light vignette, etc.). For example,would they like the wedding pix of their grandparents modified to look like they were shot in the 1800's when they were really shot in the 1960's, or would they find this confusing and unnecessary. Would they feel that the keyhole vignette effect that they run across in their parents' wedding album was kitschy or an interesting true vestige of that era? etc.</p> <p>Anyway, unless the couple specifically asks for something else, my vote is almost always for minimal / traditional post processing. The attached took me well under a couple of min to do using LR and Tiffen DFX.</p> <p>Just my $0.02,</p> <p>Tom</p> <p>PS - I was too lazy to do something with the disembodied head that's between them. I'd certainly make it less prominent, and maybe eve remove it completely.</p><div></div>

<p>One of the other amazing feats that organic lenses can pull off is produce a striking three dimensional stereo rendering with a single lens and single exposure. If you can't see it, there is something wrong with your eyes. ;-)</p> <p> </p>

@John - That's a great idea. Taking your concept further, I would probably saw a strong positive, spherical Fresnel into several rectangular pieces (along the x-axis), not just two, and swap the outermost pieces (along the x-axis) to roughly approximate a strong negative cylindrical lens. Thanks! @Rodeo - Diverging the beam by the inclusion of such a lens will, of course, greatly decrease the optical joules per square cm at a given distance, but other than some minor reflection and absorption losses in the Fresnel, it shouldn't decrease the total energy output (Joules, not Joules / sq cm) by very much, just re-direct it. This is why I said that I wasn't interested in absorptive devices like snoods or tape. Also, I don''t think that putting horizontal strips of black tape across the front of the existing Fresnel in the 910 will have much effect. Since the source of light in such a flash is extended, and not a point source, off axis beams come from all points on the Fresnel, not just its outside edges. So, tto get good edge cutoff, I would probably have to put the tape that you suggested out at the end of a tube, which I'm trying to avoid. Thanks, tho. Best regards, Tom M

<p>Arghhh! Where did the edit button go? </p> <p>I meant to say *CYLINDRICAL* negative Fresnel lens, not just a standard spherical Fresnel.</p> <p>Tom M</p>

<p>For a project I'm going to be starting, I would like to create a soft-edged, small-source-size, flat, wide beam from a conventional, shoe-mount strobe (eg, SB 910 or equivalent). <br> I would like the apparatus to do this to be as small, portable, and light efficient as possible, so I'm not enthusiastic about snoots, gobos, a custom grid, and similar absorptive modifiers.<br> The horizontal spread should be of order +/- 30 degrees, but I would like to keep the vertical beam divergence to be as small as an SB910 (or its older siblings) will give me when the flash is zoomed out to its maximum telephoto position. <br> I came up with the idea using a negative Fresnel lens to do this. They are not as common as positive Fresnels, and a bit pricey, so I was wondering if anyone here has ever tried something like this, or if anyone could suggest any alternatives.</p> <p>Thanks in advance,<br> <br />Tom M</p>