paddler4

Indeed. Hence my initially simple post--dodge and burn. For a beginner, getting the simple advice to dodge and burn may be more helpful than detailed explanation of a complex method for doing so, like the one I gave in response to your criticism. I don't think anyone here was "expecting [the OP] to know all the tips & tricks of post processing". Exactly the opposite: several of us were offering suggestions on the assumption that the OP doesn't yet know these tricks. and if avoiding complexity is your goal, let me remind you of part of your response to my simple suggestion that the OP dodge and burn: I'm not going to respond further, as I don't see the value in continuing a gratuitous argument that is a distraction from the help the OP asked for.

Arlingtonbeech, I think this thread may have led you astray. I finally saved your image and read it into Photoshop. You don't have room to expose more. The thin white circles have luminosity values up to 96 (out of 100), so if you exposed much more, you'd blow them out. The face is darker than you would want, but the luminosity values are fine, mostly ranging from the mid 30s to the 40s and even 50s. There is plenty of detail there. I think the solution is a simple dodge and burn, using a technique like the one I described, not a change in the exposure. For example, here is a quick and dirty edit, with a substantial brightening of the face and some darkening of the background. Paddler4

Ah, samstevens, in general, I agree, but I don't think that was the problem here... I was actually being precise. "Burn" doesn't mean "use the burn tool". It means the process. When I started out, it usually meant waving a piece of cardboard with a hole cut out of it underneath the enlarger. Even people inexperienced with photo editing can find 3 or 4 (or more) ways of dodging and burning with a quick web search, most of which don't use the dodge and burn tools. For example, this page has 4 methods, only one of which uses those tools, and it's not a complete list. It omits at least one common one (which I never use), which is using a neutral gray layer, e.g., Photoshop Dodge and Burn with 50% Grey Layer – SLR Photography Guide. I've had some back and forth with one very experienced retoucher about this. He often uses selections, but I find it hard to get smooth transitions that way, so I use the method I posted, which he also taught me. One additional advantage of this approach over some others is that it makes it trivially easy to reduce the dodging and burning. Just switch the brush to black. As long as the flow rate is very low, you can make fine adjustments that aren't at all apparent in the final result--that is, you can't tell where the boundaries are.

Really. Who knew? I actually didn't suggest using the dodge and burn tools. I suggested dodging and burning. Since you assert that this is "downright crude", I'll give you the method I usually use, which was given to me by an expert photo retoucher. I'll use dodging for an example, as it would be more important in the OP's image. 1. add a curves adjustment layer, bring up the midtones (or other parts of the curve if you want to be fancy) until the area that you want to dodge the most is at least as light as you want. No harm in going a bit farther, as you'll see. 2. Invert the mask. 3. Select fairly soft white brush (hardness depends on the image), with 100% opacity and a very flow flow. I typically use between 9% and 12%. This is important, as opacity and flow work very differently when you are attempting to build up an effect. 4. Slowly paint on the burning where you want it, building it up more where you want a stronger effect. The advantage of this approach is it completely separates relative amounts of burning, which you control with the brush, and the absolute level of burning, which you can control by changing the curve itself. And, of course, you can lower opacity. Somewhat similar to your second method, except that it dispenses with the superflous steps and pixel layer.

It would be helpful if you showed us the histogram. The suggestions to spot meter off the face to better expose the face make sense, except: (1) if the histogram spans the entire range, increasing exposure in that matter will cause some areas to blow out, and (2) regardless of #1, the imbalance--much brigher lighting behind the face than on it--will be unchanged. that is, the entire shot will be brighter. So, what I would do is one of these: (1) if there is room on the histogram, increase exposure (spot metering on the face would just be one way to do that), and then burn the background in post to darken it. (2) if there isn't room on the histogram to increase exposure, use the exposure you have, brighten the face by dodging, and darken the background. Unless you are blowing out the highlights, no. The problem comes from underexpositing the face.

This isn't because of the DNG format. I shoot Canon and Panasonic, so none of my raws are DNG, and I've had it happen. However, I usually only notice it later, so I'm unable to reconstruct exactly the workflow.

I shot JPEG when I first started, as I had the misimpression that raw was much harder. Not all that long after I started, a cousin who teaches photography gently urged me to start shooting raw. I nervously shot raw+JPEG for a few weeks but quickly realized that I was never using the JPEGs. Simply rendering the raw file--e.g., importing it into Lightroom--takes no work, and the initial rendering is quite enough to show you whether it's worth keeping.

I have this from time to time, but my workflow is sufficiently varied that I haven't isolated when it happens. This might be useful: Metadata Mismatch – Ask Tim Grey

Thank you for putting this so well. One of the most common and most serious mistakes many newbies make is thinking that postprocessing is just a way to compensate for bad camera work. I remember being at an informal exhibit where a photographer bragged that his images are SOOC, as though this is a virtu, even though some of his images could have used some postprocessing. Quite apart from the fact that the scene may not give you precisely the image you want to create, the conditions are often not ideal. For example, you may be confronted with lighting that provides too little contrast or that brightens the wrong parts of the image. Learning postprocessing (it really should be called just "processing" if you are shooting raw) is a core part of becoming a competent photographer--that is, competent in the sense of being able to create the image you want. A badly exposed photo will be just as badly exposed if shot as JPEG as if shot in raw. They can't, since unrendered raw files can't be posted. If the OP wants to shoot raw (which I think is in most cases the better option) and wants feedback before he does editing, he just has to render the image in a raw converter like Lightroom, using either the default profile or any of the other available profiles and export a JPEG from that to post here without doing any editing. The advantage of doing that over posting an in-camera JPEG is that most camera JPEG settings apply substantial adjustments to color, contrast, sharpness, etc., so they don't do a good job of showing the basic material the photographer has to work with.

None. It's not as though I have a profitable business to protect. If people are actually interested enough in one of my photos to ask how I got certain results, I'm more than happy to tell them, and I hope others will do the same for me. In fact, a couple of years ago, I had a focus-stacked image of a flower in an exhibit, and I was asked to give a little zoom session explaining what I did. I considered the request a complement. Because I take some field macros of bugs, which usually involves a huge number of failures for each successful image, I've often told people how few good images I get. Why keep it secret?

One on-site mirror, one cloud-based backup (BackBlaze).

I've finally started digitizing some very old family photos--most going back only to the 1940s or 1930s, but one dating from roughly 1885--and doing a little digital restoration. It's slow going, and I won't get all that many done, but every one I complete is immediately backed up twice, so there are three digital copies of it as well as any prints I make.

Actually, it does. There are lots of types, and they store data differently. See, for example, this description. Regardless of the method used, the software will have a method for re-assembling the pieces. Alan, this disagreement is irrelevant to your question. If you want to be able to access the copies without using a restore function, a mirror (copy) is your answer. But if your only concern is being able to retrieve the file by some method, either will work. The file manager I used for my mirror, Directory Opus, makes this extremely simple. I have it configured so the active window comprises three panels: one on the left that shows the directory tree, and then two larger panels that show specific parts of the structure, both directories and files. I have all of the photos in my active catalog under one parent directories named "photos". So I just set the middle and left panel to the same part of \photos on the different drives, and drag from the copy to the mail disk. Trivially easy. It also allows you to compare file dates.

Alan, There are several kinds of backups. You can find an explanation here: https://www.partitionwizard.com/clone-disk/types-of-backup.html No service is going to do a full backup every time. That would take days or weeks. They will do either incremental or differential, I think usually the former because it's quicker and takes less space. A differential backup gets progressively longer. None of these do what you want. What you want is a mirror: a copy of the relevant drive. That's why I use a mirror for my first "backup" and a true backup for the second. What de-bakker is suggesting is also a mirrored copy, just using different devices. I have a new desktop on order. When it arrives, I'll first install my file manager / synch program of choice, and then I'll simply drag the folders I want from the mirror on my external HD to my new computer. This won't take care of everything; I'll still need to install sofware, set program paramters, yada yada. But all I have to do to put all of my photos onto the new computer is one drag and drop. I can also be selective if I want; I can put some files into different locations on the new drive with no difficulty. It's all simple, and there is nothing obscure. I use the online true backup as a last resort. I have used it on rare occasions--e.g., if I mess up a file on the mirror as well. Dan

This is not entirely correct. Most often, backup software creates "incremental" backups--that is, it copies only what has changed since the last backup. In the case of large files, this means that new or modified sectors will be copied, but unchanged sectors won't be. All sectors are present in the copy or backup either way, but in the case of a backup, the software has to reassemble the pieces. The end result is the same, but it's a bit more cumbersome because one has to tell the backup software to restore the file, rather than simply dragging it from the mirror.

A mirror is a simple copy, where the file is treated as intact. You can simply drag from one disk to another. A backup stores fragments of files; if you only change certain sectors, those are stored in an incremental backup. The restore process reassembles the pieces.

Ken, If you want something cheap and easy, use an external hard drive (as Alan said, $100) and sync software for one mirrored copy and something like Backblaze for a cloud backup. I don't know whether you use a mac or a pc, and the sync software differs. I use a PC, and there are loads of options. Microsoft's synctoy is free but a little kludgy. I like to sync from within a good 2-pane file manager that allows me to see one disk in one pane and the other disk in the other. I use Director Opus Pro, which is an extremely powerful and flexible file manager, but it costs about US $62. There are cheaper options. Tim Gray recommends Goodsync Backup Cleanup with GoodSync – Ask Tim Grey Personally, I find a mirrored copy much easier to work with than a backup, so I use the cloud backup as my option of last resort. Dan

My point is simple: unless you have too much material, mirroring to an external hard drive works well and costs very little. I've been doing it successfully for years. If you have to much material, it won't work. I'm not close to that point.

Sorry, I left out a word. You can buy a 4TB EXTERNAL hard drive for $150-200. I make do with w TB

You can buy a 4 TB hard drive for $150-$200.

I don't understand this. A RAID array and a NAS are both local copies. What do you mean by "connected to the web"? Do you mean that you keep a copy of one of these on a server somewhere, similar to an online backup? I don't recall whether it's standard or an option, but BackBlaze allows you to store your backup with an encryption key that they don't keep. My recollection is that they warn you that if you lose the key, they can't do anything to help. I do agree that a mirror is handier than a backup, which is why my first "backup" is a mirrored copy on an external drive. Not a NAS; I find it works fine just to sync to a regulare external drive.

or a power surge. Or a water leak. An off site backup is like fire insurance: a low cost protection against an event that's very unlikely but that has a very high cost if it ihappens.

The key thing, IMHO, is not whether the second backup is in the cloud, but only that you have two and that they are in different places so that if your computer and local backup are both damaged (e.g, by water), you have another backup elsewhere. re online backup speeds: once the initial backup is done, you aren't even aware of the backup going on. It runs seamlesslessly in the background. However, if you add a lot of material to your local drive at one time, don't have the computer running for very long, and have a slow upload speed, there would be a risk of not having the backup completed before you turn the computer off. I haven't done the arithmetic, but it would have to be a lot of material and a very short computing session to make this a problem for raw captures. Images edited in a program like photoshop can be 10-20x as large, but those take time to create. Even an 11 mps upload is 82.5 MB/minute. This is less an issue for me; I had 70 mps uploads in the past and have 200 mps now, thanks to optical service. But nonetheless, this is one reason why I make my local mirrored copy as soon as I upload photos, to make sure that I have one backup even if there is temporarily some material lacking a second. It would be the same issue if you physically carry a second backup offsite

All online backups take quite a while to make the initial backup unless you have very little material. The bottleneck isn't usually your backup company; it's your internet connection. Particularly if you have cable, which is asymmetric, uploads are very slow. I have Verizon FIOS, which is symmetric, but even so, it took quite a while. If my arithmetic is right: I have 1.15 TB filled on the HD that has my photos. That's 1,150,000 MB, which is 9,200,000mb. I have 200mb/sec FIOS, so if there are no bottlenecks other than the internet connection, that's nearly 13 hours. 200MB cable service is likely to have far slower upload speeds, probably more like 12 mb/s, so that would be 213 hours. Again, if my arithmetic is right. But in any case, it is VERY slow. However, once you have the initial backup done, the incremental backups are quite fast and aren't even noticeable, as they operate in the background. I've used two companies but am currently using BackBlaze. I'm satisfied. However, I keep a second backup--a mirror, actually--as one really should have two (one on-site and one off-site), and I generally find it easier and faster to restore from a local mirrored image. Any sync program can create the mirror. I use Directory Opus, which is also a superb file manager.

With a given sensor, DOF is primarily a function of distance to the subject and secondarily focal length. FL has only very small effects on DOF, contrary to popular thinking, although it does affect the balance of DOF front / back. See for example https://www.cambridgeincolour.com/tutorials/depth-of-field.htm. So if you shoot exactly the same thing in the same way, there shouldn't be a difference in DOF. However, to do these tests, you need to be shooting in full manual mode or aperture priority. If you shoot in P or some other automated mode and the camera changes the aperture setting, you will get differences caused by that, not by the differences between the two lenses themselves.