May 19, 2013, 02:01:51 PM

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - TrumpetPower!

Pages: 1 ... 36 37 [38] 39 40 ... 65
556
EOS Bodies - For Stills / Re: ISO 50
« on: January 19, 2013, 07:35:35 PM »
so regarding of the subject (motive with no high lights = motive with a small DR) you can overexpose and get a benefit of the over exposure in the shadows


That's great information, thanks! I'll certainly use that once I've got a 6d and am doing tripod macro focus stacks with low dr objects.


If you're going to the trouble of focus stacking...well, first, with macro work, there's often all sorts of specular highlights with lots of colorful interference patterns and other things going on, such that you might have a great deal more dynamic range in the scene that you want to preserve than you initially realize.

But, back to point...if you're focus stacking, there's no reason you can't either do "standard" ETTR (if it truly is a scene with a limited dynamic range) or HDR (if necessary). You've already got a workflow that involves lots of scripted steps; what's one more step to script?

I would note, though, that the exposure adjustments for either ETTR or the type of not-tonemapped HDR you'd want are really best done in the camera's native linear space before any sort of gamma or other tone curve is applied. I don't know if ACR / Lightroom does that, but most tools based off DCRAW do. I'd especially recommend Raw Photo Processor: http://www.raw-photo-processor.com/

If your tool of choice doesn't do exposure adjustments in linear raw space, then you're best off nailing exposure in camera, and ISO 50 therefore becomes (in that situation) a better bet than ETTR.

Cheers,

b&

557
EOS Bodies - For Stills / Re: ISO 50
« on: January 19, 2013, 04:33:39 PM »
Expanded ISO means digital gain (negative gain for ISO 50).  The exposure is at ISO 100, then pulled down a stop.  In your example, you'd lose the highlights.

-- assuming, of course, there are highlights to be lost.

There almost always are, which is why ETTR is generally not such a great idea. But, when there aren't, or when you truly don't care about losing them, then, yes, ISO 50 or ETTR is a sometimes-useful tool to have in the toolbox.

b&

558
EOS Bodies - For Stills / Re: ISO 50
« on: January 19, 2013, 03:22:16 PM »
There might be a bit of analog jiu-jitsu contributing to the mix, but, with that caveat, the linear data recorded at the sensor with a shot at, say f/8 @ 1/400 @ ISO 100 ("sunny f/16") will be identical to the same shot at f/8 @ 1/400 @ ISO 50. At some point in the processing chain, however, the values in the ISO 50 file will get halved (by simple arithmetic, not electronic amplification), resulting in an image one stop darker than the ISO 100 file. That will reduce noise overall. However, the sensor is still saturating at the exact same point. The net effect is that a pixel that the sensor recorded as, say 256, is being rendered as 128...and that there's nothing in the original data that gets mapped between 128 and 256.

Thanks to the gamma curve that gets applied after linear processing, the end result is that there's no data in the last stop. It therefore gets rendered as pure white -- and, thus, a loss of a stop.

You can do the exact same thing yourself, assuming your RAW processing software is capable of linear exposure adjustments.

It's potentially useful in scenes with low dynamic range, or in scenes where you don't care about highlights but do care about shadows.

The common term amongst photographers who do that sort of thing is, "ETTR." I generally strongly caution against doing that, as it's very easy to blow out the highlights, and there's so much wonderful and delicate color to be found in the highlights that is so easy to clobber. But there are certainly situations in which it can be useful.

In general, the meter in most cameras underexposes the linear data by one to two stops, and the processing pipeline applies an equal and opposite amount of digital overexposure to compensate. This is generally a very good thing, because sensors clip so readily and so unforgivingly and modern sensors have so little noise. But, yes, if you're very careful, you can make use of that "extra" headroom.

Cheers,

b&

559
Canon General / Re: Don't order camera gear from Dell.com online store
« on: January 18, 2013, 06:06:52 PM »
I can easily imagine the guy taking all the pictures of the unboxing. I did the exact same thing when I got my 400 f/2.8, and something similar when I got my iPF8100.

It's pretty clear that there's some sort of fraud going on. While it's possible it's the irate customer is the one committing the fraud, I have no trouble believing that the distribution center got compromised, that several boxes had cameras replaced with flooring, and they do so little business that the replacement was sent from the same compromised batch (or that an employee was doing the replacement on the fly, and the same one handled both orders). One would think that Dell would have ordered an immediate inspection of the inventory, including sampling of that particular lot, but it's not at all hard to imagine the sort of corporate incompetence that would result in that not happening. Especially if order fulfillment was (as Dell admitted) outsourced to the lowest bidder.

I trust B&H, and they'll continue to get my business. Dell...well, everything about Dell that I've ever dealt with has been about nothing but the lowest bidder and driving down prices. I value my money too much to waste it by being cheap.

b&

560
There's no way to actually understand white balance without also understanding linear raw, which also gets into exposure and profiling and dynamic range (of both the scene and the camera) and lots more.

In uniform illumination, it's possible to get truly perfect exposure and white balance (and near-perfect color reproduction and contrast and all the rest) by knowing what to do with a quality profiling chart and all sorts of arcane software. In less tightly controlled situations, you can use similar techniques with a ColorChecker to get as close as the actual lighting in the scene permits.

If you're truly insane, Elle Stone outlines a procedure here:

http://www.freelists.org/post/argyllcms/Profile-input-white-not-mapping-to-output-white,68

that does that. I have a follow-up post in that thread with refinements and simplifications...but, alas, only the images and not my actual text made it to the archives. But Elle's process works, is logically equivalent to what I do, and the inspiration for what I do.

If, for whatever reason, shooting a color chart isn't an option, I have two suggestions, both of which work better than anything else I've seen discussed in this thread. (Both assume you're shooting RAW and setting white balance in post-processing.)

First, forget all the expensive white balance targets. None of them is as good as a styrofoam coffee cup.

Styrofoam is the right level of reflectance to get a noise-free sample without clipping, and it's spectrally flat. Indeed, with only two notable exceptions, nothing is better than styrofoam as a white balance target.

The first exception is Tyvek, which is 98% - 99% reflective and even more spectrally flat than styrofoam (which is 80% reflective or so). Your local office supply store sells Tyvek envelopes. Tyvek tends to be on the glossy side and prone to glare...which is good for knowing if you've got glare in the scene, but not so good for setting white balance. The other exception is Spectralon, which is ludicrously-refined PTFE (Teflon) with a glare-free surface. Expect to pay as much for a Spectralon target as for an L lens.

But back to the coffee cup. You can put it in your scene and get an even sampling of the illumination from every direction; you can then eyedropper from any part of the cup to get a white balance from light in that direction. Or, you can put it over your lens and get an average of all the color in the space where you're shooting, perfect for an in-camera custom white balance.

There are, of course, times when it isn't even practical to put a coffee cup in the scene. In such cases, it's still easy to get an excellent white balance. All you have to do is crank the saturation to maximum, fiddle with the white balance knobs until the colors in the scene look the least weird you can make them, and then return the saturation back to wherever you want it.

So, there're my two recommendations for white balance for the not-insane.

I'd also especially recommend avoiding using anything other than styrofoam (or Tyvek or Spectralon) for white balance, and especially avoiding "white" paper. With the exception of a small number of expensive fine art papers, paper is actually light yellow and has fluorescent blue dyes added to trick the eyes into thinking that the paper is whiter than white. It's a similar story with white clothing, with the added bonus that you usually get a non-trivial amount of skin showing through the fabric. Very few walls painted white are actually free of colorants, even though the base paints are generally themselves pretty good potential targets. Even those walls painted with truly white paint...well, it doesn't take long for them to take on colors from the environment.

I could go on, but this is already more information than is healthy....

Cheers,

b&

561
HDR - High Dynamic Range / Re: Best HDR Software?
« on: January 12, 2013, 08:40:46 AM »
The best way I've found is to think of it as a variation on the graduated neutral density filter theme.

Imagine the archetypal example, of a straight horizon, where you'd position the filter's transition on said horizon. Take two shots, one properly exposed for sky and the other for foreground. Put them as separate (aligned) layers in Photoshop (etc.). Add a mask to the upper layer, and use the gradient tool to create the same transition as the filter would have. The advantage, of course, is that you have complete control over how hard / soft the "filter" is, and over how many "stops" it is.

Now, imagine that the horizon isn't flat, but instead has a very prominent mountain peak sticking up out of it covering half the sky. If all you had was a filter, you'd be screwed. But, use this same technique and, rather than the gradient tool, use a very very large soft brush to create a custom-shaped mask, and you've got a custom-made GND filter just for your scene, something that would be insanely expensive if you tried to do it with tinted glass.

Extending the technique to even more complex scenes is easy to imagine from there. All you're doing is masking in and out the different exposures and thereby creating a virtual multi-step odd-shaped graduated neutral density filter.

Cheers,

b&

562
The (ungripped) 5DIII with the Shorty McForty and a wrist strap is the ultimate P&S. At first glance by a non-photographer, it just looks like a high-end P&S. The next time you're at a party, etc., and would otherwise reach for your iPhone, put it in green square mode and turn on live view. Maybe even pass it around the table. Nobody will realize that it's the reigning IQ / low light monster that it is until they see the prints.

If I were a wedding photographer, I'd definitely do some reception / dance shots like that, as an inconspicuous "guest" in the thick of things and just another random schmuck there with a camera.

Cheers,

b&

563
Lenses / Re: Prime Lens for 6D
« on: January 11, 2013, 04:32:43 PM »
Take the grip off the 5DIII, put on the Shorty McForty, and add a wrist strap, and you've got the ultimate point & shoot camera. You can even put it in green square mode, turn on live view, and pass it 'round the table. People are likely to mistrake it for a high-end P&S and unlikely to think that it's the reigning IQ / low light monster that it is.

Use the 50 f/1.4 or the 35 f/1.4 (or whatever) for when you're being a photographer. Use the Shorty McForty when you'd otherwise reach for your iPhone but want a bit more image quality.

Oh -- and the Shorty McForty makes an awesome body cap....

I recently had an opportunity to buy a mint-condition 50 f/1.0L for $4k. I love the thought...but, realistically, that lone extra stop just doesn't give enough over the already-superlative 50 f/1.4, especially with the high ISO capability of the 5DIII. The 50 f/1.4 is at least as good from f/1.4 on, so the only reason to get the f/1.0 is for faster than f/1.4...and, frankly, it's not all that great from f/1.0 through f/1.4. It's pretty mushy wide open, even at the plane of focus. I might consider the f/1.0 for a small premium over the f/1.2L, but I already think the f/1.2L is overpriced....

b&

564
Lenses / Re: Prime Lens for 6D
« on: January 02, 2013, 08:52:21 PM »
If nothing else, the Shorty McForty is an awesome body cap.

Pro tip: mount the Shorty McForty, take off the grip (if you have one), turn on Live View, and you've now got the ultimate party camera. Not many people will realize it's the IQ / low light monster it is, and just think you've got a high-end point-and-shoot. Put it in green square mode and hand it off to people and they're still not likely to realize it.

Of course, there's also an awful lot to be said for the 400 f/2.8, as well....

b&

565
EOS Bodies / Re: Canon DSLR Body Rumors for 2013
« on: January 01, 2013, 03:27:57 PM »
The big megapickle camera will not be branded as the 1DSx, for the exact same reason that, lo these many years ago, the compact Mac with a 68030 CPU was branded an SE/30 instead of, in keeping with the IIx and the IIcx, the SEx.

Cheers,

b&

566
Technical Support / Re: Printing large images - help needed!!
« on: December 29, 2012, 07:25:13 PM »
Wait a minute, I assume these programs for enlarging aren't free? Photoshop alone can't do the job? I was worried about the big one now. Maybe I'm out of my depth on this one.   :-\

I'll again suggest that you leave the up-sizing to the print shop. They've already paid for the software, and they've already figured out which of the various methods available with the various applications provide the best results for files with the types of detail in your images when sent to their printer.

If you up-size it yourself, they're stuck with whatever you give them. If you give them the file at its native size, they can then feed it to whatever will do the best job.

The same goes for color correction.

Again: make the file really shine on your computer, native resolution, and hand it over to the print shop as is, accompanied by a discussion about how you want the final print to look. (What's the expected viewing distance? Do you prefer "crunchy" sharpness, or would you rather preserve fine detail at the expense of a bit of softness? How much noise can you tolerate? That sort of thing.)

I'll also note that up to a doubling of linear resolution there really isn't any significant difference between Photoshop and anything else, and Photoshop does a good job (though not necessarily the best) at all reasonable ranges of extrapolation. Further, all modern printer drivers do their own extrapolation and interpolation, such that you often don't even want to do anything in Photoshop unless you have reason to suspect that the printer driver won't cut the mustard. And, lastly, sharpening (using whatever technique with whatever tool) is a critical part of scaling as well as printing, and it takes a bit of experience to know how to properly sharpen a file for print -- again, experience that your print shop has that you're paying them to have.

Cheers,

b&

567
Technical Support / Re: Printing large images - help needed!!
« on: December 29, 2012, 12:22:33 PM »
The single best piece of advice anybody possibly can give you is to find a quality print shop you trust, to prepare the files so they look their best on your computer, to hand them over to the print shop without mucking about with them yourself, and to communicate with the shop what it is you want.

Different enlargement techniques work best for different printers, wokflows, and more. Your print shop should know what works best for their setup; that's mostly what you're paying them for.

And printing that large is certainly possible with the cameras you describe. If you'd like a preview of what to expect, you can print a full-size crop on your own printer. First scale the image without resizing or interpolating to the final print dimensions; this will decrease the PPI. Then, crop or set the canvas size to your printer's paper size (obviously without scaling). Tape the print to the wall and stand as far back as your expected viewing distance. If you need to stick your nose in the print, you might have a problem, but, with those dimensions, at a few feet or so it should look gorgeous.

Your other option, much much more expensive for a one-off but cheaper in the long run, is to either buy the equipment to do it yourself or use one of those no-frills print shops like the ones in warehouse retail or Internet-only storefronts. You should expect to make many experimental prints before learning what does and doesn't work. If you're going to do a lot of this sort of thing, it's the only way to go...but not if you're only going to do a few every now and again.

Cheers,

b&

568
Lenses / Re: 8-15L cut it yourself gelatin
« on: December 21, 2012, 02:19:08 PM »
Supertelephotos take similar filters.

b&

569
Lenses / Re: The great battle: primes vs zooms
« on: December 17, 2012, 06:11:31 PM »
What am I shooting, and why?

If it's an event and I have no clue what I might be shooting, I want one or more zooms.

If it's one specific shot and I know exactly what I want, I probably want just one prime.

If it were, say, a football game, I'd want one of each: the 400 f/2.8 and the 70-200 f/2.8.

Since this is a money-is-no-object exercise, I went with the list of primes and the unstated assumption that I'd have an assistant to carry them and a half-dozen bodies so they could be switched out at a moment's notice.

But your list also missed out on some of the most important primes, like the TS-E 24, the MP-E 65, and the Great Whites....

Cheers,

b&

570
Lenses / Re: 2.8 vs F4
« on: December 17, 2012, 01:39:53 PM »
I'm more likely to leave out the "point." Thus, it's "eff two eight" and "eff four." I don't think I'd leave off the "point" for bigger apertures, though...it'd be, "eff five six," but "eff one point four." Not sure why.

b&

Pages: 1 ... 36 37 [38] 39 40 ... 65