Category Archives: Digital

Digital Photography (general topics)

Pentax K-5 and an Exposure “Oops”

We’ve all done it you know, when you snap a shot and haven’t checked your exposure, when the result you get is just a black image. Well, with the K-5, this isn’t necessarily a lost shot. In fact, the result can be quite interesting!

DxOMark just published their review of the Pentax k-5 sensor. The outcome of their tests placed the Sensor in a three way tie for fourth place with only medium format and full frame sensors ahead of it. The outcome was so good that DxOMark made this statement: “No need for suspense: this new 16.3 MP sensor is simply the best APS-C we have tested so far, sometimes able to compete even with very high-end full-frame cameras.” In particular, this applies to dynamic range and that is the crux of my post because dynamic range, which the K-5 is currently holding title as the best in the world, is how you can sometimes recover the “oops” I mentioned.

So, here’s the RAW image with ACR defaults (shot at f/4.5, ISO 80, and 1/200s shutter):

Now, this is with:

Exposure: +4.00

Recovery: 0

Fill Light: 0

Blacks: 5

Exposure: +4.00

Recovery: 0

Fill Light: 0

Blacks: 0

Exposure: +4.00

Recovery: 0

Fill Light: 100

Blacks: 0

Now with some noise reduction and sharpening:

Obviously I could do some more with the image, since it has been recovered, but I think the outcome speaks for itself. Other shots I took today in the same light were ISO 16000. This was ISO 80, meaning the exposure was almost 8 full stops off of what it should have been. That, folks, is wicked dynamic range.

The Pixel Peeping Trap

peepA lot of people can’t resist getting a nice, close, look at their images and it is so easy to do now with digital photography. Gone are the days, okay they’re almost gone, where we took our trusty 35mm film camera out to some event, snapped a roll or two of pictures, and then had them developed onto 5″ x 7″ prints. Unless you had real issues with hand shake, odds are that the prints would look nice and sharp. However, when you produced prints like this you had effectively shrunk the image to about 30% of its actual native size and that has the effect of sharpening the image quite a bit.

How did I arrive at 30% of actual size for the 5″ x 7″ print? Well, professional negative scanners will scan a negative at about 4000 dots per inch and, for a 35mm negative, that results in an image that’s about 5512 x 3674 in digital terms. The average photo printer will print between 240 to 300 dots per inch, with 300 being the desirable one, producing a print that is between 15″ x 22″ to 12″ x 18″ in size, depending on the printer. Photographers (and not just the pros) with good technique and proper equipment for the job will produce sharp images printed at these sizes which are, based on our equipment to print, 100% size in effect.

So, how does all this relate to pixel peeping on the monitor? Well, many people, myself included, will tell you that when you print you’ll want a minimum 240 dots per inch for a good quality result, anything less will start to look bad. Well, a monitor is substantially less dots per inch! The general LCD monitor for today’s computer is displaying at 96 dots per inch or 40% of the resolution of your printer output (at 240 dpi) and if you were to print your film image at that resolution it would be 38″ x 57″ and look like, well, crap if we held it in our hands and looked at it. Now, if you stepped back quite a distance and looked at it, it would start to look better, but odds are that you’re not sitting 15 feet back from your monitor and so viewing it at 100% on your display is a very innaccurate look at the image.

Now, most photo software, especially ones like Photoshop, Lightroom, and the like, do a very good job reinterpolating your image for display when you view it fit to screen. They are accounting for a lot of the differences in resolution and displaying according to the capabilities of the medium. According to some sites, and I haven’t specifically verified this, Adobe recommends that you view your images at about 25% unless you have specific retouching requirements. That’s probably a little too much to get a realistic impression, 30-40% is more in the ballpark based on our previous numbers and that is often what I’ll use. Why? Because then the display on screen is a much more realistic approximation of the final output to paper. If the image looks good, sharp (if sharp is what I want), and attractive at that size on screen, it will, 99% of the time, look that way when I print it.

Net effect, if you insist on evaluating the sharpness and quality of your images at 100% on screen, you’ll only come out of it unhappy. In fact, you may discard the images as a result of that and it is entirely possible that they would have been razor sharp when printed, so you lost out and maybe the rest of us did as well. This is a trap, avoid it.

Finally, it is obvious that not all of us are printing our images. I only print some, I have a very good printer for this, but I still only print what I intend to hang on my wall. Most of the time we intend to publish our pictures on the web in some capacity. If it’s Facebook, it’s going to be shrunk substantially (down to about 13% for me), but for other sites, such as Flickr, it will be resized several times and your original will be left intact. However, regardless of the site, you’ll notice that many of the professional photographers never publish the full-sized image. There are some reasons for this:

1. Protection against theft.

2. People will misjudge the image because they’ll pixel peep. Yep, they will, there are a lot of pixel peepers and they’ll want to see how the pro is and judge him or her based on that. Inaccurately, I might add.

I’m not a pro, but the above two issues still apply to any amateur. In fact, the first one may be greater for us because we probably don’t police the use of our images as strongly as we don’t have a vested financial interest. Some, less scrupulous companies know that. However, you’ll notice on my site that the longest size I upload is 1200 pixels, around 25% of the image size produced by my camera. I do that because it’s a good size for my site and, more importantly, it’s comparable in display to a printed version of the same image when someone views it on screen.

So, stop peeping at the pixels and just get out there and shoot!

RAW vs JPEG

As a digital SLR user, you have a couple of options when deciding what the camera is supposed to do when it captures the data off the sensor: RAW or JPEG. In regards to this, I land firmly in the RAW camp, here’s why:

1. JPEG is an 8 bit format, RAW is between 12 and 14 (though it can be less, it really depends on a variety of factors). Either way, RAW has more information available. That extra information allows for processing control that is lost if the camera makes the JPEG decisions for you. You can, for example, often recover blown highlights or underexposed regions, to a point. It’s not a panacea, but why lose that if you can avoid. A note for purists… You could do a lot of the same with film in the old days. Somebody with the right talent in a darkroom could often recover what mass-market machines could not.

2. RAW is more forgiving because all you’re really doing is capturing the sensor data, not telling the camera how to interpret it. So, for example, I leave my camera on auto white balance because it doesn’t matter if I save in RAW, the white balance isn’t applied yet anyways and I’ll set it in post processing. For JPEG, if you make a mistake on white balance, or leave it on auto, it’s going to be much harder to fix and the fix, by the nature of JPEG, will be destructive to the image.

3. RAW offers “non-destructive” editing options. By that I mean that when you process the RAW image data, the processing options you make are not baked into the original, you can easily undo them well after the fact, and so it is safe to save the original with your processing choices. JPEG, it’s baked in, so if you want to avoid lossy destruction of your original image, you have to save as a new file which is, by the nature of JPEG, going to lose additional information in the process. Successive saves of JPEG is a bit like copying from old VCR tapes, each copy gets worse.

4. RAW offers fine-grained control over the development choices. For example, with sharpening in Adobe Camera Raw, I have four control mechanisms on the sharpening that allow me substantially more precision than the camera options for JPEG development. Same as true for noise reduction, lens correction, and more. Net effect, it creates flexibility in your outcome that is lost with letting the camera do it.

5. Scaling a RAW image with higher bit depth is going to produce a better outcome than scaling a JPEG. JPEG has lost, already, bits of colour information and it has alredy discarded pixels in the compression process. Scale and you lose more.

6. Format conversion is more flexible in RAW. I can save to 16 bit TIFF and retain sensor data that cannot be saved if I go from JPEG to TIFF.

Now, all this wonderfulness of RAW isn’t without some disadvantages:

1. File size. RAW files are substantially larger than JPEG and so you can fit less of them on a card and less of them on your harddrive. On the other hand, disk and storage media are amongst the cheapest things you’ll buy in your photography hobby, so load up on them and forget about file size.

2. You have post-shooting work to do. You can’t just take the RAW file off the camera and slap it up online, you have develop it. So, this is probably the biggest reason that people avoid RAW. However, I shoot a lot of pictures and only spend the time on what I intend to publish. As you get practiced with it, and become more familiar with your gear and the software, it becomes substantially easier. I’m not a pro, and if I spend more than a couple of minutes on an image, I start to evaluate whether it is good enough to post. Having said that, I like to play with processing effects such as cross-processing and saturation/colour balance to create interesting outcomes and so those images take a little longer.

Now, you don’t have to jump into RAW shooting with both feet if you don’t want to. Depending on the camera, you can sometimes choose to shoot both. Pentax offers this option on all their current dSLRs and, I think, Canon is starting to do this too. Basically, for Pentax shooters, there’s a RAW button on the left side that, when pressed, will save both the RAW and the JPEG image. That lets you try processing yourself to see just how you stack up against the camera when it comes to it without risking a loss of an image because you’re not sure what to do. Also, with Pentax, you can set it do this mode all the time. So, as I said, risk free.

This, by the way, is a much longer version of a response I made on Photographr.info a few days ago. You should visit the site, a very friendly place to ask photography questions.

The Art Gallery of Ontario and Photography

The AGO forbids photography of art work in the galleries, though it does permit photography of architectural elements of the building itself. The premise for this is explained in one of their blog postings and, well, it’s a spurious claim to say the least. For example:

“While our visitors often point out that some other major art museums in the world allow photography of artwork, many of those collections are no longer subject to copyright restrictions, or are under different copyright rules than those in Canada.”

On the first point many, if not most, of the AGO works are also in the public domain so, in fact, are not subject to copyright law, Canadian or otherwise. So, right off the bat, the assertion is a false one. On the second point, copyright is multi-nationally enforced with common agreements between many countries. For the most part, the rules are very much aligned and that makes it pretty easy to deal with anywhere and Canada’s rules are not that complex or unusual.

“We didn’t set the copyright rules but we are required to respect them.”

They also didn’t read the rules, clearly, so I’m left being unsure of how they are thus able to respect them. For example, Part III, section 29 of the Copyright Act deals with fair use rights that are non-infringing. For further reference, which I also invite the AGO to examine, see the text of the Copyright Act itself.

If the AGO wants to ban photography, itself an art form that it showcases, then they may do so. It is, after all, within their rights. However, it’s particularly pathetic that they feel the need to lie about it in the process. That tells me they have no real excuse, more of a “we felt like it” attitude, and then used a poorly executed attempt to justify it with a law that doesn’t support them.

In the end, I might suggest to the AGO that they become aware of the Internet. It’s not all that hard to see if they have a case for their argument, Google turns up a link to the act as their first hit. People do check assertions every now and then, I know I did.

By the way, before I get asked… I do support banning flash photography in the galleries. While there is some dispute as to the effect the flash has on some of the art, it’s still a reasonable request. Why take a chance? Besides, the flash can disturb other viewers. Not allowing tripods, however, is a little absurd.

One final thought on this. It is permitted to sketch works in the AGO, using pencils only (go figure that one as well). The AGO is inconsistent: either the work cannot be copied or it can. Sketch, photography, or whatever, it’s a copy. They’re inconsistent on this front and that makes their policy even more absurd.

Pentax made one mistake

I really like Pentax cameras, I have for quite a long time going back into film and forward into digital. Pentax made a serious mistake though.

If you look at the Pentax dSLR line, especially the K10D, K20D, and the K7D, the camera is designed for the “semi-pro” shooter, clearly aimed smack at the SLR crowd. It doesn’t have fancy settings, otherwise known as scene modes, and they have provided innovative features such as a sensitivity priority, shutter and aperature priority, and raw+jpeg quick button. These are features that other dSLR makers are just starting to add. On top of that, the cameras maintain compatability with Pentax’s very solid historical lens line and can even mount the screw mount lenses and medium format lenses with easily obtained adapters.

So, they’ve clearly invested effort in creating a photographer’s camera, where did they screw up? Well, it’s not picture quality, but it’s related: long exposure noise reduction. Basically, the idea is that with long exposures, the chances of increased sensor related noise (such as hot pixels) increases and by taking a second exposure with the shutter closed, the camera can provide the proper image using dark frame subtraction. Sounds good on the surface, but the second exposure is for the same length of time as the initial exposure.

Okay, you might think that makes some sense, and it does, to a point. However, if you do night photography or astrophotography, you may be holding the shutter open for many minutes, potentially much longer. I tried some shots with a 5 minute exposure and that means the total time I waited before I could taken another was 10 minutes. At that point you start hunting for a way to turn this “feature” off. Good luck with that, you can’t turn it off on the K20D. You could on the K10D, but with the switch from CCD to CMOS, the option is gone, though I understand that you can sort-of turn it off on the K7D (still forced on bulb exposures). This has now cut, in half, the number of pictures you can take in dark settings with a long exposure.

So here, Pentax has designed and built a line of dSLR cameras that pretty much puts total control into your hands and then takes away the one thing sure to exclude anyone interested in low light or long exposure photography: an off switch for the noise reduction. This doesn’t make sense. I’m sure that there were likely technical issues that pushed them in the direction, but it doesn’t matter. People doing this sort of photography will do their own dark frame subtraction after the fact, it isn’t hard to accomplish and they will absolutely hate the inability to shut this function off.

This is a big mistake by Pentax. It makes me wonder if I made a mistake too. I like the idea of night shooting, I want to do star shots and long water exposures. Now I think it’s just going to be too painful to do. I don’t want to stand around for minutes on end while my camera takes a picture of the inside of my shutter. Pentax really needs to release a firmware upgrade to enable turning this thing off.