Field Reports | Have Camera Will Travel | Gary Gray

How to Up-Rez an Image

Wed, 19 Mar 2008 09:49:19 -0600

Field Report

by Gary Gray

As a follow up to my field report on the 300 DPI printing myth, I'm now going to show you a technique for enlarging your digital image files using Photoshop CS3. I've been using this technique since CS3 was released in 2007 and have found that I get better results than I do using Genuine Fractals (The Industry Standard.)

As a general rule, one should avoid adding pixels to their image for prints. Creating image data that doesn't exist can result in less than desirable image quality, however, sometimes you simply don't have enough image size to begin with and when a large print is called for, up-rezing is a necessary evil. Some programs will do it for you when you print by re-rezing the image size on the fly when sent to the printer. The problem I've had with this method is that it is hit-and-miss and wasting a print to find out can cost money. Up-rezing before you print is a better method I believe (for most of us.)

The first question I'm usually asked is "who taught you this technique?"

My answer is "nobody", I figured it out on my own through trial and error. Maybe somebody out there has a better technique. I haven't seen one that gives me better results.

I've read of and tried numerous different techniques in Photoshop the most common advice being dished out is to size up in either 1%, 5% or 10% increments. These techniques will give you varying results and more often than not you'll tend to lose a a little something or introduce pixel halos around the edges of the details. I've also found that Genuine Fractals, does a great job compared to the standard methods available in Photoshop, it too tends to create pixel halos. My trick will give you as much detail and result in virtually no haloing in your images. Don't believe me, try it yourself.

I've created an action in Photoshop and assigned it to one of my function keys. You may wish to try this approach as well, as it will allow you to quickly make enlargements of your file in increments. Essentially, we'll increase the file size in 25% increments and making a sharpening adjustment after each increment. For the purposes of this report, I'm using a full resolution 16 bit Tiff File shot with a Canon EOS 5D. I've found this procedure to be fairly consistent with just about any camera I've used and on original file sizes ranging from 1024 x 768 up to full size images from the Canon EOS 1DsMKII.

I'll begin with an image from my Canon 5D with dimensions of 2916 x 4374 pixels.

Lets start with creating the action.

1. load the image you wish to uprez into Photoshop, then click on the "create new action" icon at the bottom of the actions pallet to begin recording your predefined action. Give your action a a name. I labeled mine 25% Up-Rez, and assigned it to Function Key F3 + Command Key. When I hit Command+F3, the upsizing routine will run, giving me a 25% larger image.

2. Next we resize the image to be 25% larger.

Select Image>Image Size and make your adjustments in the dialogs as follows.

Resizeing the image to 125% of normal, constrained proportions and resample the image. We're using Bilinear resizing here, I've found this method is best for keeping details without introducing halos.

3. Next, we sharpen the image.

Select Filter>Sharpen>Smart Sharpen and make your adjustments in the dialogs as follows.

Check the Advanced button, Set Amount to 90%, Radius to 3.0 pixels, Remove Gaussian Blur and check the More Accurate box. Shouldn't need to touch anything else, but feel free to play around with these settings to explore different results. Click OK.

4. Click the Stop Recording button at the bottom of the Actions Pallet.

You've now created an action that will resize your image up by 25%. Run this action more than once to keep making your image larger in 25% increments. In the case of this image, we've started with a file that is 2916 x 4374 pixels, which when set to 300 dpi for printing (you can go down to 180 dpi and still get good results), you could print the original image at 12 x 18 at 243 dpi and get an excellent quality print. With a 25% uprez, you'll now have the ability to print size of roughly 15 x 22 inches at 240 dpi. Rez up by another 25% and you can now get a print size of roughly 20 x 30 inches at 227 dpi. I try not to resolve an image lower than 180 dpi in practice, but if you've read my earlier essay, you'll know that you can print at 180 dpi all day long without losing print quality. In the case of this image, I've now created a uprezzed image file by using a simple resizing routine that I can safely print up to 24 x 36 at 180 dpi with no significant loss of detail from the original image and in truth will look pretty darn good when viewed from just a few inches away. Once you've resized your image to your liking, feel free to adjust the sharpening and contrast to suit your liking. I normally apply this technique to images before I make serious color or sharpness adjustments. I've never had a problem using this approach.

So, what about Genuine Fractals? Does this technique give as good a result? I think so, and it will save you the cost of another software package.

Here are some cropped samples of the above master image which I've resized using my technique and using Genuine Fractals. Judge for yourself if this technique is worth while.

Both images were resized to 6835 x 4557 pixels. These are 100% crops of portions of each image for comparison. The Genuine Fractals image is the watermarked image on the left, the photoshop resized image using my technique is on the right.

Personally, I think my photoshop upsizing technique renders slightly more detail then Genuine Fractals and whilst this comparison doesn't show it very well, also reduces haloing.

Give it a try, you have nothing to lose.

The 300 DPI Print Myth

Thu, 31 Jan 2008 13:52:14 -0700

The 300 dpi Myth Field Report

by Gary Gray

What is the best resolution to print at? What's the largest good quality print I can make with my xxx megapixel camera?

Time and time again, these questions come up on the internet forums.

Every expert chimes in. Some know, some don't. Some are guessing. Some are responding with what they've read elsewhere.

Traditional wisdom says you print at the native resolution of your camera. In practice, people will often times resize their images for the mythical 300 dpi. I don't know where 300 dpi came from, but that's what you'll hear most often.

Is 300 dpi the best resolution to use with an inkjet printer? What is the lowest resolution you can use and still get a suitable print?

For my test image, I've taken a photograph from my catalog, made with a Canon 1DsMKII using a Canon EF 70-200mm f/4 L lens.

Nothing fancy, other than it provides a good range of detail from a fairly sharp lens on a high resolution body.

I've conducted a very simple and basic test using the above photograph. Using Photoshop PS3,

I've re-rezed the image to different DPI settings without sharpening the image and have maintained the 8 x 10 inch aspect ratio.

I've printed a 36 x 30 inch contact sheet from Adobe Lightroom, using no print sharpening and with all 9 images at their resolved DPI's, ranging from 75 dpi up to 416 dpi (native.)

The printer used is a HP Z3100, large format Ink-jet. It is state of the art and in perfect color calibration.

The paper is Premier Photo Luster Micropore, a very good paper, commonly used by large format printers for exhibition grade photographs.

Each file is a PSD formatted file from the original RAW image.

Below is a thunbnail view of the contact sheet containing 9 ea, 8x10 inch images which I'll be using for this discussion.

The resolution of each image is contained in the file name. As you can see, there are 9 images total at the following resolutions.

75 dpi, 100 dpi, 125 dpi, 150 dpi, 175 dpi, 200 dpi, 250 dpi, 300 dpi and 416 dpi (native resolution.)

The process is subjective. I take each photograph and analyze it under a controlled light source using the 416 dpi, native resolution image as a benchmark. My analysis is simply done using a loupe to examine the fine details of each shot and make subjective comparisons for each image. I'll then give my opinion as to what a minimal acceptable dpi would be for my particular setup.

Obviously, you can't see these prints. You'll have to take my word for it.

Looking at the 416 native resolution image, it is crisp and detail is very well resolved on the master 8x10 print. I'm concentrating on the building. Looking at the planks, window and the trees surrounding the building. The rocks and slag on the hillside below the building also give me a good reference point. The detail in the roof of the building is present but approaching saturation.

I began by comparing the 75 dpi image to the 416 dpi image. You never know, I may have underestimated the potential quality at 75 dpi, so I wanted to find out right away.

Results. At 75 dpi, the image, plain and simple, is not very good. To the naked eye, it is fuzzy, as if slightly out of focus and stair-stepping is evident in the sloped tin roof of the old mine building. Right off the bat, it is obvious that an image printed at 75 dpi isn't going to impress anybody. It reminds me of old 35mm enlargements done poorly. The scene is recognizable and the print is representative of the scene, but I wouldn't show it off.

Next up, 100 dpi. A noticeable improvement over 75 dpi, but there is still a visible difference from the 416 dpi print, even with the naked eye. With magnification, the details in the boards in the lower center of the frame are simply blurry but the stair-stepping is greatly reduced from that found in the 75 dpi image.

Next, 125 dpi. Things are looking better. Compared to the 100 dpi print, I'm now seeing a better contrast in small details. The large pine trees are showing much better, the tiny rocks and loose lumber is more crisp looking. Compared to the 416 dpi print, it's not looking too bad but not quite as sharp. The 125 dpi print still appears more soft. It's almost passable though. For a snapshot or something non-critical, I could use it.

Next 150 dpi. At first glance, compared with the 416 dpi print using the naked eye, I can't tell the two apart. Under magnification, you can see a difference. Most notably in the yellow trees behind the old mine building and particularly the yellow tree at the far left-center on the edge of the print. The fine detail in the roof of the building is slightly blurred but present on the lower res image. I could sharpen this image slightly and make it work just fine.

Next 175 dpi. Visibly, it looks identical to the 416 dpi print with the naked eye. Only under magnification can you see the difference and only if you study hard. Mostly, there is a loss of contrast in some fine edges on the lower resolution image. Rocks in the hillside, some of the loose planks, the edges between planks on the building. Very subtile and requiring a very close study. I could hang these two prints on a wall and you couldn't look at them with a naked eye and tell me which was which. With a little sharpening in post processing, you could print all day long at 175 dpi and get a good quality print.

Next 200 dpi. Visibly identical to the 416 dpi print. I really had to look hard with magnification to find a difference. The difference shows up mostly in fine details in shadows and only just a tiny bit. The shadows in the pine trees are where it can be found and only with very close and determined study under magnification. I'd use this print for just about anything if I had too. I feel comfortable with this resolution.

Next 250 dpi. With the naked eye, you can see no difference between this resolution and 416 dpi. Under magnification using my loupe, I can find nothing different in general detail. Tonal graduations are about the same, fine areas in the shadows are the same, edge contrast is identical. It would require a microscope to see a difference.

Next 300 dpi. Same as 250 dpi. There simply is no visible difference between the 416, 300 and 250 dpi prints, either with the naked eye nor with a loupe.

Conclusion...well, I already knew what the conclusion was going to be, as I've made this comparison many times in the past with different photographs. You can get a usable and good quality print down to 150 dpi. With good post processing, the average person isn't going to see what is missing and it would required a very trained eye with a magnifying glass to tell you anything different. In a perfect world, I wouldn't scale my photographs any lower than 175 dpi if I wanted to make prints for sale. This particular image would scale nicely to a 22 x 27 inch print at 150 dpi. With proper sharpening, it would look virtually identical at that size to the same image printed at 416 dpi. Nobody could tell the difference with the naked eye.

So, what does this mean? You may have seen different charts floating around on the internet showing the megapixels of the camera and the quality of the print you can expect at different sizes. Poor, Good, Excellent, these seem to be the most common reference. This means I'm going to give you a chart too. Here's my reference chart. I'm calling a 150 dpi print good, anything above excellent, anything below, poor. You'll have to judge for yourself the quality you get and what you need. What I'm saying here is that with a good quality camera and a good sharp lens, you can expect prints to look good at 150 dpi. Anything more is gravy.

So, if you have a 8 megapixel DSLR (Canon EOS 30D for example), you can reasonably expect a decent quality print at 150 dpi with a print size of about 16 x 24.

If you have a 10 megapixel DSLR (Canon EOS 450D for example), the largest decent quality print size you can expect will be roughly 17 x 26 at 150 dpi.

If you have a 12 megapixel DSLR (Nikon D300 for example), the largest decent quality print size you can expect will be roughly 19 x 28.5 at 150 dpi.

If you have a 16 megapixel DSLR (Canon EOS 1DsMKII for example), the largest decent quality print size you can expect will be roughly 22 x 33 at 150 dpi.

If you have a 21 megapixel DSLR (Canon EOS 1DsMKIII for example), the largest decent quality print size you can expect will be roughly 25 x 37 at 150 dpi.

Keep in mind, these are ball-park figures. You'll hear people tell you they've made 24 x 36 inch prints off their 8 megapixel cameras that look great. We'll maybe they look great, maybe they don't. A lot of this will be determined by the type of image you are printing. You may very well get a portrait shot or other low detail scene to print this large without a discernible loss of quality.

You may be able to do down to 125 dpi or maybe even 100 dpi with some types of images. A lot of this is going to be determined by what the expected viewing distance from the image is going to be. As for me, my expected viewing distance is a loupe. I personally make my decision by what I can see under magnification. If you can't see it with a loupe, you certainly won't see it with the naked eye, at any distance. You can of course re-rez your images to larger sizes using Photoshop. This is tricky business though. I'd recommend rezzing in stages and applying very mild sharpening at each stage. There are a number of tricks for stretching your print sizes, but that's a different matter, one I'll cover at a later date.

Other things that come in to play here are of course, the quality of your original image. Is it sharp and clearly in focus to begin with? What is the quality of the lens used? What is the quality of the printer used? What is the quality of the paper used? All questions that only you can answer when you make your prints or send your images out for printing.

One thing for certain though, using 300 dpi is a myth. You don't need to print at 300 dpi to get a good print and if anybody tells you so, they don't know what they are talking about. You don't even need to play it safe and print at 300 dpi. You're wasting your time and effort if you do so.

For me, this is my approach and it has served me well over the years. Your mileage may vary.

Happy Printing!

The Art of Technobabble

Wed, 16 Jan 2008 12:55:53 -0700

The Art of Technobabble (photographically speaking) and the New-Age Photographer.

by Gary Gray

I recently attended a Landscape Photography work-shop hosted by fairly renowned Colorado photographer, Kathleen Norris Cook. There were about a dozen or so other photographers in attendance but only two professionals, I being one,. Kathleen has worked primarily with Medium Format film throughout her career; however, most of the attendees including myself were using Digital SLR's. What struck me about this particular experience was the fact that she wasn't exceptionally knowledgeable about the use and application of the Digital Camera and spoke primarily from her perspective as a film photographer. Her knowledge and skill as a Landscape Photographer is without question, but she didn't seem prepared to assist her students with understanding the use of their particular equipment and how it is best used in Landscape Photography situations. Many students were struggling with their camera and as a result of their struggles, I found that many of them were gravitating to me for better understanding and knowledge of their particular camera during the three days we were in the field.

During our final critique session, Kathleen and I were having a brief discussion on the merits of Digital vs film and it was during this discussion that she made a well intended comment that I was a “gear-head.” Well, I know I'm a gear-head, but I've never really called myself a gear-head so I tried to understand her statement from a film photographers perspective. She's old-school You lug your film camera into the field, you load it with film, you study or pre-map your location and scene, you get set up, you wait for the light to be right and you take a half dozen shots. Afterwards, you spend days and days peering at the negatives on a light table, selecting the best shots and then painstakingly make the best possible enlargement prints you can get. You don't worry about megapixels, and sensor cleaning and frame rates and all the things us owners of Digital SLR's are accustomed to worrying about. From her perspective, I was too technically involved with photographic equipment to be concentrating on the actual art of photography. After thinking about it, I think she may have been right.

When I was much younger, my first camera was a Voightlander Vito III 35mm range-finder. It had a fixed 50mm lens, maximum shutter speed of 1/250th a second and was made of metal. The viewfinder was about the size of a pencil eraser. I never knew anything better existed, nor would I have cared if I did. I learned photography on my own and this is what I started with and I was happy.

I quickly discovered that I liked black & white. Color was the rage in the 60's, color television, color film...wow, color. Not for me. Black & white photographs had an edge to them, could convey mood and atmosphere that color couldn't approach.

Coming from 35mm SLR cameras, we had a slew of bodies to choose from and a slew of lenses to choose from, but there never seemed to be a big discussion centered on which brand was better. It was more a question of what type of film you liked to use. Kodachrome, Velvia, Provia, TMax, TriX, blah, blah, blah... Every film had its' own characteristic look, the color saturation, grain, acutance. This was our technobabble of the past. Myself, I simply went to the local drug store and picked up what ever film was on sale. I never got into the technobabble of the day and I was happy.

As the years moved past and I continued into adulthood, I remember with great enthusiasm the first “microprocessor” camera. The Canon AE-1. At the time of the release of the Canon AE-1, I was knee deep in my engineering studies and this new electronic marvel struck a chord in my photographic soul. We now had something to talk about other than the different types of film. We could now delve into the technical aspects of camera design and the benefits rendered thereof. Of course, our photographic skill wasn't improved by these modern manifestations of 35mm film photography, but living amongst other technically minded individuals, we could now have seriously technical discussions about cameras that were previously confined to the domain of “high-fidelity audio.” Our horizons had been broadened and we were happy.

Flash forward to 2008.

35mm film photography has been replaced by the march of technology. We now have the Digital SLR entering a stage of maturity and with a superior image quality to any film camera I've ever used. I didn't jump on the digital band-wagon immediately. I waited for a generation or two until the Digital SLR finally caught my attention. The last film camera I purchased was a Canon EOS-3, a semi-pro body with a microprocessor and custom function programability. Eye control focus system. Weather sealed body. It was the peak of technology for a 35mm SLR. I used it very little, but I was satisfied that I had about as good a camera as I'd ever need. It sat on a shelf for many years, my interest had waned, life taking other directions until this new digital camera technology permeated my thoughts sufficiently enough for me to venture forth into a brave new photographic world. In patience I watched as the first 1 megapixel cameras hit the market. The Kodak DCS series were the first I paid attention to. Basically, a Nikon film body modified for digital imaging with a 1.5 megapixel CCD sensor and a 2.6x crop factor. It was a monster.

These first incarnations of the Digital SLR were firmly based on the latest film 35mm SLR technology of the time (early-mid 1990's) and while most of us born before 1985 understood 35mm film photography, the new breed of photographer would grow up with a whole new perspective on the SLR camera. I didn't know it at the time, but the new technobabble was also introduced with this camera.

For certain, the digital SLR functions and behaves in most ways the same as the older film cameras. Heck, the newest SLR's are merely a manifestation of the natural technological advancement of all things analog towards a digital replacement. For the most part, the photographic rules of exposure, lighting & composition haven't changed as a result of the Digital Revolution. What has changed is the terminology. megapixels, CCD, CMOS, digital sensor, crop factor. Common everyday terminology for the modern photographer. Other phrases, once unheard of, are now part of our vocabulary. Post-processing, work-flow, pixel density just to name a few. With the new age of digital photography one must also drift away from the darkroom full of chemicals, enlargers and papers and towards the Desktop Computer and Inkjet printers. I see now why so many of my contemporaries have given up on photography. The monetary investment of acquiring the new knowledge and new equipment is a daunting prospect. Many old-timers simply can't come to grips with the computer age and still cling to their film cameras and negatives, and amazingly enough, are still making great photographs. What would take the old-timers days and weeks to accomplish can now be done in a matter of hours. I'm one of the old-timers who was lucky enough to cut my teeth in the digital world during its' infancy. I worked in Graphic Arts professionally for The Wall Street Journal as an Engineer, working on digital film scanners and image reproduction equipment. The late 70's and early 80's were an amazing time. I watched the film and digital world slowly blend into one coherent technology and I watched as the personal computer slowly replaced the dark-room. What I haven't come to grasp with, and perhaps this is a sign of old age, is the current vocabulary of the new-age photographer. With my tongue now firmly planted in my cheek, I'm going to attempt explain this phenomena from this old-timers perspective.

Pixels

A little speck of material on the digital sensor that captures light. The consumer market is driven more by this word than any other word in the digital vocabulary.

“How many megapixels does your camera have?” is the question I'm most often asked by the casual gawker.

“All of them.” is my usual response, usually coupled with a wry smile as I slowly turn away.

In the old days, we looked at the grain of the film and most technical arguments centered on the relative (or perceived) improvements in film technology. There is no pixel “grain.” It's simply a matter of how many pixels are crammed into the sensor of your camera. We also have the derivative terminology from the word Pixel. This includes megapixels, Pixel Sharpness, Pixel Density and distant cousins, Bayer or Foveon sensor (which is a theory on how to arrange and utilize those pixels.) Throw in a few good marketing monikers such as “X3”, or “MKIII” and you have what must be an advertising coup in the digital camera sales world. My new EOS-1DSMKIV 30 megapixel X4 Foveon Sensor camera that will shoot at 25 frames per second in 16 bit mode at ISO 120,000 will take a better photo of my 1 year old grandson's cake smeared face than my 8 month old 27 megapixel camera at 14 bits and ISO 60,000 at 16 frames per second. See how those pixels add up and make your camera sound better. Bigger numbers when coupled with alphanumerics and added to the camera name, you can't go wrong. Just pick a camera with the longest name and most/highest digits and of course...the most megapixels. I can't wait until the megapixel becomes obsolete and is replaced by the word “Giga-Pixel”, followed some 10 years later with “Tera-Pixel” and ultimately ending with “Google-Pixel.” Maybe, someday, we'll drop the word pixel all together. The Pixel will some day be obsolete, replaced by some bio-sensor with resolution down to a molecular level, capable of being grown in your basement or kitchen window. It can be maintained by imbedded nanobots. You'll never experience another dead pixel. A self healing camera sensor. Don't laugh, it will probably happen some day. I'll be dead of course, but today's photographer can look forward to the day when technology has left them behind and the new age photographers of that distant future will look upon them as from the "pixel era" and as the dinosaur. Becoming a dinosaur is a condition we should all look forward to experiencing. The alternative is to die young.

Crop Sensors

Now, this one is amazingly simple to me. Anyone who never picked up a film SLR though, is not going to understand this until they've become a more seasoned photographer. What I mean by being more seasoned is they have been subjected to enough ridicule from old-timers like me for not understanding their photographic heritage. The crop sensor is simply smaller than a standard 35mm film frame sized sensor. What makes it better is that it is cheaper than a full frame sensor. In my opinion, a smaller imaging surface is not better when it comes to photography, but try to explain that to the new-age photographer. The new age photographer sees it as an advantage. “The Crop Sensor Advantage”, another marketing term, grasped upon and clung too by the ill-advised and naïve. The new-age photographer uses his crop sensor advantage to be a better tele-photographer. They can now take a 500mm telephoto-zoom lens and magically turn it into a 800mm telephoto-zoom lens. This crop-sensor advantage allows them to take that pulitzer-prize winning photograph of a common sparrow from 30 more feet away. Never mind everything else, he's convinced that this advantage is the result of higher pixel density. It enables the "focal length challenged" photographer.

Focal Length Challenged is another one of those technobabble phrases associated with digital photography.

Basically it means "I can't afford a better telephoto lens." But, having a camera with higher pixel density will solve the problem. You thought I was through with the pixel thing didn't you? No chance. Everything in digital photography revolves around the pixel, don't forget this.

What the heck is “Pixel Density” anyway? Pixel density is cramming more pixels into a smaller surface area. Ten million pixels in a sensor with 2/3 the surface area is better than 12 million pixels on something larger. I haven't figured it out yet. I still go for the 12 million on a larger sensor. Photographically, it's technobabble, nothing more. The only advantage is to the bank account.

Cousin terminology related to Crop Sensors and Pixel Density are phrases such as Sensor Well, Signal-to-noise Ratios, Sensor Read Noise, Dynamic Range, Sensor Unity Gain, and the list continues to grow. People actually go to college, study engineering and write their Masters Thesis on subjects such as these. The new-age photographer must learn this terminology if they are to be conversant on the subject of Digital Cameras. One simply can't make a proper and informed purchase decision unless they have analyzed the Signal-to-noise Ratio of every sensor in every body on the market and calculated the effect of Sensor Well photon capture ratios in relationship to the dynamic range of the sensor at Unity Gain. If you don't understand the significance of these vital relationships, you simply can't make an informed buying decision when it comes to purchasing the very best Digital SLR and if you buy the wrong DSLR, you are really in for big trouble, because the true "New-Age Photographer" would never buy the wrong DSLR.

So, having pondered this entire digital revolution for about 10 years and having realized that what I learned from film still applies to digital photography and in ways that are no less important, is this. Return to the roots of photography. Understand what it takes to make a good photograph, not what it takes to make a good camera. The best photographs come from the best photographers, not from the best cameras. Let go of your inner consumer, forget the megapixels and signal-to-noise ratios. Don't worry if your camera will shoot at 5 or 10 frames per second or at ISO-64,000. Your results will be judged more on what you see than what the camera doesn't.

At this point in the game, there are only two basic things to worry about if you want to be a good photographer.

Can you be there when it happens?

Do you have the ability to see it when it happens?

If you can't answer yes to those two questions, you are wasting your time and money. Every digital SLR camera on the market is capable of making the image.

It's not about the camera, it's about you and your skill as a photographer. Never lose sight of this.

Just my opinion, but I'm sticking to it.

Nikon D300 Field Report

Wed, 09 Jan 2008 11:29:53 -0700

This is Hans, my 6 month old German Shepherd, taken with the Nikon D300 and a Nikon 80-400 VR Telephoto Zoom Lens.

Nikon D300 Evaluation

by Gary Gray (with a little help from Hans)

Anybody that knows me, knows I've been using Canon Digital SLR's. Reason being, I've always found them to have an image quality that was superior to Nikon across the board. Don't get me wrong, I've always liked Nikon cameras, but when it came to getting that little extra detail and lower noise, Canon has been a cut above and at a better price point for my professional needs since the dawn of the Digital SLR. All of this changed when Nikon released their latest Digital SLR's, beginning with the Nikon D300 and followed shortly thereafter by the D3, their first full-frame sensor body.

Here is the price / equipment comparison I'll be using as a reference for this report. This would be a common starting point for somebody purchasing either system. Both lenses offer a good focal range and image stabilization. You can find these sold as kits in this configuration.

Nikon D300, $1,799.99 MSRP, street price at the time of this writing $1,799,99 Body only (Amazon.com)

Nikon 18-200mm f/3.5-5.6 G ED-IF AF-S VR DX Zoom Nikkor lens, street price $679.00 (Amazon.com)

Total cost for a base Nikon D300 setup: $2,478.99

Canon 5D, $2,499,99 MSRP, street price at the time of this writing $2,109 Body only (Amazon.com)

Canon EF 24-105mm f/4 L IS USM Lens, street price $1,039.00 (Amazon.com)

Total cost for a base Canon 5D setup: $3,148.00

Both systems are very good setups and I have no problems using either in my work.

The Nikon D300 has been on the market since November, 2007. It sports a 12.3 megapixel, APS-C CMOS sensor. For those unfamiliar, the APS-C sensor, also referred to as the "cropped sensor" is a little smaller than the standard 35mm image format sensor found in its brother the D3 and the Canon 5D and 1DsMKIII bodies. The smaller sensor provides an economic advantage by being cheaper to manufacture, thus allowing the manufacturers to produce high-quality digital SLR's at a lower cost. This smaller sensor also provides other advantages and disadvantages to the photographer. In a nutshell, the cropped sensor reduces the field of view provided by any given lens by a factor of 1.5 for Nikon and 1.6 for Canon. So, a 100mm lens mounted on a D300 will give a field of view of a 150mm lens, vs a 100mm field of view if mounted on the D3 or Canon full frame bodies. To understand this effect on field of view, simply take an 8 x 10 photograph and cut off the outer 1/3 of the image to roughly the size of a 5 x 7 photo. The remaining image is what the sensor captures vs what the lens is actually providing the sensor in terms of image. Some folks believe that this is a telephoto advantage, but don't mistake the issue, there is no amplification or magnification, just a more narrow view, so the image appears to have a telephoto effect. On the other end of the focal view spectrum, the crop sensors do not provide as wide a field of view as the full-frame sensors.

As a landscape photographer, I've always found the wider field of view of more benefit to my type of work. Other factors such as depth of field come into play when one compares the full-frame and crop sensor bodies. If you adjust your field of view to be equal between the different bodies, you will always be using a shorter focal length lens on a crop sensor than on a full frame sensor. As a result of this type of adjustment, you'll find that the full frame sensor provides a more narrow depth of field for the same field of view for a given distance to your photographic subject. It can get confusing and I'm not going to waste a bunch of time here trying to explain it in depth. I will say though, that the photographic rules remain the same with any camera/lens combination, you simply have to account for the effects the smaller sensors generate when using Digital SLR's and standard lenses.

So with a crop sensor body you get a more narrow field of view with the benefit of having more pixels to reproduce your image than if you take a full-frame sensor with the same lens and then crop the image to get a closer view. If you are one of those people who can afford any type of lens, you can do just as well or better with a full-frame body by using a longer lens (at greater expense.) All that happy stuff aside, I've always believed that for a given budget, the crop sensor bodies tend to give a photographer a slight edge for shooting wildlife at a greater distance. But that edge came at a price. By cramming more pixels into a smaller space, these smaller sensors generate more noise in the image. There are trade-offs and this has always been my primary concern with using crop sensor bodies.

My benchmark for this discussion will be the Canon EOS-5D. The Canon 5D when released in August of 2005 it was the lowest noise digital SLR on the market and the first affordable full-frame sensor body and a whopping 12.7 megapixels. Full frame sensor bodies had been on the market before in the form of high-end professional grade cameras, but unless one had deep pockets, one simply couldn't afford them. With the 5D, images up to ISO 3200 were now possible, with noise levels that are substantially below any other Digital SLR at the time, including the Nikon cameras, and more importantly, affordable. A landmark camera. In the years following the release of the 5D, nothing came close to the low noise levels until the release of the EOS 1DMKIII in early 2007 and then the Nikon D3 in late 2007. At the time of this writing, the 5D was still on the market and at a price level competitive with the Nikon D300. Many photographers have to make a decision between buying either the Nikon or the Canon, so I'll give you an idea of how these two stack up to one-another, from my first hand experience.

On an equal footing, I'll start with the basics.

The Nikon D300 is a 2.5 year newer body technology wise. Nikon has crammed this camera with features and performance. From a handling standpoint, both the 5D and the D300 are about the same size and weight. Neither feels more or less comfortable in my hands or more or less uncomfortable strung over my shoulder or around my neck.

The very first thing I noticed with the Nikon D300 is the LCD display. The little screen on the back of the camera where the user interface and image is viewed. When you look at the D300 LCD next to the 5D LCD it is like night and day, in favor of the Nikon.

The Nikon LCD screen is much higher resolution (VGA) much brighter and renders colors much better. It's also larger, at 3 inches. Some photographers say this is a non-issue, but put them side by side and there is no doubt about which looks better. I find the LCD screen on the D300 is good enough to use for examining photographs in detail. On the other hand, the 5D's 2.5 inch LCD is only useful for making functional changes in the system menu and confirming you have a photograph. In bright sunlight, the 5D LCD screen is almost useless.

The 5D has a slightly larger view finder that covers about 96% of the frame, meaning there will be slightly more image than what you see in the viewfinder. Not a big issue, but a reality. Another reality for me is that I wear eye glasses and sometimes I can't see the entire viewfinder in the 5D as a result. It's that big, almost too big for my eyes.

The D300 viewfinder is slightly smaller, not a significant thing. It's big, bright and also gives the photographer a 100% view. What you see is what you get. I find it a little better for my photographic use as a result.

The Camera controls on the 5D are simple and intuitive. You can pick up a 5D and figure out how to use it in short order. More advanced options are available in the custom setting menu but most everything you'll need is at your finger-tips. The body isn't cluttered with buttons and switches and you run little risk of accidently changing a setting by inadvertent handling of the body. The main disappointment with the 5D body design seems to be the lack of a dedicated mirror-lockup button on the body. To do a mirror lockup one must either access this function through several button/dial maneuvers or program the function into the custom dial setting. The custom dial setting on the 5D works fine but if the camera self powers-down, you lose your last settings and it reverts to what the default setting was. Either way, getting to mirror lockup on the 5D was cumbersome and inconvenient.

The D300 on the other-hand is a maze of buttons and switches. I found myself having to practice several times to lean how the functions worked and where the proper buttons were to access them. Every thing is labeled, but I find the Nikon body to be less intuitive to operate. It takes more study and practice to master the controls. One advantage the D300 has over the 5D though is that you can change the mirror lockup function without having to wade through the camera settings. A single flip of a dial and you are there. This is a good thing.

The Canon 5D is not weather sealed. It is a rugged body and will take some abuse but I normally put it away if I'm outdoors and it begins to snow or rain. I don't want water leaking in through the seams or buttons and ruining my camera. Still, I've used it in low temperatures and wet conditions in the mountains and near the ocean, and using simple common sense about how and where I use it, I've never had an issue. To get better weather protection from a Canon camera, you must buy one of their 1 series cameras at a significantly higher cost.

The Nikon D300 has more protective sealing around the body. I don't think it is a fully weather-sealed body, but based on Nikon's description, the camera should be more resistant to water and the elements. I've read and heard of anecdotal evidence that the Nikons do indeed provide a better body for shooting in inclement weather, and I trust this is true.

Battery life

Both cameras utilize current Lithium Ion battery technology. Nikon claims that under normal shooting conditions one can expect about 1000 frames on a fully charged battery. From my testing and comparing it with the 5D, I would say they are about the same. I don't use the preview screen much and I allow my cameras to go to sleep when in use. I can't really see a difference at this point. Once I've had a chance to do some serious shooting, I'll report back with anything that I feel is out of the ordinary.

Lenses

It's been generally accepted that Canon has a little stronger lens lineup than Nikon. Canon makes a wide range of zoom and prime lenses to match just about any budget. Some of Canon's cheapest lenses perform very poorly and their more expensive "L" lenses are professional grade.

Nikon has a good lens lineup, but the distinction between their lower grade lenses and top glass is a little more blurred I think. Nikon's overall selection is a little thinner and their prices on better lenses tend to be a bit higher overall. What you save on the cost of a body may be lost acquiring glass.

It should be noted, second party lenses are made for both brands of camera by several companies, most notably, Sigma and Tamron. With few exceptions, I've found second party lenses to be under-performers from a professional photographic standpoint. For a budget conscious person who wants a less expensive alternative, they are an valid option though.

Overall, regarding lens selections, I feel that Canon has a better price/performance ratio than Nikon, as well as better selection of choices.

Things that aren't equal.

The Nikon D300 has newer features than the Canon EOS 5D. Many of these newer features can be found on the newer Canon bodies however and I fully expect the replacement to the 5D will also have an improved feature set. It's not really fair comparing a 2.5 year old camera's features to a brand new model. Still, I'll point out the differences.

Sensor cleaning. Canon 5D = NO, Nikon D300=Yes.

The jury is still out on how effective built in sensor cleaning is on Digital SLR's. We're sort of in version 1.0 of the technology. A common gripe of digital SLR owners over the years has been their propensity for collecting debris on the photographic sensors. Some claim the 5D is one of the worst offenders in this regard. My experience with the 5D is that it is no more prone to sensor debris than any other DSLR I've used, including the 1DsMKII full frame body. I think the 5D was shipped with excessive debris already in the camera and once it was cleaned out, I've had very few problems. Still, it is there and I have to look at my images and clean out the little dust specs from any image I want to sell. This adds time and effort to post processing.

The Nikon D300 has a vibrating sensor that will function when you turn the camera on or off, the idea being that it will loosen up dirt particles and shake them off the sensor. In theory, this should result in fewer sensor cleanings and less debris showing up in photographs. With limited experience using the D300, all I can say is the the camera came out of the box clean and in the time since getting it I've changed lenses numerous times and I've yet to find any dirt on the sensor showing up in my test images. It looks promising.

Live View

The newer Canon bodies are including Live-View but the Canon 5D doesn't have this feature. Live view allows you to see the scene in real time on the camera's LCD display. The D300 has this feature and it works in two different modes, hand held and tripod. From my initial testing, it seems to work very well and based on what I've heard and read from other sources, the Nikon Live view is functionally superior to Canon's present version.

Custom Lens Settings

With the 5D, you put a lens on the camera, you take your photograph. The camera communicates with the lens for auto-foccus and exposure and does nothing more.

With the Nikon D300, you can make fine adjustments to lenses that have CPU's plus compensating adjustments for slight variations of focus for a set of lenses. In addition, you can enter a number of lens into the Nikon storage for non-CPU lenses, such as older manual focus lenses. Nikon is smart here. What this allows is for older vintage Nikon lenses to communicate with the camera, even though those older lenses have no electronics. Nikon has an aperture ring on the back of every lens they've made and it will communicate mechanically with the Nikon body. You can still use these old lenses with the camera in manual or aperture priority mode and have the metering still functional. A very handy and forward looking design. With Canon bodies, you pretty much have to use an EOS lens if you want to use the functions of the camera without having to resort to a totally manual exposure measurement. The Canon bodies will still use other vintages and brands of lenses via special adapters, but Canon has more or less forsaken their older pre-1988 lenses. Personally, I find the Nikon approach much more enjoyable and functional. I can buy used Nikon lenses from the 60's and 70's and still use them on my brand new camera. Some of these older lenses are gems too. So the lens advantage isn't fully in Canon's corner at all, Nikons will work with a much wider variety of older lenses without having to resort to custom adapters.

Auto Focus

The Canon EOS-5D auto-focus system is a 9 point system with 6 invisible supplemental AF points. This system also includes separate modes for single shot focusing and AI servo tracking. Generally speaking, I've found the 5D's auto-focus system adequate for most needs. When shooting high speed action and fast moving subjects, it struggles a bit. By today's standards, this is run of the mill performance.

The Nikon D300 in contrast, uses a 51 point auto-focus with 3D focus tracking with three dynamic auto-focus modes. In my initial testing on my favorite subject, Hans the German Shepherd, I've found the D300 capable of tracking and staying locked on to this fast moving pooch in every test I've conducted. This is state-of-the-art auto-focusing on this body and by comparison, it makes the 5D look inadequate. With approximately 500 test shots, I've yet to take an out of focus shot using any mode on the D300 in any situation I've encountered. Simply great performance.

Frame Rate

Canon EOS 5D shoots at a maximum of 3 frames per second. This is not blazing fast. The buffer size in the camera though is fairly deep for a camera this old and will allow you to shoot about 16-17 continuous frames before bogging down. This isn't the camera for sports shooters, it simply won't keep up with the action. For general purpose photographic needs, it gets by just fine.

The D300 will shoot at a base mode of 6 frames per second and boosts that speed up to 8 frames per second with the addition of the optional MB-D10 Multi-power battery pack. This is fast. The only camera faster is the EOS 1DMKIII at 10 frames per second at almost 3 times the cost. You can use this body to shoot sports all day long and it will get the job done. I'd also venture an educated guess and suggest the auto-focus system on the D300 even outperforms the 1DMKIII, if it doesn't it is going to be very close because I feel it is better than the Canon 1DsMKII, which is no slouch.

One other item of note here regarding shutter lag (time it takes from pressing the shutter button to taking the picture), the 5D is a little bit sluggish compared to the D300. In my day-to-day use of the 5D, I've often found this slightly slower shutter lag to be an issue, mostly when I'm trying to time a shot for just the right moment. You are more likely to get a shot after the moment with the 5D than the D300. Not that the 5D is horrible, but just noticeably slower compared to many other cameras on the market.

Flash Sync

Canon EOS-5D, 1/200 second max.

Nikon D300, 1/250 second max.

Not much a difference at first glance, but considering the 5D is still more expensive and used by many professionals, 1/200th a second flash sync is sub-par by any standard. Why Canon decided to cripple this aspect of the 5D's performance is a mystery, even Canon's less expensive models out-perform it. Silly.

Now let us look at the 'big picture" (pun intended)

The million dollar question is. How good are the images and how well do they compare to one another? Well, I took a few test shots to get a feel. My first thoughts before doing a pixel peeping comparison was that both cameras have 12.x megapixel sensors, therefore, both camera should produce roughly the same resolution and comparable image quality. After doing a few test shots, I discovered that my assumptions are correct, for the most part. The ability to produce a high resolution, detailed and low noise image from either camera is almost identical up to ISO-1000. You wouldn't be able to tell either camera apart by looking at a 24 x 36 inch print.

Now, that said, I find that the Canon EOS-5D still outperforms the Nikon D300 above ISO-1000. The Canon produces cleaner images with more detail from ISO-1000 to ISO-3200 and there is no question about it. Below you'll find a few 100% crops showing the performance using the two sample setups. Keep in mind, the differences between the two test lenses can account for some of the difference in detail, but you can see that both lenses are doing a good job of resolving detail at 12 megpixels, and that's good news for Nikon owners, as the 18-200 VR holds it's own against the more expensive Canon L.

As you can see here, at lower ISO, there is no difference in noise level or detail. Both are excellent. These are 100% crops of JPGS converted from RAW. Aperture priority mode, F/11, matrix metering, bad light, tripod mounted, based on the following scene. Both cameras were in autofocus mode, using the center focus point, aimed at the UL emblem on the dog-chewed dremel tool.

The Hans Taste Test Image

The above image was a slap together scene I used to analyze these two cameras. I call it my dog taste test. Every item in the scene above has been the victim of my German Shepherd Puppy, Hans. Thank goodness he hasn't found my cameras yet. I used available light (tungsten) for the most part. The test images were equalized for white-balance and slight changes made for exposure for max highlight and shadow without clipping, using Adobe Lightroom. No changes were made to color saturation or sharpness. The differences in contrast I attribute to the different lenses. You'll also notice a difference in depth-of-field, with the 5D being more shallow. This is because to create these images with the same field of view, I had to increase the focal length of the 5D by a factor of 1.5. These objects are about 3 feet from the camera and the focal length is roughly 52mm on the D300 and 80mm on the 5D.

Nikon D300 at ISO-200 (base ISO)

Canon 5D at ISO-200

Nikon D300 at ISO 400

Canon 5D at ISO 400

Nikon D300 at ISO 800

Canon 5D at ISO 800

Here at ISO 800, this is where the similarity ends. The D300 is still doing a great job compared to the renowned 5D image quality. You'll never see the difference in print up to this point.

Here are ISO 1600 shots. You can see the Nikon is starting to lose it a little compared to the 5D. I'm using a low setting for noise reduction on the D300, trying to keep the noise down a little without losing much detail. It's not that effective in my opinion and further noise reduction may smooth things out but at what loss of detail? Even with noise reduction on, noise is beginning to interfere with the image and very fine details are beginning to break apart.

Nikon D300 at ISO 1600

Canon 5D at ISO 1600

Nikon D300 at ISO 3200

Canon 5D at ISO 3200

Again, ISO 3200 and the difference is really starting to show. Nikon Detail is falling apart and the Canon is still going strong. This is why the 5D is such a good photographic tool. It simply holds image throughout its range of function all the way to ISO 3200. Does this mean the D300 can't be used at the higher ISO's? Of course not. When looking at a full size image, it still looks very good to the naked eye at normal size. You can get excellent prints at ISO 1600 or higher from this camera, it just can't keep up with the full frame sensor on the 5D at this point, and this is one reason many photographers choose full frame over crop sensor bodies. They hold better detail with lower noise at high ISO's.

So, what does all this detail at high ISO mean for me and you. For me and my style of photography, nothing. I hardly ever shoot above ISO-800 and I have the ability to do so with excellent results using the 5D. I've taken thousands of photographs in many different situations and less than .05 % of those shots are above ISO 800. It just simply isn't required for most photographic situations I encounter as a professional photographer. Now, if you plan on doing a lot of low light shooting such as stages and music concerts in poorly lit venues, or if you need the ability to shoot in poorly lit rooms without a flash using only available light all the time, you may find yourself needing and using ISO-1600 or ISO-3200 and this may become a bigger factor. For me and 99% of the general public, it's a non issue and I strongly suggest you let go of your inner tech-no geek and not weigh the issue too heavily when choosing a camera body. It's a gear-head issue for the most part, not a photographic and artistic priority.

So, lets talk about a few intangibles now. You'll notice that I didn't mention picture styles and default setting for JPG shooting. Reason being is that I never shoot JPG's. I shoot everything in RAW and if you are serious about getting the best quality from your photographs, you'll shoot RAW as well. Still, there are some folks that insist that JPG is good enough. Well, I don't think you'll find much difference between the cameras shooting JPG files. The default settings and picture styles will give you enough variation and adjustment, you'll find some setup that meets your needs. But, if you are like me and like shooting raw, there are a few things I've discovered comparing the two cameras.

The Nikon D300 has a few additional exposure features over the Canon 5D. One is called SRS for Scene Recognition System, supposedly giving it a more intelligent exposure system than previous DSLR's. The D300 also has a highlight/shadow compensation capability which helps prevent blowing out highlights and smearing shadows. Canon's new cameras have a similar dynamic exposure feature. What I've found in my comparison is the D300 produces a better looking raw file out of the camera. The color seems better controlled on the D300 vs the color from the 5D. Exposures are very accurate, even in snow scenes. Automatic white balance seems to be very good as well, which brings up another plus in favor of the D300. The D300 has the ability to set a custom white balance on the fly, without having to take a reference photo. It still takes a reference exposure, but this exposure is loaded directly into the camera's storage and can be used immediately, plus you can take more than one reference shot and store it on the fly. A useful feature, something the Canon 5D doesn't do. With the 5D, you still have to take a photograph and tell the camera to reference that photograph as a white balance source; extra steps and extra work. In general I find the raw files from the D300 to be more punchy and better exposed than with the 5D. A very impressive output by any measure.

Other features found on the D300 that you may find helpful.

The D300 supports HDMI output for video hookup to a television. I've tested it on my 1080i High Def television and it works great. You'll have to buy a cable though, the camera doesn't come with one.

A really cool feature in the D300 is the ability to set the camera up for interval photography. You can program it to take shots unattended at a certain start time with definable intervals and for number of frames to take. Couple it with a infra-red sensor and you could place this camera in a remote location and have it take pictures of wildlife that strolled into the frame, or even more interesting, time-lapse photography. Set it up to photograph a budding flower over time. I really like this feature.

Conclusion

The Nikon D300 is a fantastic photographic tool. In almost every instance, it's performance, handling and agility far exceeds that of the Canon 5D. The Image quality is without a doubt every bit as good. The only area and I do mean ONLY, where the 5D out-performs the D300 is with noise above ISO-1000. In every other regard, the D300 is a superior tool. When I look at my kit and decide which camera to use, in most situations I will be picking up the D300 and leaving the 5D behind, there is no question about it. When you couple the fact that the 5D is priced several hundred dollars more than the D300, not including the lens, I can't see a justification for purchasing a new 5D at this time. When Canon releases a replacement body for the 5D, I'm certain that the gap in features between the two will lessen. I'm also quite sure that the price point for the 5D MKII or what ever it's named will also go back up to around $3000. Maybe Canon will get a clue and stop crippling their higher end cameras. Nikon has taken the lead here and there is no doubt about it. If you are looking for a great camera and lens combo at a very reasonable cost, you can not go wrong with the Nikon D300 with a 18-200 VR kit. It's a winner in every regard.

To learn more about the Nikon D300, click here.