dslr.ws

Jeff Kellor at DCResource stayed up until 3:00 am posting his review of the new Nikon D300. What does he say about it? Let’s find out.

First off, the battery life is excellent. You should get 1,000 shots on one charge, a huge improvement over the Nikon D200′s estimated 340 shots per charge, and a decent improvement over the Canon 40D’s 800 shots. Assuming you shoot more like a regular person than a CIPA test — for example, not using a flash — you should get even better results. This is assuming you’re not using live view, though.

He likes the magnesium allow build: “the D300 doesn’t let you forget that it’s an $1800 camera.” He also says that the grip is well-sized.

The live view sounds decent, aside from the slow contrast-based auto focus system and the lack of a live histogram:

So how does it look? Quite good, in my opinion. The view is bright, and incredibly sharp on the 3-inch LCD. Low light visibility isn’t great with the screen at its default brightness, but it gets better if you max it out (there’s a shortcut in live view mode to do quickly adjust the brightness). In both of the live view modes, you can zoom in on an area of the frame to make sure that your subject is properly focused. While you can preview white balance in live view mode, exposure is not simulated (the image always looks the same, regardless of the shutter speed or aperture). On a related note, there’s no live histogram available here, which some of the other live view D-SLRs offer.

Also, you can’t use live view and the self-timer function at the same time. That seems like a weird and arbitrary limitation, but there you have it.

One interesting bit to note is that the continuous shooting speed drops considerably if you’re shooting 14-bit RAW NEF files. In this review, shooting RAW, JPEGs, or RAW+JPEG ran at about 5.9 fps shooting speed (this is without the battery grip, which should bump that up to 8 fps.) However, shooting 14-bit RAW dropped that burst speed down to just 2.5 fps.

I’ve always been jealous of Nikon’s auto ISO function, which my Canon Digital Rebel XT doesn’t have, and which seems like a basic photography control that you want on a digital camera. The D300′s sounds easy and covers what you want it to do:

The D300 lets you set the ISO manually, ranging from 200 to 3200, with low (ISO 100) and high (ISO 6400) options available. It also has an auto ISO mode available. You set the slowest shutter speed and the highest sensitivity that you want the camera to use, and it does the rest.

On to ISO noise performance. The D300 gives a very respectable performance in the night test, even if it can’t quite hit the ridiculous ISO levels of it’s bigger brother, the Nikon D3:

You can see some minute amounts of noise at ISO 400 and 800, but it’s barely noticeable. Noise starts to pick up at ISO 1600, reducing your print size to small or medium, though you may be able to go larger if you shoot in RAW and use noise reduction software. At ISO 3200, the image has a kind of blotchy appearance from noise and presumably noise reduction. You may be able to squeeze out a small print at this sensitivity with a little help from NR software. There’s too much detail lost at ISO 6400 (H1.0) for the photo to be of much use.

NIght shots are hard to compare (since conditions are never consistent), but I’m thinking that the D300 did better than the Canon EOS-40D in this test. Both the D300 and the 40D wiped the floor with the Sony Alpha DSLR-A700, which isn’t so hot in low light at high sensitivities.

The studio ISO noise test they do also shows the D300 coming through with flying colors:

This is why I love digital SLRs — the D300 produces buttery-smooth, noise-free images through ISO 800. At ISO 1600 do we start to see some noise, but it’s barely there. At ISO 3200 the noise becomes noticeable, but it’s still quite low overall. That trend continues at ISO 6400, but the results are definitely still usable, especially if you’re shooting RAW. Just send those photos through noise reduction software and you’ll get yourself some nice 4 x 6 inch prints. Once again, I think the D300 beats both the 40D and the A700 in terms of high ISO performance — and I think you’ll agree if you compare the test photos.

Jeff mentions that the D300′s photos are soft out-of-the-camera, which isn’t unusual for a DSLR, especially Nikon. You still have a lot of detail in the files, though, and a little unsharp mask in Photoshop will bring them out. Or, just bump up the in-camera sharpening level, if you don’t like to do it on the computer.

The exposure seems to often be on the bright side, which is something to keep in mind:

Another issue I found is that the camera seemed to overexpose by 1/3 to 2/3 of a stop fairly often, so I quickly became familiar with the D300′s bracketing feature.

Or just shoot with -2/3 exposure compensation dialed in, and hope you’re not losing too much dynamic range at the high end of the histogram.

Interestingly, he reports that he couldn’t see any difference between the 12-bit and 14-bit RAW files:

One of the selling points of the D300 is to record 14-bit RAW files, which allows the camera to capture more color data. I took a photo of the test scene at both 12 and 14-bits, stared at them for awhile, and couldn’t see a difference. If you want to see for yourself, then feel free to download the 12-bit and 14-bit RAW files.

I haven’t had a chance to download the originals myself to look at, but I’ll be doing that soon, and would be interested to hear anyone else’s opinions about it; feel free to leave comments below.

In the end, he’s very pleased with the camera, calling it a strong competitor to the Canon 40D. A few negative points that he mentions include:

• Images are soft at default settings; camera tends to overexpose
• Battery grip (and proper battery configuration) required for 8 fps continuous shooting
• No exposure preview, live histogram, or self-timer in live view mode
• Sluggish contrast detect AF in tripod live view mode
• Continuous shooting burst rate drops to 2.5 fps when shooting 14-bit NEFs
• Flimsy door over memory card slot
• Remote capture software not included (Canon and Sony bundle theirs)
• Most expensive midrange D-SLR

There you have it. Be sure to check out the full review for all the details and a bunch of full resolution sample photos.

Rob Galbraith posted an interesting look at the “fixed” auto focus in the Canon 1D Mark III. This includes the sub-mirror fix that Canon offered, as well as firmware 1.1.3, which offers AF adjustments as well.

He reports that the sub-mirror fix is a big improvement, and the autofocus performance is no longer degraded in warm temperatures. Firmware 1.1.3 improves autofocus in all conditions, including better performance in bright light, and a better “feel” to the AF.

However, it’s not all rainbows and unicorns in 1DMkIII auto focus land:

But in our view, there’s more work to be done. In bright conditions, the EOS-1D Mark III still doesn’t match the EOS-1D Mark II N; at times the two are close, and at other times the EOS-1D Mark III’s autofocus performance is below what we consider acceptable. The EOS-1D Mark III can autofocus much better outdoors on a warm, sunny day now than it could at the time of its introduction, though this is a reflection of how much room for improvement there was. There are also specific situations — mostly involving shooting indoors or in dimmer outdoor conditions — where its AI Servo autofocus is actually superior to the EOS-1D Mark II N.

Looking at all of what we’ve shot with the two camera models in recent weeks, however, the EOS-1D Mark II N is the one that has delivered the most in-focus frames. The culprit that continues to cramp the EOS-1D Mark III’s autofocus style is bright sunlight.

Be sure to check out the full article, it’s incredibly in-depth and chock full of information. Rob flew out to Phoenix a number of times to work with Canon directly on this issue, and has a huge amount of technical and test information to pore over.

And, much like when they first mentioned this problem, they now have a ton of full resolution high-speed photo sequences (29 different ones) for insane depths of Canon 1D Mark III autofocus research. It’s a total of 1.4 GB of full resolution auto focus test sequences for your downloading and viewing enjoyment.

Seriously, go to Rob’s site for all the details. There are a lot of them.

John Harrington posted an interesting comparison between the Nikon D3 and the Canon 1Ds Mark III.

Some of the highlights include the D3 apparently blowing away the Canon on high ISO noise performance. The Nikon shows significantly less noise at 3200 ISO, and perfectly usable images at 12,600 ISO, which probably isn’t an ISO number you ever thought about before:

Look at the results! I would be comfortable delivering an image from a dark Congressional hearing, or a candle-lit church ceremony with this iso. The colors are more than fine, as is the sharpness. Moreover, the noise looks more like film grain than the noise of days gone by. The noise in the Canon above looks less like noise than it’s predecessor, but it looks more like noise than the Nikon does.

I don’t really agree with the analysis of the noise results at ISO 25,800, though:

Next up is the Nikon, at iso25,800. Yes folks, that’s not a typo. And, again, sorry, but Canon can’t go that high. Can you say “I can make an image in available darkness?!?!” Again, the colors are more than fine, as is the sharpness. Moreover, the noise here again looks more like film grain than the noise of days gone by.

However, the image shows some splotchy colors, kind of like an overly compressed JPEG. It’s still an amazing result for 25800 ISO, though, and I can’t even begin to imagine what kind of lighting conditions I could shoot with at that point.

He then moves on to a resolution comparison at 3200 ISO, and because of the lower noise of the D3 he actually prefers enlarging the lower resolution Nikon D3 image over what he gets from the 1DsMkIII:

Here’s the same Nikon iso3200 file, but it has been rezzed up to be the same size as the Canon file. No sharpening has been applied. It appears there is less noise, and it’s smoother overall. I know it’s not as sharp, but when you’re so deep into this photo, either the viewing distance of the viewer or the reproduction abilities of the output device will all but take care of this concern.

Here’s the same Nikon iso3200 file but it has been uprezzed and sharpened to my tastes. It appears equal to, or better than, the Canon image. In other words, by up-rezzing the file to the size of the Canon file, I can actually achieve a more pleasing final result than the native file results from the Canon.

Unfortunately, the comparison images for these are from a Barbie doll, which looks appropriately plasticky. I’d have preferred to see an image with much more detail for a resolution and detail comparison; a Barbie doll’s plastic surface is probably the best-case scenario for noise reduction and up-rezzing, since it’s supposed to look smooth and lacking in detail.

He also posts some interesting theories about the future, and what a Nikon D3X might bring to the game:

Following in the footsteps of the D1, and the D2, so too will there be a D3x. I’ve heard that it’s comparable to the Mark III as to file size. If the D3x has a comparable filesize to the Mark III, and the high ISO’s of the D3, it’ll trounce all over the Mark III, but I’ll not hold my breath for that “perfect storm” of capabilities. It’s coming in the Spring, certainly in time to get into the hands of photographers before the Olympics.

He also briefly discusses the Foveon sensor, which seems to have fallen out of the headlines lately, but has always seemed promising.

In the end, the reviewer says that he prefers the D3, if only “by a nose,” and stating that it’s only for the way he shoots:

For large commercial jobs, where I am thinking I want a higher native resolution, it’ll be the Canon I reach for. For low light situations, which are all too common, it will, without a doubt, be Nikon. In fact, with the noise issue out of the way, and most client deliverables needing to be down-rezzed anyway, I can see that the Nikon not only would be a better solution to speeding up my post-production process, but moreover, storing the RAW/DNG files will save me significant hard-drive space in my archives over the larger Canon files.

Personally, I’m hoping to see a similar comparison, but with a different subject than a Barbie doll. No offense, Barbie dear, but you’re not really a good candidate for detail and resolution comparisons. You need more acne and pock-marked skin. Ken might dump you, but you’ll be famous on camera websites throughout the world! It’s a fair trade.

The cameralabs.com website has posted a review of the new Nikon D300 digital SLR camera. Let’s jump right to the image quality comparison, in which they shot the same scene with a D300 (using a Nikkor DX 17-55mm f2.8 lens), and a Sony A700 (with the Sony DT 16-80mm lens.)

There’s a clear difference in the images when using their default settings and shooting JPEGs:

It’s immediately apparent from the crops below that the output from the D300 is noticeably softer than that of the Sony Alpha A700, when both cameras are set to their default JPEG settings. Of course there are differences in the optics used here, but with both bodies fitted with high quality zooms, the major difference is with in-camera processing – particularly in terms of sharpening and contrast. Boost the D300′s – or indeed reduce the A700′s – and the images can look very similar. It is interesting to note how their respective Auto White Balance settings have produced noticeably different results though with the D300 JPEG looking warmer than the A700s.

For the record, the A700 white balance looks more accurate to me — although you need to take that with a grain of salt since I wasn’t on site when they were shooting these and don’t know how it should look. But, having said that, the snow on the mountains in the D300 scene isn’t as accurately white as the snow in the A700′s photo.

Shooting RAW with the D300 gives significantly sharper output, although they aren’t clear on what settings they used to convert the RAW image:

The crop from the RAW file is noticeably sharper than the JPEG using the default settings, and of course there’s a lot of potential for further adjustment in Capture NX. As you’ll see in our results and gallery pages, the default JPEG output from the D300 can be a little soft, but this is in-line with most higher-end DSLRs which expect the owners to either tweak the in-camera settings or perform further adjustments later. Certainly the output from the D300′s RAW files converted using Capture NX can look excellent, and was much preferred to the default JPEG settings.

For noise performance, they compared the D300 with the A700 again, and also with a Canon 5D. The 5D isn’t really in the same league; with its full-frame sensor, the noise is much reduced and the detail is greater, and it outperforms the D300 and A700 fairly easily. So the review pretty much ignores the Canon 5D results as being too good to compare with the others (Canon fanboys can start eating their cake at this point), and compares the D300 with the A700. The result is that they give different results, but the review is hard-pressed to say that one is performing better than the other, stating instead that it’s really down to personal taste. Be sure to check out the actual image comparisons in the review to see how you feel about the noise performance.

In other features, the reviewer is very pleased with the 100% frame coverage of the Nikon D300′s viewfinder:

The D300 may have the same 0.94x magnification as its predecessor but Nikon’s upgraded the coverage from 95% to a full 100%. It’s a joy to frame your shot knowing what you see is exactly what you’re going to get, and it’s a feature normally reserved for top-of-the-range pro models. It’s also a nice step-up from the 95% coverage of the Canon EOS 40D and Sony Alpha A700.

They also find the autofocus to be full-featured and well-implemented:

With the D300 set to Continuous shooting and the nine-point dynamic area, it effectively tracked and photographed vehicles approaching face-on at 40kph, over a 100m stretch with 100% accuracy, using the DX 17-55mm f2.8 zoomed-into 55mm. Set to 51-point 3D-tracking, it coped with tennis players quickly moving around the frame, with the focus point visibly following them through the viewfinder with pretty decent accuracy. As always you’ll need a quick lens for the best success, but with the D300, Nikon’s really got fast-action covered.

On to the negative points. The LCD screen was hard to see at sharp angles or in direct sunlight, which limits the usefulness for live view. The contrast-based autofocus was slow, and the Panasonic L10, which also has contrast-based AF in live view, was much faster. The reviewer is very unhappy with the lack of a live histogram in live view, which other DSLRs have. Knowing that your exposure will be perfect seems like one of the big advantages to using live view, and it’s disappointing that the D300 doesn’t have that.

They also report that the anti-dust system wasn’t effective, but point out that most anti-dust systems aren’t.

In terms of Nikon users considering an upgrade, they point out that the resolution isn’t really much improved over the Nikon D80 or Nikon D200, and that if you upgrade, “you’d be upgrading for build, handling and features” rather than image quality. They also point out that it’s more expensive than rivals without giving better image quality; you’re paying 40-50% more than a Canon 40D or Sony A700, but getting roughly the same image quality, albeit with some more features and better ergonomics.

But they conclude that it’s an excellent camera, and receives the “highly recommended” label, with the caveats mentioned earlier:

But unless you’ve already bought into the Nikon system, we’d advise closely comparing it to the rivals mentioned above. The D300’s features may on the whole be a step-ahead of the competition, but you may not personally need them, in which case your money is better spent on a cheaper body with better glass.

All in all a very good review. Be sure to check it out for the full details and sample photos.