D300 - Initial observations

[simon_hickie1]

Hi Richard. I think it would have nailed the exposure spot on. The image below had the green bird box in the centre as the focusing point. Matrix metering again. The highlights are retained well. On the D80 in identical lighting, composition, metering mode and focusing point, the exposure was 2/3 stop more, resulting in blown highlights on the washing and on the bench - and I don't think it was a particularly 'difficult' scene.

 

Even with only just over a hundred test shots under my belt, I feel that the results are well worth the investment. The metering is such that I can be pretty confident that the camera is going to do its best to get both ends of the histogram in the right place. For really wide dynamic range landscapes, I'm sure I'll need to use ND grad filters, but I feel that the days of fiddling with exposure compensation for what seems like every shot are over.<div></div>

[richard_driscoll]

Very many thanks for all the help Simon.

 

I guess now I'll wait and see what the anticipated D90 does, though I have a nasty feeling it's going to be more

like the D40 than the D300 in the metering department.

[richard_driscoll]

Actually, I just looked at the histogram for "birdbox, washing and bench." It shows the shadows sitting nicely on

the left end of the histogram and a spike on the right which is almost certainly the sky. I'm sure the D40 would

have done exactly the same in this situation.

 

On the D80 would the shadows have moved up from the left hand end do you think? As you reckon it gave 2/3 stop

more then I guess the answer must be is yes. If so it is much worse than the D40.

[simon_hickie1]

I like the naming of the image! As it happens I have the D80 version here. Notice how the detail has been lost on the bench. The histogram is well spread, but clearly blows at the white end. We could also be seeing greater dynamic range effects at play here too.<div></div>

[photo5]

Actually this happens with D80 JPG's but in a RAW file you'll find plenty of detail in the bench. I disliked the D80's habit of "matrix spot metering". In other words, the focus point was biased higher than the rest of the image. If you shot a dark brown building with a bright sky above, the sky above would be blown to smithereens and the dark brown building would be light brown. A horrible defect in the metering. Yes, the D80 metering, like the D50 (which the D80 was designed to be an upgrade path from, as the D200 was designed as the upgrade path from the D70 series), was more point and shoot than SLR.

[jake_mecklenborg]

I've shot about 2,000 shots so far on the D300 and it is overexposing by 1/3 to 1ev in a lot of outdoor landscape situations, rarely underexposing if it's done it even once. More of an issue obviously if you're shooting jpeg versus RAW, and a big deal if you're shooting jpeg in a fast-changing photo j or sports situation. To me it's a lot easier to check the histogram and simply dial down the exposure compensation than it is to change metering mode and play that game. It's usually the case with sports where you're shooting jpeg instead of RAW that the light doesn't change /that/ much from one camera pointing to another on the field, where this is going to be a major issue. And even when the jpeg is slightly blown out, it's not so much of an issue with this camera as with the D70/80 because the image quality is just plain better.

 

The bottom line with the D300 is that a lot of things are just a *little* better which adds up to a lot, and a 10% better treatment of shadows and highlights is a huge difference as anyone who is experienced knows. All this jabber is a non-issue if you're a hobbyist shooting RAW all the time since you're going to have a helluva time getting beyond this thing's EV range in ordinary shooting.

[tri-x1]

In the D80 shot of the bench above, on my screen (recently calibrated) the bench has plenty of detail abd no blown highlights that I can see,

 

I have 5,000 clicks on my D300 and although I shoot raw most of the time, my jpegs haven't shown any tendency toward overexposure except when using very wide lenses on bright days. Then switching metering from matrix to averaging cures the problem.

[simon_hickie1]

Hi Wayne. On the D80 shot, areas of the bench max out at 255,255,255 as do parts of the white pillowcase. In the same places on the D300 shot, all colour values are less than 255,255,255.

[richard_driscoll]

Yes, the D80 shot is clearly blown in several places. I've been trying to reconcile the differences between the two but with little success. The D80 version has received 2/3 stop more exposure and yet the shadows are still similarly placed on the left of the histogram whereas I'd have expected them to have moved up to the right. I think the D80 midtones look brighter as one would expect. According to the EXIF data both are normal contrast (tone) so maybe as Simon says above it's down to increased dynamic range in the D300 with perhaps a difference in default tone curves.

[photo5]

Jake, Nikon overexposes whenever there is green in the photo. That's been true since I began noticing it with my N70, the first camera I owned with Matrix metering. My D300 meters perfectly in most every situation so far (with only a little more than 5,000 clicks on the body).

[velikaptan]

you can pick up brand new D200 from B&H for $999.00 its a great camera, i have both D200 & D300 not much diference

[simon_hickie1]

Interesting point Dave. I'll have to dig out the F70 to see how it behaves.