Dynamic range comparison btwn D90 and D300?

[larsbc]

Anyone done a comparison or can you point me to some published samples/test results?

I borrowed a friend's D90 to compare to my D300. I looked at some .nef files from the D90 tonight and I was a bit disappointed in how little highlight recovery I could do on the files. Unfortunately, I didn't shoot the two cameras side-by-side so I don't know for sure if my D300 could do better. At the moment, I have just my experience which is saying that it could.

Can anyone give me their experience regarding dynamic range and the D90 vs D300 with regard to raw files?

Thanks in advance.

larsbc

 

[_sem_]

Anyone done a comparison or can you point me to some published samples/test results?

I borrowed a friend's D90 to compare to my D300. I looked at some .nef files from the D90 tonight and I was a bit disappointed in how little highlight recovery I could do on the files. Unfortunately, I didn't shoot the two cameras side-by-side so I don't know for sure if my D300 could do better. At the moment, I have just my experience which is saying that it could.

Can anyone give me their experience regarding dynamic range and the D90 vs D300 with regard to raw files?

The DPR and DxO, IR etc tests say DR is practically about the same. Prehaps lossless compression could make a little difference, but nothing seriously considerable. There may be a difference in the metering algorithm however. And the amount of highlights recovery depends on the picture controls and WB. I mean, the metering is the same with all settings, but what you see in the histogram is closer to the right edge with vivid than with normal, and ADL can make odd differences for example. I guess there wouldn't be much difference if you set both to UniWB.

 

[gerryd04]

Can't help with comparisons but do know that my D90 is better with neutral PC, so it may have been influenced by picture control settings that your friend has set. Nikon does tend to use less aggressive curves with higher end bodies.

--
Gerry,
http://gerryd.smugmug.com/ discount code on homepage

'There's a fine line between fishing and just standing on the shore like an idiot.' - Steven Wright

 

[digby dart]

The D90 and D300 are among those cameras tested on DXO website, the results include dynamic range.

Remember though, besides the width of the range, where the range begins and ends is also very important for either highlight or shadow detail recovery. The dpreview tests often give these start end figures in their reviews.
--

 

[larsbc]

The D90 and D300 are among those cameras tested on DXO website, the results include dynamic range.

Remember though, besides the width of the range, where the range begins and ends is also very important for either highlight or shadow detail recovery. The dpreview tests often give these start end figures in their reviews.

I don't know why it didn't occur to me to check Dpreview's reviews...I think it's because I thought the D300 review came out before they started testing dynamic range in a consistent way. So thanks for the suggestion because I found what I was looking for.

When I compared the D90 test results to the D300, it showed that the D300 had about a .2 ev advantage in highlight range. I was surprised by that, because my impression from playing with the files yesterday was that it had a greater advantage than that.

Then I compared the D60 to the D90 (because I have a D60 I use for travel, and its dynamic range is really frustrating me, which is why I'm looking into a D90 as its replacement). The D60 has a wider range BUT quite a bit less highlight range, which is what I'm most concerned with.

As far as the camera's dynamic range hierarchy goes, it does reflect what I've found from my experience with these cameras. The D60 is distinctly worse than my D300. The D90 falls somewhere in between them.

Thanks for everyone's input.

larsbc

 

[spbStan]

In a study done by Nikonians measuring thousands of image files a chart was created:



The results are consistent with my personal experience and that of:
http://www.imaging-resource.com/IMCOMP/COMPS01.HTM

The D300 has a lot of things going for it but DR, and high ISO is not ones that it beats the D90, or course nothing else does in its class or in even in the Nikon line except for the exceptional D3x that beats all comers.
--
Stan
St Petersburg Russia

 

[zirtico]

Most DR comparisons are exaggerated. I find DPreview to have the most accurate figures. If you really want to measure it yourself, then put yourself in a room without windows or do this test at night. Fill up the cameras frame with a uniformly lit solid colour; doesn't have to be grey. Then keep changing shutter speed from 1/500th at f/3.5 at which point the frame should be black, and keep increasing until it's totally white. Then use few black and white frames and check the RGB values for a small set of pixels on all these frames. The range goes from when it's (0,0,1) or some co-ordinate with 1, up till 254. That should give you a fairly good representation of your camera's DR.

The difference between the DR between the D90 and D300 are probably in fact so small 0.1 or 0.2 eV that there could be that much difference among different D90s and D300s.

 

[larsbc]

In a study done by Nikonians measuring thousands of image files a chart was created:



The results are consistent with my personal experience and that of:
http://www.imaging-resource.com/IMCOMP/COMPS01.HTM

The D300 has a lot of things going for it but DR, and high ISO is not ones that it beats the D90, or course nothing else does in its class or in even in the Nikon line except for the exceptional D3x that beats all comers.

Thanks for that. Do they also have a chart that comes the highlight range and shadow range of the cameras? I'm mostly concerned with the highlight range.

larsbc

 

[digby dart]

Also in the old days of film we exposed for the highlights and developed for the shadows, if the dynamic range favours the shadow end, stooping down 1ev or more then process back up 1ev, just be careful the shadows don’t get muddy. Should have mentioned this in my original suggestion, I am besieged with the flue at the moment so the brain is on slow motion.
--

 

[rhlpetrus]

They are same and the best for APS-C in RAW (not necessarily for jpegs, that depends on settings). Check DxO Mark tests (choose cameras from drop down menu):

http://www.dxomark.com/index.php/en/Camera-Sensor

Anyone done a comparison or can you point me to some published samples/test results?

I borrowed a friend's D90 to compare to my D300. I looked at some .nef files from the D90 tonight and I was a bit disappointed in how little highlight recovery I could do on the files. Unfortunately, I didn't shoot the two cameras side-by-side so I don't know for sure if my D300 could do better. At the moment, I have just my experience which is saying that it could.

Can anyone give me their experience regarding dynamic range and the D90 vs D300 with regard to raw files?

Thanks in advance.

larsbc

--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)

 

[_sem_]

Thanks for that. Do they also have a chart that comes the highlight range and shadow range of the cameras? I'm mostly concerned with the highlight range.

I think the highlights and shadows headroom depend a lot on how the camera is set up in picture controls (brightness, contrast, saturation; ADL). The headrooms shown by DPR are the defaults, but you can modify the PC if you want more highlights and less shadows headroom (judging from what you see on the camera display & histogram). Also, set ISO to the fake 100 and the highlights headroom will almost vanish. It will vanish precisely if you use UniWB (the idea is that you know you've got none, so there's no guessing how much and you just watch the histogram). And, with extreme WB conditions the headroom tends to shrink at both ends (because channels blow at different points and there is less non-clipped overlapping range).

If you're after more DR at once, the main restriction is the output DR of display or prints. Notice that Nikon does not mess with HDR processing (compression of sensor DR range to output DR) and only allows you to shift a certain DR window up and down in raw conversion - but if you set contrast low the image will still be blown at one or both ends. So if you want to use the DR provided by the sensor using NX you may need to do pseudo-HDR (multiple conversions at different EC, then some sort of HDR blending). Alternatively, DxO and Adobe can process more input DR at once, but such results with Adobe tend to look rather flat. Actual multiple exposures will give you a better results than pseudo-HDR noisewise and regarding colours in the shadows.