Quick and Dirty Comparison of Dynamic Range of Nikon D80 vs. D700

[see_r]

<p>I have a Nikon D80 and have been reading wonderful things and seeing excellent examples of dynamic range with the D700. So I brought my D80 (and memory chip) into the camera store and ran a "quick and dirty" comparison really only intended for comparing the dynamic ranges of these two DSLRs. I've attached my results. Here's what I did with both cameras:<br />1) put on my AF Nikkor 50mm 1:1.8D; 2) set to 200 ISO f4.5 at 1/100 s; 3) took two shots (pretty much the same shot with each camera), the first of which had a highlight that was about 4 1/2 stops overexposed (in the first images, left side, out the window on the street) and the second of which had a shadow that was 4 1/2 stops underexposed (in the second images, right side, middle dark wall); 4) opened up the raw file, made no changes and pasted into a single document; 5) res'ed down and jpg to upload here.<br />I know the D700 is a great camera and perhaps I have missed something in the haste of my experiment. But I'm not sure I see an improvement in dynamic range from D80 to D700. If anything, I see the opposite--at least when evaluating the extreme highlights and shadows.<br />Please let me know your thoughts.</p><div></div>

[Karim Ghantous]

<p>I think it might be interesting to bring up the exposure in the under-exposed images and see what levels of noise and detail you get. That will be the real test. (Also do the opposite for the highlights.)</p>

[ben_goren]

<p>Chip,</p>

 

<p>A <em>much</em> more meaningful comparison would be to shoot the same scene,

<em>properly exposed</em> with both cameras, and do real-world post-processing

<em>appropriate to each camera</em> before downsampling and comparing.</p>

 

<p>It’s a common mistrake to think that a fair comparison means using two different

cameras with identical settings. But, in the real world, it makes much more sense to independently

strive for the utmost from each camera, and then compare the results. After all, who cares if you

have to set the frobnotz on one camera to 11 while the other only needs it set to 42? It’s the

end result that matters, not the dial settings you used to get there.</p>

 

<p>Cheers,</p>

 

<p>b&</p>

[johnw63]

<p>I don't know. I think a simple test like this has merit. If I spend THAT much for a new camera, I don't want to take the shot, see little improvement, but THEN, if I spend a lot of time flogging the shots in post production I can pull the extra range I paid for.</p>

<p> </p>

[see_r]

<p>Thanks for your responses. Ben, yes I agree--it would be nice to borrow a D700 and take it on a trip, shooting it part of the time and the D80 part of the time, and then do the best I can with each respective set of images. Of course this would mean me needing to borrow the D700 for awhile in advance of said trip because I would need to do the learning curve.<br />and actually, I just did a follow up, going back to the NEFs and setting the exposure compensation for each to -4 to evaluate the blown out streets....I do see improvement with the D700</p>

[see_r]

<p>Oh, and here's what I got at minus 4 exposure compensation (the d700 image is the larger and the d80 the smaller, obviously)</p><div></div>

[elliot1]

<p>Although camera features have changed dramatically from the D80 to the D700, dynamic range is not one of them (it has increased only slightly). The minor differences you are seeing can easily be balanced in post processing. According to the DXO Mark site, the D80 has a dynamic range ranking of 11.2 and the D700 is rated at 12.2. (The D3x is rated at 13.7). I don't believe this is a significant difference and my own real world testing has yielded similar results to yours. Yet overall you will get improved pictures under most shooting situations with the D700.</p>

[dbcooper]

<p>I agree with Elliott. I have a D700 and kept my D200, which should be pretty much similar to a D80 in terms of IQ. The dynamic range difference isn't earth-shaking - it's about a stop overall, meaning the D700 is 1/2 stop better before clipping at either end. The difference is very subtle to my eye...more noticeable with shadows, and slightly harder to blow out highlights - that's all. I honestly can't remember reading many (really any) reviews or forum comments (here or elsewhere) that say the DR difference is huge. It's interesting to see that per <a href="http://www.dxomark.com/index.php/eng/DxOMark-Sensor">DxO Mark</a> the D90 and D5000 out-score the D700 and D3 in terms of DR by 1/3 stop, and the D300s matches the D700 and D3.</p>

<p>The big improvement that I see is much better IQ in low light in terms of all-around lack of digital noise, better shadow area capture, and much less noise when some shadow areas are 'brought back' with post-processing. Maybe that's where the DR idea comes from. The other thing is the D700's remarkably low noise at high ISO, but just about any camera (film or digital) will exhibit less DR at high ISOs (clipped shadows, higher contrast).</p>

<p>Low-light performance is why I got a D700. I like to shoot in marginal light and low light at native ISO, and use long exposure times in other situations, and it's getting the job done for me. I seldom shoot at high ISO due to the increased contrast and lower DR (esp. shadows) that brings, but it's nice to have it if I need it. </p>

<p> </p>

[Ed_Ingold]

<p>Chip (and others),</p>

<p>What exactly do you expect to see with respect to dynamic range - an HDR effect? Highlight detail? Putting it bluntly, you gotta' learn to see better.</p>

<p>In the examples posted, the D700 (upper) shot clearly has more shadow detail than the D80 (lower) shot. Furthermore, blocking due to noise is clearly visible from the D80. I would not expect to see (and do not see) any significant difference in highlight detail between the two cameras. That's simply the way digital imaging (and slide film) work - blown highlights are beyond repair.</p>

<p>To use either camera correctly, you could drop the exposure a bit to capture details outside the window. If you meter the brightest part and open up a couple of stops, it will look bright in the image but not blown. This would darken the shadow area, but you are always able to restore that end of the range to some extent.</p>

<p>That is where the real power of the D700 comes into play. When you open up shadows in post you also magnify any noise. I used the Shadow/Highlight tool to do this (without spending a lot of time) to illustrate my point. The D700 image cleans up nicely with good color whereas the D80 image begins to look blotchy with muted colors.</p><div></div>

[Ed_Ingold]

<p><em> if I spend a lot of time flogging the shots in post production I can pull the extra range I paid for.</em></p>

<p>If you know what you're doing, it doesn't take a lot of time. My guess is you don't want to do anything. In that case, you picked the wrong hobby.</p>

[midan_smith]

<p>I have a D80 and upgraded to a D700, and the D700 is ten times the camera the D80 is. I loved my D80 but unless you are shooting with zoom and really need that 1.5x crop, the D700 does everything else better. Focusing: miles better. Skin tones: miles better. The dynamic range comes out when you shoot a 14bit nef and make changes in your raw processing software, and you see how much extra information the D700 has captured. I can also hand hold shots in poor light at f/8 and get very good images with the D700.</p>

[see_r]

<p>[corrected message below]</p>

[see_r]

<p>Edward, thanks for your response. I've attached the results of what happens when you pull the shadows out in the nef. For illustrative purposes only, I've done this at +4 exposure comp and then enlarged selected shadow areas. These results show the superiority of the D700 vs D80 in maintaining the shadow areas.<br />I should note that everything I have done here is not meant to be representative of how I would actually use each camera, from exposure to end result, but as a quick way to compare the cameras. Yeah, if you really want to get into photography you're never going to want to just go with the defaults--in this case-the default tonal response curve. I have my own personalized method of "HDR" that I use when I'm doing photography and not trying to compare cameras.</p><div></div>

[sem_svizec]

<p>I think the dynamic range is an odd story. It appears an important thing in the camera reviews. On the other hand, Nikon and other camera makers don't seem to care so much for it. The in-camera raw conversion and external raw converters mostly give up at around 9 stops, which is about what the display can show these days, but less than what the sensors produce (assuming that one is not very demanding regarding noise when evaluating DR). A 10-stop NEF will result in a conversion blown either in the highlights or the shadows (or both) with a NX converter, no matter how low contrast you set; NX converters do not do DR compression.<br>

If you want to compare extreme DR capabilities of cameras, you need to shoot with perfect exposure setting and process results with pseudo-HDR (multiple raw conversions with different exposure compensations to rescue shadows and highlights, then HDR blending). Certain raw converters (DxO, Adobe) let you get more DR in a single conversion, but the rendering (DR compression) options are limited compared to HDR programs.<br>

The current cameras are also not made to help you achieve maximum DR the sensor can hold. You have to exploit the highlights headroom but the histogram does not show how much you have. For best exposure you need "UniWB" (Rawnalyze is a good tool for exposure evaluation).<br>

Whenever feasible, true HDR with multiple exposures will give you better results than what you can get squeezing maximum DR from a single NEF. This is because the shadows range is (i) prone to noise (ii) "sparsely" recorded due to the gamma curve. But even with multiple exposures the DR is limited because of flare & ghosts (the highlights tend to spoil the shadow areas).<br>

</p>