Nikon D90 better than Nikon D300s, D300. Second spot: Nikon D5000
http://www.dxomark.com/index.php/eng/DxOMark-Sensor/Camera-rankings
Nikon D90 has the best APS-C sensor results @dxomark
(48 posts) (10 voices)-
Posted 3 years ago #
-
So the D5000 is still "better" than the D300S? Hmm...
Posted 3 years ago # -
Well, those are all lab results, but yes!
Posted 3 years ago # -
I had the D90 before and I swear the D300s is better in terms of picture quality.
Posted 3 years ago # -
that's pretty weird, for sure... but a great bargain for the D5000 !
Posted 3 years ago # -
The D5000 looks like a real good buy. I wonder why they haven't tested the recent Rebels?
Posted 3 years ago # -
That's interesting that Nikon didn't put the D90's sensor in the D300s. I think this is the first time I've realized that. Does anyone know if that's because of the limitation on bit depth readout on the D90 sensor (i.e. 14-bit not available) or the readout speed or what? I shot 20,000 shots at 14-bit and 10,000 shots at 12-bit on my D300 and I wouldn't take even odds that I can tell you which are which. Let's get honest, I couldn't tell you which are from my D300 or my D90, either, so I'm sure all those at the top of that APS-C list have great sensors.
Posted 3 years ago # -
Very true Jonny. Despite this list, I would still buy a D300S over the D90/D5000. It has more of the features I want.
Posted 3 years ago # -
admin said:
I had the D90 before and I swear the D300s is better in terms of picture quality.I believe it, admin. You have to take these DXO results with a grain of salt because I don't think they actually measure resolution. They take their data and analyze it based on stated resolution, so the AA filters can make a big difference in IQ (not to mention autofocus and metering systems).
Posted 3 years ago # -
But overall it made Canon look bad, that is good enough for me, but then again they are not very nice to my dear old D300!!!!!!! I am going to need sleeping pills or something to sleep tonight
Funny thing though, I sold a D5000 to a real estate agent the other day and I am hired to take some pictures for him, I hope he does not find out that his camera is better than mine
Pete
Posted 3 years ago # -
I have the d90 and I really love it, although if I manage to sell it for a recent amount of cash I'll get the d300s - nevertheless, I love the IQ from d90
Posted 3 years ago # -
I like imaging-resource's comparometer. They have their standard pics for the D300s up so that you can compare it to the D90 if you like.
IMHO, based on their results, I would say that the D300s is noticeably sharper than the D90 at ISO 6400; but, the D90 seems to have fewer really obvious color blotches.Posted 3 years ago # -
bmxdad said:
But overall it made Canon look bad, that is good enough for me, but then again they are not very nice to my dear old D300!!!!!!! I am going to need sleeping pills or something to sleep tonightFunny thing though, I sold a D5000 to a real estate agent the other day and I am hired to take some pictures for him, I hope he does not find out that his camera is better than mine
Pete
Pete, I'll trade you straight up: my D90 for your D300.
I know you were only joking. What's the practical difference between 5 points (1/3 stop) on their scale?
Posted 3 years ago # -
Oh come on it still relates doesn't it? I'm sure the fine people at DxO labs know what they are doing and as long as Nikon cameras hold the top DSLR spots we accept their rankings. Once canon or Sony get past then we can cry foul ball! :^)
Posted 3 years ago # -
Oh come on it still relates doesn't it? I'm sure the fine people at DxO labs know what they are doing and as long as Nikon cameras hold the top DSLR spots we accept their rankings. Once canon or Sony get past then we can cry foul ball! :^)
"Are those kids French?"
LOLPosted 3 years ago # -
Looks like the D700 is the best buy for the money but it also shows purpose for all of the cameras through these ratings.
It's easy with these ratings to see which cameras are suited for studio, sports and etc, in making a choice of what camera fits ones venue.
I also like how these ratings exploited the Advance Digital Cameras for what they are.Posted 3 years ago # -
Not really Bland. This only test the sensor itself nothing more. It doesn't factor in AA filters and the other things Jonny mentioned above. This is just piece of the overall image quality puzzle. That's why these ranks are so controversial.
Posted 3 years ago # -
Interesting......sounds like Nikon should build only two cameras.....one for the Po folk...and one for the deep pockets. Image quality could be close, although features would of course be more prevalent on the high end camera. Probably not to be.....profit margin...Pro badging....more mon for Nikons fun....at the high end!
Posted 3 years ago # -
How can the effects of the AA filter not be taken into account, as they are basing their entire work on raw files, and therefore the AA filter is in the chain?
Posted 3 years ago # -
NikoDoby said:
Not really Bland. This only test the sensor itself nothing more. It doesn't factor in AA filters and the other things Jonny mentioned above. This is just piece of the overall image quality puzzle. That's why these ranks are so controversial.I understand what you and jonnyapple are saying and I agree to what you both are referring to. But imo this rating shows the average photographer a great deal of information when looking at cameras that are marketed similar.
Posted 3 years ago # -
LOL.....see the shot......take the shot..... no sensor bias ...no lens bias....nothing else to worry about!.....If it's good....it's good.....if not....try again!
Posted 3 years ago # -
soap said:
How can the effects of the AA filter not be taken into account, as they are basing their entire work on raw files, and therefore the AA filter is in the chain?Because they are NOT testing resolution or sharpness.
"DxOMark Sensor measures only the RAW image quality of a digital camera; therefore, DxOMark Sensor is NOT an evaluation of overall camera image quality or performance.
Resolution is NOT factored into the DxOMark Sensor ranking."
I believe the medium format cameras don't have AA filters.
Posted 3 years ago # -
Sorry in advance for the long post.
soap, that's a good question and one that is hard to answer based on what DXO puts on their website. They actually claim to take into account the AA filter, but I think that's only a half truth. They don't ever say exactly how they take their data, but here's what I'm guessing they do. I may be wrong, but I don't think I'm too far wrong.
They're careful to say that they don't test the quality of optics, so they must not have lenses on the camera when they're doing the tests. Fair enough. They're claiming only to test the sensor's performance. That means they must have well-calibrated light sources and they probably illuminate the entire sensor uniformly with that calibrated light. Then they must do something to measure the signal to noise ratio on the light measured by the sensor. I'm picturing that maybe they find the average level on a 12-bit raw file (4096 available levels) for each channel, and they take the standard deviation of each channel (R, G, B) and divide the average for each channel (say 3500, for example, for a grey light source in the red channel) by something proportional to the standard deviation for that channel (say they find the std dev is 10 levels in that red channel) to give a signal to noise ratio (of 350 in my example). They use that in their score for the dynamic range performance at that ISO setting to find out how many stops of usable dynamic range there are above the point where the signal to noise is 0dB (that's where changes will be indistinguishable due to noise). Maybe they average the SNR for the three channels to get the stated SNR.
Without a lens, there's no image being focused on the sensor and the AA filter isn't doing its job. The reflectivity (or transmittance, if you prefer) of the AA filter -will- affect both the DXOmark score and real-world images by changing how much light gets to the sensor (if some light reflects from the surface of the AA filter back toward the lens you have a smaller effective sensitivity since the sensor never has a chance to capture and record that light), but this is why I think it's a half truth: it doesn't take into account how much the AA filter affects resolution by low pass filtering, which is what I'd be more interested in since that's what the AA filter is designed for. This could be a problem if you had a situation like the Canon 7D—maybe its AA filter is as severe or maybe even a bit stronger than the one on the D300s, which would explain the paradox of worse resolution on the 7D's sensor in spite of having a smaller pixel pitch than the nikon's. DXOmark won't pick that up, and it's actually pretty difficult to find out since you'd need to keep the optics the same in your test (maybe with a 14-24 and an f-mount adapter for the canon and the same exact lens on the nikon?).Color depth might be done by finding out what the spectral response of each type of pixel is. Better scores would come from less overlap in the colors of light that R, G, and B are sensitive to while still covering the entire visible range. The signal to noise ratio would come into play here, as well, since noise will kill your color sensitivity just like it will kill your sensitivity to subtle changes in contrast. If this is true, they probably have a wicked complicated way of figuring how many colors can be distinguished at each ISO value based on the sensor's response to three or more well-characterized, pre-filtered light sources.
Feel free to critique what I've said. Like I say, I may be wrong.
Posted 3 years ago # -
shutterdancer said:
LOL.....see the shot......take the shot..... no sensor bias ...no lens bias....nothing else to worry about!.....If it's good....it's good.....if not....try again!Shutter, I actually like your explanation better. ;-)
I like the DXOmark sensor website and I think it's useful, bland, but what I was trying to say is that it's not the whole picture. What it tells you about is things like how well the electrical engineers did in designing the sensor, the amplification circuits, and the RAW processing circuits. The rest of the camera has to be tested some other way.
Posted 3 years ago # -
Actually Jonny my theory is that DxO shows pictures of bananas to a bunch of chimps and based on which photo is chosen they rank the cameras. Though I could be wrong and they might actually be using groups of gerbils instead.
Posted 3 years ago #
Reply »
You must log in to post.
