Canon 7D vs. Nikon D300s vs. Sony A500 ISO Test – Round 2
Written by Eric on November 2nd, 2009 | 37 Comments
If you didn’t get enough in the first Canon 7D vs. Nikon D300s ISO Test (and I know a lot of you didn’t, based on the lively comment section), we’re back for Round 2.
This time, we’ve got a couple of additional variables to mix things up a bit — and to see if the claims of some of the Nikon shooters in the comment section of the first round bear fruit.
First off, we’ve thrown a new camera into the mix – the new 12.3-megapixel Sony A500, which is a covers a sensitivity range of ISO 200-12800. And, we’ve still got the same Canon 7D and Nikon D300s from the last test.
This time, based on popular demand, we’ve turned off all noise reduction, dynamic range and auto-lighting optimizers. The exception being the Sony A500, which Sony must have felt it necessary to keep noise reduction turned on all the time as there are only two settings – Normal and High. As a result, we’ve gone with the lesser of the two evils and set the Sony A500 to “Normal” Noise Reduction.
Finally, we’ve bumped the exposure +1/3 stop from the last test.
If you’re expecting more of the same, get ready for a surprise.
Sample Images
Here is a sample of the complete scene. Below are 100% crop samples taken from the focus point of each image. No post processing was performed on any of these images other than the crops shown below.
You may download samples of each image for personal inspection by clicking on the links below each sample (right-click and choose “Save as…”). Do not republish the images on the Internet or elsewhere without express written permission, which may be obtained by email.
Nikon D300s vs. Canon 7D ISO 100
Nikon D300s vs. Canon 7D vs. Sony A500 ISO 200
Nikon D300s vs. Canon 7D vs. Sony A500 ISO 400
Nikon D300s vs. Canon 7D vs. Sony A500 ISO 800
Nikon D300s vs. Canon 7D vs. Sony A500 ISO 1600
Nikon D300s vs. Canon 7D vs. Sony A500 ISO 3200
Nikon D300s vs. Canon 7D vs. Sony A500 ISO 6400
Canon 7D vs. Sony A500 ISO 12800
Canon 7D and Nikon D300s RAW Images
The following 7D and D300s were opened in their native RAW formats using, Canon Digital Photo Professional and Nikon ViewNX, respectively. No edits were made before the files were exported to JPEG format for display here. Sorry, but I haven’t had a chance to install Sony’s software yet – look for that soon though.
Canon 7D ISO 6400 RAW Export Sample
Nikon D300s ISO 6400 RAW Export Sample
Conclusion
If you’re one of the Nikon shooters that commented in the last round about Nikon’s noise reduction, feel free to chime in with an “I told you so” whenever you feel like it.
I think the biggest lesson to take away from this is that Nikon’s in-camera noise reduction blows and should probably never be used. In the above samples, I really don’t see much noise worth reducing, especially in-camera.
As for the RAW image conversions . . . I’m a huge fan of Lightroom and use it for all of my day-to-day photography, but the differences between how Lightroom and ViewNX handle noise in a RAW file for the D300s is a tough pill to swallow. I’ll cut Adobe a break on the 7D files since it’s not officially supported yet, but the D300s noise is out of control at high ISOs in Lightroom 2.5. However, if you’re working with images at higher ISO settings, don’t dismiss opening up the manufacturer’s native RAW conversion software if things aren’t quite looking right in Lightroom or ACR.
Finally, you can see the Sony A500 is really hurting from the noise reduction as ISO creeps on up. I’ll also point out that the A500 seems to have missed the focus point, which was upon the Kodak emblem on the film canister. In the full images, the focus point was actually about half an inch behind that point on the joker card in the scene (not where I put it). And yes, I tried manual focusing with these cameras; however, my eyes don’t have the accuracy that they used to. Again, we’ll be looking at the A500 a little closer very soon.
I’ll leave the rest of it to you in the comments below.
The Nikon D300s, Canon 7D and Sony A500 are each available from B&H Photo. You can order these cameras or search for other photo gear by using the following links:
By making your photography purchases at B&H Photo through these links, you are helping Photography Bay to continue bring quality camera tests, news and reviews. Thanks for your continued support.
November 2nd, 2009 at 9:07 am
Let’s see how the results look when you have equal magnifications of the images! It’s obvious that the Canon 7D images are rendered at a higher magnification than either Nikon or Sony examples giving the Canon examples a clear disadvantage in this “Test”.
November 2nd, 2009 at 10:03 am
Nah, that explanation why you missed the focus in A500 is just not right… You TRY till you get it RIGTH. What kind of test is this otherwise? How can I compare how much details does the A500 noise reduction eat while you even couldn’t focus right?
I’d like to ask for a test to be made again – this time in the proper way.
Also – perhaps we’d see the photos from all cameras at noise reduction set on normal/middle value? This would give us some more idea on how does the NR works.
November 2nd, 2009 at 10:42 am
This kind of comparison is just nonsense to me but in reality it would never end. Picture quality is at the shooter not the camera. Beside what photoshop can not do?
November 2nd, 2009 at 11:16 am
Wait for the sony A750,that’s the one to compare, except if you are allready sure the a750 will be FF.
November 2nd, 2009 at 11:44 am
once more, Prepeare to get flamed! :)
nice post. Is sony using the same sensor that it sapplies to nikon? That would be internesting result if they were.
in addition, Another site that I know actually gave and Edge to the D300s (I would like to post the site, but I don’t know this site’s policy.. so ..)
November 2nd, 2009 at 12:49 pm
As far as I can see from the comparison, 7D win hands down in terms of details and noise. D300s NR is stronger and the A500 is strongest yet.
November 2nd, 2009 at 1:04 pm
Hi Eric,
I also enjoyed this High ISO comparison.
Personally it would have bin nice if you did the ISO 6400 Raw comparison with no noise reduction at all.
Or maybe better, done with dcraw perhaps….
That would have shown us a more equal processing.
There’s still an amount of noise reduction applied in the software.
Never the less, you’ve done a nice job here.
Thanks
November 2nd, 2009 at 1:40 pm
@Kenneth Bailey – The “difference” in magnification levels is because the 7D is 18-megapixels, while the A500 and D300s are 12-megapixels. As a result, taking the same crop of 600px x 150px will cover a smaller area of the scene at 100% magnification. In short, there is no “difference” in magnification levels.
@Sky_walker – The AF point was on the same location with the A500 as it was with the other cameras. Additionally, if you will click over to Round 1 that is linked above, you can find the D300s and 7D at Normal noise reduction so you can see how NR works for all cameras. Those original files are available for download as well.
November 2nd, 2009 at 3:17 pm
Horrible test. You can NOT use AF for tests like this.
Learn to manual focus on a specific area and even the playing field. You are not testing resolution or noise performance if you can’t understand how to manual focus.
Not saying this because I love Sony – I own a 7D – but lets be fair about things, the Sony a500 and a550 are exception cameras at high iso but your test makes their image quality seem wrong. I can get an a200 to produce a sharper crop than what you were able to come up with.
November 2nd, 2009 at 4:45 pm
make a comparaison with photographing a grey chart & will talk.
i tink that noise level is less on the nikon anyway,but for prints above A3+ the canon will lead thanks to higher resolution & will see more details.
finally, it’s really abvious to compare a pic at 100% screen,no one does it in everyday use!!!!& the beter way to handle noise is to shoot in raw & reduce it after thanks to photoshop,nx2,DPP, lightroom…..try & ull see!
go make photos guys, what ever the camera u use!
November 2nd, 2009 at 7:54 pm
I would like to know how you achieved a 600px x 150px crop on images with such different resolutions? Did you first magnify the images to 100% on screen and then ‘eyeball’ the area of the images you wanted to magnify or do you have some sort of new fangled measuring device that precisely measures an area to the exact same measurements and coordinates! I would consider it to be extremely difficult to match a 12.3 megapixel camera to a 18.2 megapixel camera having the same sized sensor; because that 18.2 megapixel camera image has to be reduced beyond the native resolution of the 12.3 mp camera to achieve the same results. There is a much smaller distance to travel, if you will for that 12.3 mp camera to achieve a 600px x 150px image.
November 2nd, 2009 at 8:08 pm
@Kenneth Bailey – Using Photoshop Elements, I set the selection rectangular marquee tool to “fixed size” mode at 600px wide and 150px high. Zooming images on-screen doesn’t really matter because 600px covers the same area no matter how I’m “viewing” it. As a result, the images are shown at 100% across a 600px x 150px section of the original file.
November 2nd, 2009 at 8:38 pm
It’s pretty obvious, at least in my view, that the 7D image has to be magnified more to achieve the same fixed 600px area coverage. Any time you are increasing magnification you are reducing resolution.
November 2nd, 2009 at 9:32 pm
@Kenneth Bailey – The 7D has a higher resolution, so 100% resolution on it is going to be a bigger image. I did not res-up the image if that’s what you are getting at.
November 2nd, 2009 at 10:56 pm
On the internet, everyone is an expert!
Great test Eric, I appreciate you posting your finds.
November 2nd, 2009 at 11:01 pm
Seems pretty close between the Canon & Nikon personally i always shoot raw &
remove the noise in Photoshop with Nik dfine.
Finally some real competition for Nikon
November 2nd, 2009 at 11:04 pm
@Arturo – Appreciate the comment. Thanks for stopping by.
November 3rd, 2009 at 2:27 am
I always appreciate constructive criticism whether it agrees with my point of view or not! 45 years in this biz has taught me that there are many times when I am not right, but I never stop asking the important questions. Next time maybe you can compare my 5D to the output of my Mamiya RZ w/P65+ Back.
November 3rd, 2009 at 2:47 am
Eric, thanks a lot for your tests. I think You not the prepaid tester =)
November 3rd, 2009 at 3:43 am
Thanks Eric, good job on the test.
Sad to hear that the SONY didnt focus right. Some critism can, and already has been, be given.
It would be interesting to also see åics at some of the higher ISO, where you have compensated for the higher magnification (and 1.5 / 1.6 crop). I guess anyone could try their mathskills at home, but it would be nice to see how far ahead the Canon 7D would be.
All the best,
Superfly
November 3rd, 2009 at 9:42 am
It’s becoming fairly common knowledge now that comparing 100% crops on cameras with DIFFERENT size sensors is incorrect.
Compare an exact portion of the frame instead.
November 3rd, 2009 at 10:47 am
The focus of the A500 is horrible, and its no fault of the camera’s. As a reviewer, shouldnt u get things to work properly, and in this case, get the focus correct, before doing a test?
Look at the low ISO pictures of the A500. The image is so soft and that has nothing to do with ISO. Its your inability to get the focus correct that is causing this, and as a result gives readers the wrong idea that the A500 is “poor” in comparison with the other 2 cameras.
Bad test.
November 3rd, 2009 at 4:20 pm
Some how i think they known a Lower cost Sony A500. which cost less than 7D and the D300 And the A500 was going to kick them right out of the Box. i always did say that Nikon and Canon are money pits
November 4th, 2009 at 9:48 am
Kenneth Bailey “45 years in this biz” and dont understand 18mp x 12mp.
Download the samples and see by yourself. Compare as you like.
Great test Eric.
November 4th, 2009 at 11:54 am
What I think is really being missed here is that Canon shouldn’t even be in the same league as the Nikon (but it obviously is!). We have an 18 megapixel camera against the Nikon’s 12. Many said that when the 7d first came out they were disappointed in Canon cramming 18 Mp onto an APS-C sensor… but with all that resolution if noise is roughly the same on the 2 cameras the Canon wins because with the same size prints each speck of noise will be smaller… finer grain (smaller) noise means a better (less noisy) print.
The one reservation I have about the 18 Mp is that the DLA (diffraction limited aperture) is a mere F 6.8 for this camera. The high pixel-density means that you begin to lose sharpness to diffraction at a mere F 6.8! I’d be interested to see a test of how much sharpness is lost. This potentially is the greatest drawback to the 7d potential as a landscape camera because you really need the smaller apertures for maximum depth-of-field. In sum, this looks to be a great sports, wildlife, action shooter… but depending on the diffraction loss maybe not so good for landscape, scenic… The 5D Mk II is still preferable in that dept..
Canon has really done something with their low-light technology here and I’m anxious to see how all the new tech & features trickle down (& up) to the rest of the Canon line. Also the competition will inevitably make Nikon kick it up a notch. Ain’t competition great! It’s an exciting time for photography!
November 5th, 2009 at 11:38 pm
Eric,
Very nice test! The only somewhat valid criticisms here relate to the fact that the Canon is slightly disadvantaged due to the resolution difference.
Having said that however, I’m not sure what the Nikon shooters are bragging about. The fact that the Nikon’s in camera NR sucks is not a selling feature! And the Canon still wins this test in a close comparison even without equalizing the image size. Yes the difference is much smaller with no noise reduction, but the Canon still wins!
The Canon also appears to have better dynamic range as the whites look whiter. (Perhaps that’s an exposure difference.)
Based on the excellent test pics here, any potential new camera buyer would have to ask “Why would I turn off the high ISO noise reduction in the Canon in regular use?”, followed by “Why would I pay $100 more for a camera with a lower resolution whose NR sucks, and whose larger pixels don’t appear to give it any advantage?”
You can rest assured that any Nikon shooter’s “I told you so” on this test is a myopic one.
November 6th, 2009 at 2:29 am
It would be nice if you had one additional comparison where you you resized the 7D to nikons size and compare them. That might be an interesting test.
While these test are good starting point, they do not tell as much about real situations. For example, I do the smallest contrast change to the 5D II and convert it to Black and White and I get horrible pattern noise. That’s something that will not show in these tests, unless the camera testers would add it to their procedures.
All said, thanks for a great post.
November 6th, 2009 at 11:09 am
In the ISO 6400 you can really see how Nikon’s ‘RAW’ NEF is using the Low Pass meadian filter to reduce noise. Canon is truer by not filtering that kind of noise. The chroma noise is very high in the Nikons which is a little surprising. IT would be interesting to see if you did a TRUE Nikon RAW file via Mode3 hack if the results are more similar. The Sony looks like an over filtered Noise Ninja applied image.
November 6th, 2009 at 11:56 am
Please redo the A500’s out of focus shots. Don’t leave it like that.
November 6th, 2009 at 1:13 pm
The reviewer ensures that neither Canon nor Nikon would have to come last by posting out-of-focus A500 images. Learn to use the camera before posting reviews. If the camera is defected, return it. The reviews like this belong in garbage can.
November 6th, 2009 at 2:01 pm
Thanks, again, Eric, for an enlightening comparison. Part 1 showed how well the 7D performs despite everyone’s claimed “deficiency” in using an APS-C sensor. A local store was offering such a great package deal that I bought a 7D the next day, and it’s doing exactly what I had anticipated it would. This test shows that Nikon’s in-camera processing is terrible, and that the Sony has a problem with auto-focus — both good things to note. And thanks for your explanation of the larger size of the Canon image. I appreciate your time and efforts, which informed my decision.
November 6th, 2009 at 3:32 pm
The Photos should be croped for the object to be the same size, reguardless of the pixel count, because partly the smaller the cell, the less photons per electron versus the larger cells.
Also the basic contrast should be adjusted to be the same for a truer comparison. The more contrast can give it an appearance of more noise.
Also from what I’ve read, this is just a test for the CMOS sensor, not the electronic niose reduction. The sensor is the the most inportant source of the camera.
The other factors such as the focus sensors needed inprovement, which is what I believe CANON did for the 7D.
I believe NIKON has the edge on more reliable exposures, where the CANON has the better low noise sensor.
November 6th, 2009 at 3:40 pm
Thanks for the tests…. It nice to see people do this kind of stuff. Personally, I think the most important part of photography is the person behind the camera. Because it doesn’t matter if you have a $100 or $10,000 camera if you don’t know how to use it. Yet don’t forget people will defend their personal preferences in terms a camera system. I also enjoy the Canon vs Nikon vs Sony battles as it keeps pushing manufactures to new innovations and helps keep prices lower.
Keep your head up.
November 7th, 2009 at 5:40 am
Good to see that the long-awaited DPreview 7d review basically confirmed your results here AND the validity of your tests, Eric. Congrats. The Canon low-light tech in the 7d is real.
November 12th, 2009 at 5:33 pm
The 7D images are “bigger” because 7D has a 18MP sensor while the other two is only 12MP.
I see no problem comparing 7D with D300 because they are about the same price range. It is fully justified if Nikon does not lower the price to compete with a lower level Canon. On the other hand A500 is half the price of the other two. I see A500 in the same league with T1i or D5000.
November 16th, 2009 at 7:01 pm
I think you should take the Sony images down until you can verify that the focus point was correct. I for one simply cannot believe that a properly focused camera, even on an a200, would take pictures like that at the low iso settings. If there is a problem with the camera or the lens then that needs to be sorted out as well.
November 17th, 2009 at 10:35 pm
Apart from the Sony images not having sharp focus it was a very good review between the Canon & Nikon.
If Canon has also nailed the AF it may be the crop sensor champ at least until the Nikon D400 arrives.
The D300s has a better auto iso implementation.
And the D300s has a better AF working range: -1 to +19 EV
The 7D AF working range: -0.5 – 18 EV.