7D not marked as highly by the DXO Mark as the Nikon D90

Discussion in 'Canon EOS' started by chris_gampat, Jan 9, 2010.

  1. Very interesting stuff. Sure, it's only slightly better but still in this economy it's quite jaw dropping.
    http://thephoblographer.wordpress.com/2010/01/09/dxo-mark-nikon-d90-bests-canon-7d/
     
  2. DxO Mark's results are meaningless crap, as we've discussed here many times before.
     
  3. The DXO Mark mark is determined through a very generalized process. It cannot be used to compare how two different bodies will perform in specific situations. More specific testing and results would be required to be useful. Since they are basing their findings on RAW images they would be judging different lenses from different manufacturers at the same time. It is not possible for them to eliminate the lens resolution variable from the results.
    Take their mark with a grain of salt.
     
  4. I'm getting tired of all these discussions about DXO's BS test results.
     
  5. It is not possible for them to eliminate the lens resolution variable from the results.​
    Lens resolution doesn't really affect measurements of noise, dynamic range, ISO accuracy or color depth, which is pretty much all that Dxomark report on.
    Just because you don't like, don't understand or don't agree with data doesn't make it BS or crap.
     
  6. Popular Photography January 2010 issue named the EOS 7D best in its class and runner up to the camera of the year just behind the nikon D3x. In their test it beat the nikon D300S in every aspect, except dim very low light AF. But it nailed it in high ISO and resolution.
     
  7. Right Angel, but still. The D90 was known to be better at low-light vs the D300. But no one compared the D300s vs the D90 or the 7D vs the D90. I'm saying that this could hint towards the D90 still being better than the 7D in which case I'd be shocked.
     
  8. No one made such comparisons because the D300S and 7D are in another class. The D90 is more closely matched with the rebel t1I or the 50D. In which case they are about the same quality. I've had the pleasure to handle both the D300s and D3s. They are very fine camera's and if it weren't for canon's true HD and manual video control, I would seriously consider a switch. But Canon came with the 7D which is also a much lighter camera with more MP than the D300s and still manage lower noise at high ISO is impressive. best nikon APS-C is the D300s, and the best APS-C of all brands is the 7D. PERIOD.
     
  9. The 7D has a smaller sensor and a higher pixel count, so it is quite possible that the D90 could win an image quality test. But regardless of which one does win, they're going to be so close that it's not going to make a difference in actual use. These days, I think the image quality of most cameras exceeds what 99.9% of users need, and that 0.1% isn't concerned with the difference between a 7D and D90, they're looking at 5D II, 1Ds III, D3x, or a Blad. What matters most is the features that make the camera versatile and easy to use for specific, real world tasks. The 7D beats the D90 hands down in any number of ways that matter to someone actually using it:
    • faster shooting and a bigger buffer
    • much better video
    • vibration free macro and tele shooting. D90 has no mirror lockup and in liveview mode it actually gives you double vibration , the mirror moves twice, dropping and raising again rapidly. 7D can shoot from liveview without moving either the mirror or the shutter: it's a macro shooter's dream.
    I could go on, but 7D is so cool that even a D90 owner like me can see it.
     
  10. The 7D is a great camera, and everytime I use it only reinforces that fact.
    Copy and paste from DxO web site.
    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
    Folks its just a number and nothing more.
    Lets add to the fact. DxO only tests one body. Not much of a sample size. I work in the manufacturing industry, and sampling one of something is not a test. Tolerances from one of anything can be on the high side of acceptable or on the low side of acceptable.
    Take the number for what is worth...
     
  11. Just because you don't like, don't understand or don't agree with data doesn't make it BS or crap.
    I for one do understand DxO's methodology. While I won't use the word crap, I will say their results are at best irrelevant to real world photography, and at worst false and misleading. They are pixel peepers who derive their performance estimates from theoretical models. As has been discussed a thousand times, final image and print quality cannot be judged by pixel peeping or theory.
    They're not the only ones who fall into this trap. R. N. Clark is a man I respect who offers a wealth of information at his site, along with beautiful photographs. Never the less, he publishes dynamic range estimates for sensors which are based on pixel noise measurements (pixel peeping) and theoretical models of how that translates into DR. Compare any of his estimates to a standard shot of a transmission step wedge and you will find his predictions are off from real world performance by 2-4 stops.
    Pixel peeping doesn't work. It's useful to know how pixels perform, but that knowledge must be placed in context. This context is missing from DxO's tests.
     
  12. I also understand the DxO Mark methodology (it's hardly rocket science), and I'm very happy to stand by my use of the word "crap" - a perfectly valid description of the welter of arbitrarily-defined and essentially meaningless numbers they provide, poorly presented and lacking in real context, which are being touted as valuable information and somehow indicative of what a given body will deliver in terms of image quality: many of which scores bear no earthly semblance to the Real-World evidence available and the hands-on experience many have had of the cameras in question.
    Crap indeed.
     
  13. Wow Chris, you really pushed some buttons. Do I sence some hostility from Canonites. The 7D might be a more capable camera or not? But for pure value, the D90 wins hands down. I played with the 7D briefly, found it inadequate for my needs and never looked back...my D300 rules for APS C.
     
  14. do they measure the final print, or do they measure 1's and 0's on a screen at 400% zoom? it's all crap, just buy a camera and shoot it. my d3 could take no better picture at normal to moderate iso's than a d90.........and high iso shots are often crap shots of boring subjects. just jump on the Digital Gravy Train and enjoy the ride.
     
  15. DxO Mark's results are meaningless crap, as we've discussed here many times before.​
    Good thing I read this response so that I did not have to type it myself. And Chris, it is not even vaguely interesting let alone very.
     
  16. Wondering what all this DXO stuff was about I finally looked at their site. What a shock! Oh my goodness, I found out I'm using a 4 year old, hopelessly out of date, entry level DSLR!
    Woe is me, woe is me, my camera would suck in DXO measurements.
    What shall I do? What should I do? WHAT WILL I DO!?
     
  17. A camera RAW file should tell everything about camera sensor + AD-converters etc. . The overall D0X score may be somewhat arbitrary, but the individual measurements about color depth, noise, dynamic range are physical quantities. And as such they matter in every day use. I use an S5 along with a D700. The S5 deals very good with high contrast situations (that is, high dynamic range), whereas the D700 excels in low light conditions (that is, low high-ISO sensor noise). Resolution, on the other hand, depends on how many pixels a sensor has, light, and of course lens quality. As mentioned, lens quality should normally not interfere with high ISO noise nor with dynamic range. Furthermore, RAW files are also indendent from the noise-suppression algorithms for producing the final photograph. These algorithms improve with time. These improvements are one of the reason that CANON could increase the sensor resolution while obtaining good high-ISO performance at the same time (another improvement along these lines is optimizing the microlens arrays for light transmission). As to me, pixel count is not a priority though. To me low high ISO noise and dynamic range are crucial. Others have different priorities, such as focus tracking, frames per second, etc.
     
  18. If raw images were indeed a true raw image the DxO results might be worth something despite their lackluster methodology. Unfortunately, all raw images are preprocessed to one degree or another as manufacturers attempt to optimize sensor performance. As manufacturers don't really reveal what exactly occurs during this preprocessing (nor would I likely understand it even if they did) it is difficult to equate results from overly constrained tests (like DxO) to the real world. This is doubly true if any of the manufacturers are "gaming" the tests... which occurs so often in other technology sectors that I wouldn't be surprised if it is occuring here as well.
    Thus, the final proof is still in the prints (or web images).
     
  19. "The D90 was known to be better at low-light vs the D300"​
    The OP should double check his sources, according to dpreview tests:
    "Yet, for all the similarities in publically announced sensor specifications and consistent color rendering, there are clear differences in per-pixel-sharpness and contrast. Both are shot with the same lens but no amount of re-shooting could get the D90 to match the D300's output."
     
  20. John Robison - I understand your despair. I share it. My camera is also hopelessly inadequate, someone who I never even met has revealed it to be so. What can we do? We'll have to give up photography, condemned to a life of abject failure as artists. Life can be so cruel.
     
  21. Heh. You guys's Nikon jealousy is cute.
    But seriously, it's just a measure of a few aspects of image quality from the raw file, and they're metrics where you'd expect the D90 - with its larger sensor and lower pixel count (look at the pixel pitch quoted) to perform better. You can't have one camera do everything. There are other measures of performance, like AF speed and frame rate, that are the 7D's specialty and where it's closer to the D300S's performance.
     
  22. Nikon and Canon have top pro models that compare well as do the lower models. State your case make your choice. In the middle things are not even. The D90 is a fine camera but it has a lesser build and is less of a pro camera than a 50D which IMO is not quite up to the build of a D300. All Nice cameras but tough to equate one on one. Haven't handled a 7D yet so I'm not quite sure how that fits in to the scheme of things.
     
  23. Heh. You guys's Nikon jealousy is cute.​
    Oh give it a break. I bought a D90 for my 17 y/o daughter. It's just a cheap little camera that has a little better high ISO performance than the D80 she was trying to use indoors.
     
  24. I had Nikon envy a couple of years ago and almost bought a D300. No more. Since buying a 7D the D300/D300S has absolutely no appeal for me whatsoever. It's certainly a superb camera but the 7D has enough capability to cement its appeal as the top APS-C camera available.
     
  25. enough capability to cement its appeal as the top APS-C camera available​
    Well i'm not going to counter that statement with a pro nikon one, but how do you define 'best'? i would rate the leica x1 as a better camera than the 7d, but that's only for me. personally I don't piss around with digicams much nowadays.
     
  26. Ouch. That hurt. It's a very respectable site so I must believe it's true. I think I'll sell my 7D and the 7 lenses I have and get a D90 and similar lenses. Goodbye Canon. It was nice while it lasted (20 years) but I must move on.
    Happy shooting,
    Yakim.
     
  27. I have a 7D and a 1Ds MKII and the only "negative" thing a can say is that when I use around 14 to 16 stops of ND filters, I noticed a slight bit of vertical banding on the shots taken, which easily are cleared with a noise software, other than that, it is an amazing and comfortable camera to shoot with. When I say comfortable I mean in terms of the optional bits and bobs it provides and for sure the other two cameras on the comparison could not provide.
     
  28. Thought I had a camera problem. Turns out my test chart has blurred print.
     
  29. I find the DxO Mark results consistent with the cameras I've personally used. I have no reason to believe the results aren't accurate for the items being measured. In the end, it's a number. A single number is meaningless for declaring one camera is "better" than another.
    For ISO noise, it looks like the D90 has slightly better results than the 7D. From what I can see on the SNR graph, it looks like noise in a given size print is about the same. So I'm not sure why the blogger is claiming that DxO Mark shows the D90 as being better, other than he's ignorant about how to interpret the data and wants to expose such. :)
    Looks like about an even match between the two in terms of prints. Make a series of prints from both cameras. Spread them out on a table. You're not going to see a difference.
    Eric
     
  30. OP writes
    the performance of the D90 compared to the 7D is "quite jaw dropping."
    Hmm. Nothing I've ever seen at DxO mark qualifies as "jaw dropping," let alone minor differences like this. A healthy serving of hyperbole, anyone?
    But it was a clever way to drive people to the OP's blog....
     
  31. Oh no! this means all photos taken with the 7D now has to be thrown out and new shots have to be taken using a D90.
     
  32. Remember the days when you just went out and shoot and enjoyed the photographs you made?
    Its funny to see the Nikon and Canon folks arguing about stuff for which you need a microspocope to distinguish between the results.
    Does this discussion go between Sony, Pentax and Olympus owners as well or their cameras are so much superior than Canon and Nikon that they don't need these kind of silly battles?
    I own a Canon 40D and I'm as happy as I can be, and I have close friends who are Nikon users and I am happy for them too as both brands are as excellent as they can possibly be. I guess the "who is better" arguments might be necessary for new users who are trying to choose between Canon and Nikon and need some kind of decision point (even if its pointless). Once you choose a brand I dont see the point of fighting like 4 year old kids.
    Oh well, I guess we need this competition so we can get better products each year for our particular brands but don't see significant differences that justify "jumping ship".
     
  33. David, I partially agree with you. Full frames are just so much better in low light, true. But it also depends on how picky your client is. I know photographers that still use the 40D and 5D Mk I and still get happy clients.
    Sure we're competing with one another as to who can deliver the better images, but we also charge a premium for it. Not everyone can afford that premium.
    For celebrity photographers, wildlife photographers, and news photographers the cropped sensor is very important.
     
  34. What are all these "best" qualification discussions about?
    Who is so desperately waiting for the 256.000 ISO? combined with 40Mpix and 25 Fps? etc.
    If the camera is blocking your qualities as a photographer, then it could be an argument,
    but to be honest, looking at what we could produce in the film era (256.000 ISO?), amazing it is.
    However, for a lot of people this fantastic DSLR gear is allways to expensive if it isn't (nearly) for free.
    Please, look at your own qualities as a photographer, specify your actual needs and your budget, get real and select the gear that fits best. Keep on dreaming of the ultimate thing and don't forget to make pictures.
    Cheers.
     
  35. To be honest, I think crop sensors are destined to become a playground for amateur and enthusiast photographer.​
    By that logic, is the Canon 1D IV an excellent example of a crop sensor that can't handle the dark and is just for enthusiasts? :)
    I'm getting similar noise from the 1D4 at 6400 as the 5D at 1600. From what I've read, my results are on par with what others are seeing. That's easily two stops improvement with 25% smaller pixel pitch. Just four years difference in technology. I have to believe this same sensor technology will be in the 1.6x crop sensors within the next 4 years.
    It wasn't all that long ago that "full frame" 35mm film was considered the hallmark of rank amateurs and real professionals wouldn't use anything smaller than medium format. :)
    Eric
     
  36. And I was looking to 7D as my next camera. Not good news my friend. Now I have to get another manufacturers camera and get new lenses. This is not good, wondering if I can sleep tonight. Oh DXO is crap. Never mind all is good again.
     
  37. Time to close this thread I guess, what was the purpose of this post anyway ?
    Topic: "7D not marked as highly by the DXO Mark as the Nikon D90"
    Ok, whatever makes you happy :)
     
  38. DXO ratings are non-sense
     
  39. I would worry more about how prints look than those DXO scores. The skill of the photographer is far more important than DXO scores or whether a camera has an FX sensor or DX sensor.
     
  40. Angel Bocanegra [​IMG], Jan 09, 2010; 02:12 p.m.
    But Canon came with the 7D which is also a much lighter camera with more MP than the D300s and still manage lower noise at high ISO is impressive. best nikon APS-C is the D300s, and the best APS-C of all brands is the 7D. PERIOD.​
    That is, until Sony releases their next generation CMOS sensor that will replace the one in the D300s and Nikon puts their fingerprint designs on it. The cheap Sony sensor in the Pentax K-X is already passing high ISO quality of D300s.
     

Share This Page