[alunfoto]

Can you tell the difference with the eye? I don't really need video so if they are pretty close I could get the K20D. Thanks.

They are pretty close. The Samsung sensor is _essentially_ the same in the two cameras, so in principle the IQ is the same.

If/when you buy the K20D, there are some things you could try to check for the particular camera. Some owners have complained about the sensor exhibiting reticular noise on high ISO. Some owners also found that there was a band on the left side of the frame that would exhibit more noise than the rest of the image. RiceHigh has an article referring to some posts here on the issue. I've seen it mentioned elsewhere too, but can't find the link at the moment.

My K20D exhibited both faults to some degree, but I didn't feel hampered by it. I may have had a good sample, or I may have ignored its shortcomings, I dunno. I do think that there's some sample variation, however, and that high-ISO performance is where it shows up.

Currently I have two K-7, and the files they produce are indistinguishable. Naturally I have checked to see if they exhibit the same faults as was reported for the K20D, and for my samples of the K-7 this is improved. The band is gone, and the reticulum is less pronounced.

To me, like many others have posted already, it was the _other_ improvements with K-7 that made a difference. But then again I don't shoot above ISO 800 very often.

[ll_coffee_lP]

I'm not saying there are no differences whatsoever. Of course there will be variations here and there. But they won't be of the sort that make one camera consistently stand out as noticeably better or worse in a blind test.

I guess you've used a significant number of camera's then and are quite familiar with the various sensors they use. Unfortunately I'm not that blessed, I can only compare the two camera's that I have that have different sensors. I find it decidedly interesting that I can notice a difference in the small sample size that I have access to though...especially when you have much more sensor experience than I.

Or maybe, you just like to talk.

c[_]

[Marc Sabatella]

I like to point out things that aren't always obvious to newcomers. The relatively small magnitude of the IQ differences between cameras is one of those things. I've seen images from dozens of cameras, and once more, I stand by statement as intended and as explained in subsequent posts for the benefit of those who were not able to get the point and insisted on nit-picking whether it was 100% views or only 90% views that would be necessary to tell the difference, or if the statement was only true for ISO up to 1600 or if it also include 12800 etc.

But since this continues to be a controversial statement, I'm going to put up a challenge. Stay tuned.

[Marc Sabatella]

OK, here is the challenge I promised.

I expect those who misunderstood the intent of my original remark to immediately object that this is not a controlled test. We're not comparing the *exact* same image taken with the *exact* same lens with the *exact* same exposure and the *exact* same post-processing. My response is that is the differences between cameras are so small that these other variables make it impossible to spot the camera differences, then I've successfully made my point. That point being that the IQ of the camera itself is extremely low in importance on the list of things that influence the quality of the image, to the point where in practice it's mostly a non-issue for most people in most situations. I tried to make this point by mentioning pixel peeping as a way to tell the difference, and some people apparently missed the broader point.

But this might be fun anyhow. Following are three portrait images each from the K-7, K20, the Nikon D40, and Canon. I chose the latter two because I figured if we're going to see big IQ differences, older entry level cameras and new top-of-the-line models from other manufacturers will show them the best.

I'm curious to see the guesses. Good luck, and I'll post answers after a while. Thumbs here, links to Flickr "large" sizes below.


http://farm3.static.flickr.com/2786/4393196945_d49ebe0570_b.jpg


http://farm4.static.flickr.com/3452/4561979127_9b964e0c2d_b.jpg


http://farm3.static.flickr.com/2578/3911176592_6497ae4607_b.jpg


http://farm4.static.flickr.com/3360/3562954927_d030f33fbb_b.jpg


http://farm4.static.flickr.com/3366/3601990589_eb8c592315_b.jpg


http://farm3.static.flickr.com/2682/4522050650_bb99d18107_b.jpg


http://farm5.static.flickr.com/4038/4450999922_270a37a527_b.jpg


http://farm4.static.flickr.com/3397/3407607110_abbe4e62c7_b.jpg


http://farm2.static.flickr.com/1030/4598729327_8c8dda9b61_b.jpg


http://farm3.static.flickr.com/2718/4183353910_73a81c8c08_b.jpg


http://farm5.static.flickr.com/4015/4591278499_b13d4316be_b.jpg


http://farm4.static.flickr.com/3651/3370834839_62bc62b4bd_b.jpg

---

Last edited by Marc Sabatella; 05-23-2010 at 11:52 AM.

[Marc Sabatella]

Bonus question: here is a controlled test image from each of those cameras, courtesy of the Imaging Resource site. All ISO 1600. The D40 is too old to have a sample of this image, so I substituted another camera that by all rights should do poorly: the Olympus E-520.

I imagine some of you have already seen these images often enough to have memorized the differences, and there are any number of ways you can cheat and figure out which is which on these, but try not to spoil it for everyone else.

Well, I have no idea which one is which.

Probably, the subtle differences come from the fact that each of the company was very much aware of what the competitor's pics will look like, so I am sure each of the four manufactures did this or that to differentiate themselves from one another. They will take on a certain "tendencies" in terms of metering and degree of exposure and things like that.

I have to agree that most of us don't have as good an eye as others who look at these things a lot. I am also not so sure if having those eyes that can differentiate will benefit a photographer. Paralysis by analysis may well be a real danger.


But, one thing I would like to add, is that given the same subject, when you shoot with different cameras without a tripod, in a course of day or longer, pictures will start to look different, not from the IQ standpoint. Because of the weight of the body and camera, the balance of it, the way viewfinder is set up, I think that eventually there will be a some sort of difference at the end. It is sort of like shooting with zoom vs. prime at the same focal lengths.

[rparmar]

I am with Marc. I cannot tell the difference between shots on my K100D Super and my K20D, even though they are nowhere near the same in terms of number of pixels or anything else. The difference between light, focus, DOF, lens, post-processing, white balance, etc. all far outweigh any difference in the cameras.

Even a small bit of PP obscures differences. The K20D is superior for features, ergonomics and ability to crop while retaining detail. That is all.

[MSM]

I am with Marc. I cannot tell the difference between shots on my K100D Super and my K20D, even though they are nowhere near the same in terms of number of pixels or anything else. The difference between light, focus, DOF, lens, post-processing, white balance, etc. all far outweigh any difference in the cameras.

Even a small bit of PP obscures differences. The K20D is superior for features, ergonomics and ability to crop while retaining detail. That is all.

I agree with Robin and Marc on the above points. I have been considering the K7 not because the IQ is supposed to be remarkably better then the K10. My question, though, is the K7 an improvement in AF and Metering. These along with WB and better high ISO (for me up to 800-1600) is what I am after. To me the AF and the Metering on my k10d has caused me to miss many more images and is a bigger concern than any perceived gain in IQ.

I would love to hear everyone's thoughts on the AF and metering. Thanks.

[johnmflores]

Easy little game, Marc,

The best images are clearly from the Nikon that you are envious of!
The second best images must be the sharp ones where the Canon locked focus instantly!
The third best ones are the Pentax you wish you had or the older model that is still better than the newer models. Your choice.
The fourth best is the Olympus because everyone know that real men shoot full frame and FourThirds is for hacks and amateurs!

[DonDouglas]

It's hard to know at this resolution, but the first image appears to have significant noise. The fourth and last appear to have highlight clipping, but the soft lighting is very forgiving. Outdoor shots with bright highlights in clouds and deep shadows under trees would be more telling. This set does prove that for publishing on the Web, and for other purposed under diffuse lighting, pretty much any camera will do.

[creampuff]

Wow, a moderator going out of his way to post assorted images taken by other people (without any credits given... aka leeching) to prove a point, which I can't for the life of me see how relevant it is. Perhaps posting his own test photos of the same subject taken at the same time with different cameras and sized to a large enough size (same resolution) would give a truer picture of any so called differences. As it stands I don't think the photos posted prove or disprove anything. At such a small size, even a camera phone can give passable results...

[alunfoto]

I like to point out things that aren't always obvious to newcomers.

Pontificating to newcomers, you mean? Surely they must deserve more thorough explanations than one-liners? This subsequent elaboration is a sad retrofit.

But since this continues to be a controversial statement, I'm going to put up a challenge. Stay tuned.

After seeing your challenge, I had to revisit your one-liner:

IQ-wise, virtually *all* DSLR's are the same unless you're pixel peeping at 100%.

I realise now that your logic is impeccable. Downsize everything to web resolution, and noone sees a fart of difference. It would probably be true if you downsized to the resolution of the weakest camera too.

However I think it is a bit tragic that a lot of image information has to be discarded just for the sake of winning an argument.

[RoxnDox]

Pontificating to newcomers, you mean? Surely they must deserve more thorough explanations than one-liners? This subsequent elaboration is a sad retrofit.



After seeing your challenge, I had to revisit your one-liner:



I realise now that your logic is impeccable. Downsize everything to web resolution, and noone sees a fart of difference. It would probably be true if you downsized to the resolution of the weakest camera too.

However I think it is a bit tragic that a lot of image information has to be discarded just for the sake of winning an argument.

It's also a bit tragic that a thread has to degenerate into a mud-slinging contest. Marc expanded on his original statement. Why not just accept it and MOVE ON from this pointless argument?

Jim

[Marc Sabatella]

Wow, a moderator going out of his way to post assorted images taken by other people (without any credits given... aka leeching) to prove a point, which I can't for the life of me see how relevant it is.

It's relevant to my point, which despite my having explained over and over, you seem to be going out of your way to avoid getting. Please read my posts instead of taking that one line in a way contrary to how I have clearly explained I intended it. Again, the point is that differences in IQ between cameras is not hugely significant in most situations - nowhere near as important as a zillion other factors. I never said there was *literally* no difference. I said *virtually*, and by that I meant, no difference that *matters* very often in the real world except when pixel peeping. As it is, I think you are arguing a straw man here; I doubt you actually disagree so strongly with my point - mostly I am guessing it is just the way I worded it.

Perhaps posting his own test photos of the same subject taken at the same time with different cameras

Needless to say, I don't *have* those cameras; that's why I went with this approach. These images are publicly available on Flickr and are being used in accordance with Fair Use guidelines.

At such a small size, even a camera phone can give passable results...

If you click the links below each picture, you get the larger size, just as I explained in my post. If you are saying you need to see even larger than that in order to tell a difference, then you have just acknowledged I was correct in the first place.

[Marc Sabatella]

Pontificating to newcomers, you mean? Surely they must deserve more thorough explanations than one-liners?

I've provided that at length for the benefit of those few of you who failed to get my point.

I realise now that your logic is impeccable. Downsize everything to web resolution, and noone sees a fart of difference. It would probably be true if you downsized to the resolution of the weakest camera too.

Precisely. I apologize if that wasn't clear form the very beginning, but I have no idea why the statement generated to much controversy and even animosity.

However I think it is a bit tragic that a lot of image information has to be discarded just for the sake of winning an argument.

It's not for the sake of "winning an argument". It's for the sake of helping newcomers sift through the information overload provided by to see for themselves how big of a difference in IQ there is between cameras *in actual practice*. In other words, to provide as direct and relevant an answer to the OP's actual question as I think possible. But don't worry, I saved the links to the rest of the info, and will post them at some point.

But since I went to the trouble of doing this, won't someone humor me and try their hand at guessing before I give out the answers?