[Marc Sabatella]

I know what the statement means but this seems so different than everything I read in reviews.

No, it isn't. Reviews are all about pixel peeping at 100%.

Can you explain a little bit by what you mean here? Do you mean raw? Also, what's pixel peeping?

RAW has nothing to do with it. Pixel peeping means looking at the image larger than can fit on your screen (which works out to being more or less equivalent to a print a bit smaller than your screen). Doing this means you are seeing the individual pixels of the image - like looking at a newspaper or magazine print through a magnifying glass.

That's the way most reviews compare cameras, because it's really the only way to reliably see differences. And my point is, if you have to make an image bigger than your screen (like, taking a regular portrait but then viewing it so big that it fills the whole screen with just an eye and nose) in order to see differences, then who really cares? Of course, there are reasons why these differences might matter - making large prints, or if you do drastic crops of your images, or if you're a professional shooting for very demanding clients. But for most purposes, the differences just don't matter.

Looked at another way, I could post ten images each from the K20D, K-7, and any other four DSLR's from any manufacturer you care to choose, and if we're looking them at sizes that actually fit on our screen, nobody on this forum would be able to reliably tell which came from which camera.

[Marc Sabatella]

In my experience this is untrue. The CCD sensor of my K10D and the CMOS sensor of my K-7 have different IQ's. In most cases I can tell what camera I used to take the photo just by looking at the image (and not at 100%). One isn't worse than the other by any stretch, but they are different.

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 am going to switch gears a bit.

On prints (obviously the same exact papers on the same printer back to back), I actually can tell the difference between the two in many of the times. By the time you get to A3 size, it becomes even more easier to tell. Again, I don't say this in regards to what is commonly debated as "IQ." I think that the sensor engineers are sort of past all that. Looking at the both outputs by these 2 cameras, it is to me more about simulating how the film pictures used to look, that appropriate level of graininess. I am pretty sure that the tweaking of K-7 sensor was not about the sharpness (well, obviously . . . ).

So, I am sorry that I may be one of those people, or few of those people think that K-7 is far superior than K20D. I really can't even think of going back to K20D. In fact, I use K10D more than K20D. I am a tad disappointed to hear that there are no reasons to update from K20D to K-7. To me there are plenty. That viewfinder of K-7 I think is a gem. It is not quite like that of ME super, but I can certainly frame much better with K-7. Metering seemed much improved too. THe balance and the gripping is much better. Shot for shot, obviously without a tripod, K-7 takes better pictures for me.

K20D had a lot to prove at the time. K10D was more or less a game changer of some sort, and Pentax was burdened with unfair level of expectation, I think. They really went for that punchy sensor. The colors are very vibrant, sharpness was very edgy with minimal noise, and with lenses like FA ltds and DA*'s, K20D really made subjects "pop out." K-7's sensor is different. It is to me more "mature."

[jgredline]

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

Why do you say dumb things like this? This is so not true.

As for the OP question. Image quality is about the same until IS01600 when the K20D is cleaner. By the way, for awesome Jpegs out of the box, the K10D is very good as well, maybe even better than the above. All though I do love the DR of the K-7 and it does meter better.

[Marc Sabatella]

Why do you say dumb things like this? This is so not true.

Of course it is - in the sense in which it is meant. As I said, of course there are *some* differences. But whether they are enough that the average person could actually reliably tell - and express a consistent preference for one camera over another in a blind test without pixel peeping - is another matter. Some extremely picky people might be able to, and perhaps you are one of them. But I see no reason to expect the average newcomer - would.

[justtakingpics]

Of course it is - in the sense in which it is meant. As I said, of course there are *some* differences. But whether they are enough that the average person could actually reliably tell - and express a consistent preference for one camera over another in a blind test without pixel peeping - is another matter. Some extremely picky people might be able to, and perhaps you are one of them. But I see no reason to expect the average newcomer - would.

But I'm trying to be like you guys! I don't want a camera that I will soon learn is not up to snuff because I was an average newcomer and that that is the way the statement was intended originally. If I am going to invest in a prosumer high-level amateur camera that would mean (to me) that IQ and their comparisons matter very much - to some degree anyway. I'm sure it can get like comparing CPUs in the computer world where you need a computer program to see the difference. This is roughly analogous, in some ways, to pixel peeping. Sooner or later I'd have to just make a decision.

[Recercare]

Why do you say dumb things like this? This is so not true.

It is both true and not true. I suspect that most photographers have their own preferences regarding sharpness, colour and contrast, and by shooting RAW you can (at least in most cases) tweak the image to fit your style/preferences regardless of brand. There will always be subtle differences though but I think they're more relevant for the pixel-peeping photographer.... I doubt clients or the "average man/woman" will notice it.

Kind regards
.lars

[howieb101]

For the typical person who views photos, ie, not a photographer trying to justify their camera over another, within APS-C most viewers wouldn't be able to tell what photo's came from what camera and whether the images were obviously better than anothers. Again, this is at normal viewing distances, not at 100% on a screen which is obviously not normal except for pixel peepers.

Now, if a typical viewer has a preference for sharpening or saturation that is different. They could definitely say they prefer one image over another but that is because they prefer a certain type of look.

[Rondec]

But I'm trying to be like you guys! I don't want a camera that I will soon learn is not up to snuff because I was an average newcomer and that that is the way the statement was intended originally. If I am going to invest in a prosumer high-level amateur camera that would mean (to me) that IQ and their comparisons matter very much - to some degree anyway. I'm sure it can get like comparing CPUs in the computer world where you need a computer program to see the difference. This is roughly analogous, in some ways, to pixel peeping. Sooner or later I'd have to just make a decision.

People who buy DSLRs actually are generally quite satisfied with their ultimate decision (as compared to computer purchasers). The reason is that there aren't any cameras out there now that aren't capable of taking quite good photos. Some just make it easier than others. The real question is what capabilities you as a photographer have and what you are able to see and learn.

Either of these two cameras will be a good tool for that, although at this point I don't really see picking the K20 over the K7 due to the lack of a price differential.

[Marc Sabatella]

But I'm trying to be like you guys! I don't want a camera that I will soon learn is not up to snuff because I was an average newcomer and that that is the way the statement was intended originally.

Understood. My point still stands, just not as strongly. That is, I still say the differences that reviewers obsess over are generally not as significant as it might seem when reading the reviews. They'll end up mattering more to some, less to others. It's really hard to say whether you'll be one of the ones to whom it matters more. But given this stuff is almost entirely subjective, there's also very little point worrying about it. In a blind test of the minority people who *can* tell the difference, you'd probably find about half the people preferring the K20D, half preferring the K-7. So even to the extent that *some* people *might* consistently prefer one over the other, that's still not really a useful way for *you* to decide on a camera unless *you* happen to be in that minority of people who can consistently see a difference for yourself. And if you were, you wouldn't be asking.

Sooner or later I'd have to just make a decision.

I agree. I'd just suggest that someone else's opinion of difference in IQ should be way, way, way down on the list of priorities.

[jpzk]

From what I've seen, I've found to K7 to have what we could call a pleasant grain at ISO1600. Though what really stands here is where this grain does not come at the expense of detail at this particular level. Therefore... my conclusion is that the K7 and K20D are both capable of similar definition within that given range(200-1600).

The JPG however was aimed squarely at the K7's JPG processing capabilities in contrast to the K20D's JPG output. Which I find the K7 to be top notch in that respect. Though I really can't comment on the camera's onboard NR capabilities.

On the issue of my K7 NR workflow, I've received several offers for some K7 high ISO candidates and so I'm anxiously waiting those to get back onboard the NR train(so to speak )

Hi John,
I get the point you are trying to make here, comparing ISO handling and results between the K7 and the K20D.
What I would really to achieve is high-ISO shooting with the K7, in RAW, and still keep the details while the noise is removed. Even with NoisewarePro, this is rather tedious and often the end results end up as "mushy" (lissage).
As for the offers for producing high-ISO pics with the K7, I think I had replied to your post .... I'll retrieve the thread and see.
I can't hardly wait to see what workflow you will be able to come up with.
Cheers.

JP

[johnmflores]

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

I'm with Marc on this one. When I got my K20d last Spring, the first thing that I did was take it and my other cameras (a 5 year-old 6Mp Nikon D70 and a 10Mp Ricoh GX-100 advanced P&S) into the yard to shoot the dogwood tree that was blooming in our yard. I was a bit surprised that the differences between the 3 (and the 2 dSLRs in particular) weren't eye-poppingly obvious. See for yourself: 2009.05 Dogwood in Bloom

This is only one anecdotal test. In low light/high ISO the differences between the cameras will be greater. But in most cases, there will be much more variance in the lenses than in the sensors.

Now regarding the K20d and K-7, I'd hazard to guess that the differences are small enough that you'd be better of choosing between the two based on other criteria, i.e., size, price, need for faster AF/FPS, etc...

Good luck!

[DonDouglas]

I'm planning to upgrade from K10D to one of these two cameras. It's apparent that the K-7 gets better marks than the K20D in just about every respect other than dynamic range in RAW. To my eye, pixel peeping is quite unnecessary to see clipped highlights. The same is less true of noisy shadows, which are largely correctable anyway. According to the Imatest results published at IR, the K-7 is about 3/4 stop worse than the K20D in this respect, worse even than the K10D. I'd love the upgrade to the K-7's physical improvements, but not at the expense of DR. The multi-exposure extended DR JPEG feature/gimmick is not attractive to me. I'm curious what users of both K-7 and K20D think of the two considering this aspect only, especially when comparing with the K20D's dynamic range extension in RAW.

[howieb101]

Well some might call the extended DR a gimmick. Others can conclude for themselves whether its useful.

For me, at lowish iso the extended DR function works. Keep in mind that sometimes it does flatten the scene. Something akin to flattening a scene when using onbaord flash.

All up still worth iit for me as I do use it.

[alunfoto]

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

Even for a one-liner, I find this to be no more than a smartass statement, for two reasons.

First, it implies assumptions about how IQ is defined. I'm sure it's possible to define IQ in such a way that it supports the one-liner, but that would be a definition bound to come up and bite you in the ass pretty soon. Even when using the same CCD or CMOS chip, DSLR makers differ on their implementations of eg. antialiasing filters and in-camera processing like the Bayer interpolation.

Secondly, it also implies that looking at an image at 100% magnification is the only way to infer something about image quality. If this is true, then please explain. :-)