[creampuff]

Marc, as a general observation here you seem to need to prove your point all the time. Please don't bandy about with semantics, accusing others of straw man arguments, telling people to read your posts as if implying that we misunderstood what you were trying to say. I don't think we need such common ploys to attempt to deflect and undermine credibility. Sorry but I find your explanation to alunfoto here a little too rich to be believable:

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*.

------------------------------------------

Now you pursue a line of argument that differences in IQ between cameras is not hugely significant and support your line of argument by posting images that even the largest of which is no bigger than 1024 pixels on the long side. Of course at such a small size, it is easy enough for you to proudly proclaim that your argument is valid.

Of course we all know that's a lot of bunkum because at such small dimensions, it proves nothing, except that a lot of pixel information was discarded along the way as it was resized, so whatever differences between images from different cameras would largely have been lost. Well it certainly proves your flawed argument adequately.

You may think it's alright to post other people's pictures citing fair use guidelines after the fact to justify posting it here. As a Mod, I would have expected better behavior. It's leeching other people's plain and simple and I'm not even sure you had the common courtesy to ask for their prior permission nor even gave them any credits here. I can only conclude your standard of what is acceptable is different from mine.

[Recercare]

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...

Considering that VERY few people actually print their pictures it seems very reasonable to compare pictures at web-size. In addition, if you really need test photos of the same subject taken at the same time to differentiate the cameras...then you have already admitted that there's no significant difference.

If there was a significant difference between cameras you would have seen it even in a 800x600 picture. I am surprised that you don't realise this.
It's also quite childish of you to go off topic. It seems almost like you're too proud to admit that he has a point and then choose to bash his post with irrelevancies.....I mean, how Marc posted the pics is not relevant to what we are discussing here.

I would also like to add a story that pretty much says it all. Some of my close relatives and family work as pro journalists. They work with pro photographers and semi-pro photographers. During the last 10 years they have changed camera brand and models three times (Olympus, Nikon and Canon). Why did they change? Because one year Canon had the best prices and the next year Nikon had the best prices. Did the quality of the photographs change? Nope! Did the photographers complain? Nope!

Kind regards
.lars

[johnmflores]

OK, Match the camera to the photo. All photos taken within 30 minutes of each other.

Camera 1
6Mp, dSLR, circa 2004

Camera 2
10Mp, Point & Shoot, circa 2008

Camera 3
14Mp dSLR, circa 2009

Photo A

100% http://farm4.static.flickr.com/3551/3507534154_5cf48c774b_o.jpg

Photo B

100% http://farm4.static.flickr.com/3558/3498260105_907eca443e_o.jpg

Photo C


100% http://farm4.static.flickr.com/3542/3507532868_87281fd4f7_o.jpg

Respond by writing something like....
Photo A was taken by Camera 1
Photo B was taken by Camera 2
etc...

No cheating! Best of luck!

[Marc Sabatella]

Marc, as a general observation here you seem to need to prove your point all the time.

I don't how to respond to this. If I make a point, then people attack it *and me*, I'm not supposed to respond?

Now you pursue a line of argument that differences in IQ between cameras is not hugely significant and support your line of argument by posting images that even the largest of which is no bigger than 1024 pixels on the long side. Of course at such a small size, it is easy enough for you to proudly proclaim that your argument is valid.

And isn't that *exactly* what my argument was - that you have to pixel peep to see any significant differences? Again, it seems you *agree* with my actual point. At least, nothing you've posted indicates otherwise. That's why I keep saying I think you are misunderstanding my point. If you feel otherwise, feel free to clarify - but please, try to refrain from the personal attacks.

I can only conclude your standard of what is acceptable is different from mine.

Perhaps so. I don't see anything wrong with posting links to images that were publicly posted to an image-sharing site. If it offends you, don't look.

---

Last edited by Marc Sabatella; 05-24-2010 at 04:36 PM.

[Marc Sabatella]

OK, Match the camera to the photo. All photos taken within 30 minutes of each other.

Only thing that's obvious to me is that C is the 6MP camera - or that it's been resized smaller than the others. That much is obvious just fro the size of the original image. As for the others, it looks like A has more DOF than B, and B is shallow enough that I'll guess it can't have been taken by a P&S. Meaning my guesses are:

A: 10MP P&S
B: 14MP DSLR
C: 6MP DSLR

Of course, this wasn't based on IQ differences per se. Even pixel peeping at 100% as you allowed here by posting the originals, I can't say I had a *preference* for any of these. A is apparently the sharpest, but I wouldn't attribute that to the cameras but rather to the lenses and the chosen apertures and focus points.

Now will someone *please* guess with mine?

[cps_goodbuy]

A: 10mp p&s - depth of field
b: 6mp dslr - really looks like the colours I used to get with my Ist DS
c: 14 mp dslr - dull colours = what I'm getting with the newer dslrs imo (+ there is more fine detail on one of the "buds")

[creampuff]

Considering that VERY few people actually print their pictures it seems very reasonable to compare pictures at web-size. In addition, if you really need test photos of the same subject taken at the same time to differentiate the cameras...then you have already admitted that there's no significant difference.

Whether or not people print is not the issue. At web sizes, sure pictures from different cameras will appear similar. But to extend that argument and say that the differences in IQ between different cameras is insignificant is ludicrous. Particularly if your frame of reference is solely based on images that have been downsized to web friendly dimensions. If you're content with web sized images, then theoretically all we need is a PnS camera as even a current Pentax PnS can produce a 4,000 x 3,000 pixel image. Yet we know the quality of a DSLR's image betters that of a PnS camera.

If there was a significant difference between cameras you would have seen it even in a 800x600 picture. I am surprised that you don't realise this.
It's also quite childish of you to go off topic. It seems almost like you're too proud to admit that he has a point and then choose to bash his post with irrelevancies.....I mean, how Marc posted the pics is not relevant to what we are discussing here.

I think you just don't know what you're talking about. If your frame of reference is gonna be based on such a small image with a limited number of pixels, the differences may be less apparent from images from different DSLRs but certainly it will begin to show when compared a PnS cameras. Why not do your own apple to apple comparison tests between cameras before you continue on with your rant. If you want to argue on whether the differences are "significant" or not, there's no definitive answer as that's a judgment call.

I would also like to add a story that pretty much says it all. Some of my close relatives and family work as pro journalists. They work with pro photographers and semi-pro photographers. During the last 10 years they have changed camera brand and models three times (Olympus, Nikon and Canon). Why did they change? Because one year Canon had the best prices and the next year Nikon had the best prices. Did the quality of the photographs change? Nope! Did the photographers complain? Nope!

Kind regards
.lars

Your story is completely irrelevant to the discussion. Why your pro journalist relatives and family members chose to switch camera brands has no bearing to the discussion. We're not talking about gear prices nor photographic ability in this thread.

---

Last edited by creampuff; 05-24-2010 at 05:48 PM.

[Recercare]

Whether or not people print is not the issue. At web sizes, sure pictures from different cameras will appear similar. But to extend that argument and say that the differences in IQ between different cameras is insignificant is ludicrous. Particularly if your frame of reference is solely based on images that have been downsized to web friendly dimensions. If you're content with web sized images, then theoretically all we need is a PnS camera as even a current Pentax PnS can produce a 4,000 x 3,000 pixel image. Yet we know the quality of a DSLR's image betters that of a PnS camera.

I don't know why you suddenly add P&S camera to this discussion as we're discussing DSLRs. However, when I compare the last 250 shots taken with my K10d and Panny P&S I can see the difference even at websize. At 100% view the difference is obvious.
The point is that I doubt I would see much difference if comparing two APS-C cameras with all pics shot in RAW. For example, Nikon D5000 vs K-X....or a Nikon/Sony equivalent to my K10d.

If you want to argue on whether the differences are "significant" or not, there's no definitive answer as that's a judgment call.

This is true. People will always have different opinions about what is considered as significant.

Your story is completely irrelevant to the discussion. Why your pro journalist relatives and family members chose to switch camera brands has no bearing to the discussion. We're not talking about gear prices nor photographic ability in this thread.

It is relevant. The point is that if there was a significant difference in IQ between those brands/models they would have stayed with the "best" brand regardless of price. They go for the lowest price simply because they can't see any difference in IQ between the brands when judging the final results, which in this case are the published prints (newspaper and magazine).

Kind regards
.lars

[Rondec]

I guess I sort of agree and disagree with you, Marc. There are definitely differences in image quality between cameras and they show up in differerent situations where cameras are stressed. If you state up front that you are only going to use your camera for shooting web photos, will only shoot iso 100-800, and will never crop, then any camera with more than about 3 megapixels will suffice.

At the same time, I do shoot higher iso periodically (not all of the time), do crop (sometimes a lot), and actually do print photos (sometimes very large). In these situations, differences in image quality have a tendency to manifest themselves.

I will agree with your over all premise, that the similarities are much more pronounced than the differences, but I will also say that I can see differences between the digital SLRs I have used (K100, K10, K20 and K7) in every day useage. Of these differences, the differences between the K20 and K7 are probably the smallest overall.

[DonDouglas]

There are stock agencies who will accept digital files from cameras over eight megapixels, which is hardly a stringent requirement these days, but who're very specific about tonal range and how much highlight and shadow dot is allowed. The fact that the K20D is nearly a stop better in this regard makes it a more attractive camera than the K-7, regardless of how much the latter feels "like a brick" or how fast it focuses. The K-X is even better, but it's weird to have to choose between IQ and ergonomics/functionality. Why can't the top of the line be tops at everything?

[audiobomber]

OK, here is the challenge I promised.

Here are my guesses Marc. I didn't bother to pixel peep. I'm going by what I perceive as a yellow colour cast in many Nikon photos and a brown tinge in many Canon photos. I know that a K-7 has more accurate wb in Tungsten light and the K20D has slightly better high ISO performance, but I couldnt' see any way to differentiate in these photos so I called them all Pentax. I know that different lighting can make a mockery of my choices, but I feel pretty confident that #2 is from a Canon.

1. Nikon
2. Canon
3. Pentax
4. Pentax
5. Pentax
6. Canon
7. Pentax
8. Nikon
9. Pentax
10. Nikon
11. Pentax
12. Canon

PS I'm assuming you didn't skew the results by purposely choosing bad photos for some of these. Is it just a random sample?

[TheTenaciousOne]

oh wow creampuff your kick up a huuuuge fuss over a little statement. As Marc said, i dont think you get his intended point, so im going to try and write it how i take it (correct me if im wrong yea),

The fact that most people look at images (i did say most, not all) on their computers, or printed at most likely <A3 sized paper (about 16.5*11.7 inches), they have no need for the `better` IQ of a top or mid ranged dslr. You make the argument that of course you cant tell the difference on your computer, but that happens to be what Marc is trying to say. If you cant see the difference on your computer, and thats where your going to be looking at your photos on, then whats the point of getting a $10000 camera.


That said, i love my K7. I didnt really buy it for the IQ as i didnt see too much of an improvement over my old k200d, but just the weather sealing and ridiculously great ergonomics bought me over =)

Moral is - Buy the camera for the features, IQ should be good enough

[Marc Sabatella]

I guess I sort of agree and disagree with you, Marc. There are definitely differences in image quality between cameras and they show up in differerent situations where cameras are stressed. If you state up front that you are only going to use your camera for shooting web photos, will only shoot iso 100-800, and will never crop, then any camera with more than about 3 megapixels will suffice.

I'd say this is a reasonable assessment, although I think one can go *a bit* further without inviting too much dissent.

As far as resolution goes, we can actually go quite a bit farther. 3MP is already more than sufficient for printing 4x6" at 300dpi, so even at 3MP it is not *just* about web viewing. And if you have at least 6-8 MP (which is to say, virtually all DSLR's made over the last several years), you've got enough pixels that one would really struggle to tell any difference in terms of resolution when printing an uncropped image up to 8.5x11" (the largest most can generally print at home). Those same 6-8 MP also give you room to do a fair bit of cropping on pictures that will be viewed on the web or printed only 4x6".

So with most modern DSLR's, it's really only pictures printed 8.5x11" that are *also* cropped heavily, or pictures printed much larger than that, where there would be likely to be a noticeable difference, at least when it comes to resolution. Differences in color rendering and so forth tend to be much more subjective, and also easily affected by choice of WB or other in-camera settings, as well as choice of PP methodology. While I won't deny that those intimately familiar with their own cameras and their own particular PP methodology can often tell the difference between cameras in their own images, it really becomes pretty hopelessly subjective trying to claim one camera has "better" IQ than another based on these "intangibles".

That's all assuming relatively low ISO. As ISO levels get higher, differences are indeed going to be more noticeable. But even at ISO 1600 it's still the case that viewed at screen size or in 4x6" prints, the difference between the K-x (Pentax's best high ISO camera) and K10D (their worst) is fairly small to my eyes. If anyone wants to see for themselves, check out the K-x and K10D "comparometer" images for ISO 1600:

http://75.126.132.154/PRODS/KX/FULLRES/KXhSLI01600_NR2D.JPG
http://75.126.132.154/PRODS/K10D/FULLRES/K10DhSLI1600.JPG

The difference is clear enough in these images when viewing at 100%, but try or viewing them full screen or printing at 4x6". Especially considering that these are just default JPEG's, and that even five seconds spent applying an NR preset in your favorite PP program would be likely to lessen the difference, I'm guessing you could print close to 8x10" and many people wouldn't spot much difference. Of course, we are now getting to territory where more and more people *would* be able to tell. No doubt, when we push to the extremes - high ISO, larger print/view sizes - differences become increasingly obvious. Really, the only question is *how* far we have to push things before the differences become visible enough to any given person.

So despite all the bickering in this thread, I still think these are worthwhile issues to discuss and to try to put in real world context. Was my original statement unnecessarily provocative? Apparently so :-). But if it leads to a more in depth discussion of the issues, I'm OK with that.

[rparmar]

The fact that the K20D is nearly a stop better in this regard makes it a more attractive camera than the K-7, regardless of how much the latter feels "like a brick" or how fast it focuses. The K-X is even better, but it's weird to have to choose between IQ and ergonomics/functionality. Why can't the top of the line be tops at everything?

Sacrifices were made to get video on the K-7. And then improvements were made following that to produce the K-x. I think we can definitely expect to see the next generation being an improvement over the K-x but with all the features of the K-7 and a bunch more (largely useless) things jammed in as well.

[Marc Sabatella]

Here are my guesses Marc.

Thanks for playing! I think I'll cut my losses (in the sense of, let's put this part of the thread out of its misery!) and post the answers. You did a remarkably good job of picking out the Pentax cameras from the others (5 out of 6!). Only a little better than random chance beyond that, which is interesting considering we're talking about a top-of-the-APS-C-line Canon and a bottom-of-the-barrel Nikon. It's also interesting to me that you based this entirely (?) on something that is really just a matter of default JPEG processing settings (eg, AWB), suggesting it probably wouldn't have been hard for any of these photographers to have shot in such a way that you wouldn't have been able to tell so easily. But kudos for recognizing Pentax colors!

PS I'm assuming you didn't skew the results by purposely choosing bad photos for some of these. Is it just a random sample?

It's not random, but I did settle for the first three images I found that met my basic criteria: color (not b&w), taken in decent light, not ruined by poor technique, not obviously requiring specialized lenses or lighting equipment, image available in large size, comparable compositions, no identifying info (eg, watermark, reflection showing the camera used). I'll say the D40 seems to be owned by a lot more very mediocre photographers than the others - I had to look at a lot more images before finding three I thought good enough to use. But I'm just talking about poor focus, camera shake, cluttered composition, etc - not any fault of the camera. The three I chose I felt used the camera effectively.

BTW, I accidentally let #2 through despite the fact that it included a watermark that could have been used to cheat. I actually expected people to guess that one right just because it appear to show more resolution, not that one should really expect the camera to make such a difference when viewing images resized for Flickr.

1. Nikon D40
Flickr: More detail about Self-Portrait

2. Canon 7D
Flickr: More detail about 118/365 - Hotel Portrait Project: Fion

3. Pentax K20D
Flickr: More detail about Portrait - Maxime

4. Pentax K20D
Flickr: More detail about young female portrait

5. Pentax K20D
Flickr: More detail about Mathilde Portrait

6. Pentax K-7
Flickr: More detail about {~_~}

7. Pentax K-7
Flickr: More detail about Shepherd at Biertan, Romania

8. Nikon D40
Flickr: More detail about Formal(ish) Portrait

9. Canon 7D
Flickr: More detail about ~ Casual Portrait #1 ~

10. Canon 7D
Flickr: More detail about Photo365-20091208 - Just Another Self Portrait

11. Pentax K-7
Flickr: More detail about Susie at the cafe

12. Nikon D40
Flickr: More detail about Hello World- I'm Jayne.(self-portrait)