Is the K20D Resolution worth the money?

[sfa1966]

Yes, the K20D has a better sensor (to state the obvious), which (amongst other goodies) resolves more detail very apparently in a severe crop: but how many dollars is that worth?

Is it worth the money? For people who make their living in photography: yes, probably. But what about us amateurs? Speaking as someone who who prints at A4 or less, and is still yet to exhaust his first pack of Epson A3-size glossy paper (now over 8 months old), and as someone who prefers the tonality of your K10D shot in the OP, my answer, in one word, would be "no".

--
Cheers,
sfa

A very limited photographer ...

 

[yesman12]

I am surprised at how much of a difference there is between the 10D and 20D...
I have been resisting this upgrade, my wallet hopes I can continue to wait.
--
Best regards
Nick

 

[*isteve]

Hi,
my proposed testing setup was actually intended to demonstrate the
visual magnitude of the resolution advantage of 14 MP vs. 10 MP. I
thought this is what you were after?

Huh, I'm "after" nothing i only present the results, I have no hidden
agenda. Anybody can make their own conclusions, that's what the test
was for.

There was a misunderstanding regarding anti-alias-filter strength.
Whenever I use that term, I always mean the relative strength
(relative to the resolution, or the photo pixel site's size), not the
absolute one. Whether the K20D is weaker or stronger (relatively,
i.e. net of resolution differences) - I don't know, probably not
decuctible from real-world scenes.

If you don't know, and I'm also not in the know then there is no
point to discuss relative nor absolute AA impact.

But I just spot something else more worrying - do I see some
sharpening halos in the blue back of that sign post for the K20D?
But didn't you follow a RAW workflow? If so, would this be a sign of
in-camera sharpening of RAW files (which I wouldn't like and would
count against the K20D, if true)? So the K20D resolution advantage
is kind of a cheat?

In my initial post i made clear the same settings i used for both
cams, implicit those settings can only be effected in jpeg. Maybe i
should mentioned 'jpeg' as a clarifier. I didn't wanted to test RAW
because my initial goal was to test default output out of camera.
Furthermore i use Lightroom/ACR and have not a clue at this time if
processing is different between the two PEF files.

Hope this helps.

ACR adds sharpening by default. Whether its the same for K10 and K20 I dont know, but I do know that the K20 does NOT sharpen its RAW files.

Thanks for your test Richard, I dont have an issue with your methodology. At the end of the day, the difference is quite obvious whatever way you cut it, and of youi print at A3 or larger its worth every penny. I also would not want to give up the AF adjustment now I have it.

Having said that, I am still very fond of my "old" K10 and will still use one as a backup for a while. My GF has her eyes on the other one

--
Steve
When I can master technique I'll be a photographer.
When I can realise a vision I'll be an artist.
When I get paid I'll be a professional.

 

[Richard Smals]

Yes, the K20D has a better sensor (to state the obvious), which
(amongst other goodies) resolves more detail very apparently in a
severe crop: but how many dollars is that worth?

Is it worth the money? For people who make their living in
photography: yes, probably. But what about us amateurs? Speaking as
someone who who prints at A4 or less, and is still yet to exhaust his
first pack of Epson A3-size glossy paper (now over 8 months old), and
as someone who prefers the tonality of your K10D shot in the OP, my
answer, in one word, would be "no".

Off course i didn't answer my own question, because the reader must answer it. You did just that, and that is the only truth for you.

I love my K10D also and did a wonderful job the past 19 month, it did take over 70000 shots by me or my students. And it will be in service along the K20D no doubt about that.

--
Bye4now



http://www.indots.nl

I have the deepest respect for all those people who like me.

 

[janneman02]

it seems overexposed.... same settings???

Maybe lboth have slightly different metering as the "darker areas"in the K20D seem lighter as well
--

The difference between genius and LBA is that genius has its
limits.

• Janneman ( adaptation of the Kings quote from Albert Einstein)

 

[Richard Smals]

it seems overexposed.... same settings???
Maybe lboth have slightly different metering as the "darker areas"in
the K20D seem lighter as well

Ha, same exposure settings but window of light was continue changing. Waited until the house was in the sun again between shots. Plus i needed the swap lenses.

--
Bye4now



http://www.indots.nl

I have the deepest respect for all those people who like me.

 

[lock]

I do not like the software upscaling cause its no real pixels you put into them. Though the size is different, this is what the cameras do on their own if you fix one of the dimensions. And still you see more quality in the k20d. But the detail in the pictures of the k20d has been the charm of it from the moment it was brought on the market. Yesterday I read a review of a 5D user who stated that up to iso 800 the k20d can rival the 5d in terms of everything...And I've heard similar comments..not bad for an ap-c camera !

lock

 

[Roentarre30]

Your comparison shots do show a great different with 100% crop.

Yes, it is worth it.

Not sure why people are so antagonistic about k20d. They could basicly go and use ist DS which is so much cheaper with better tonality ... blah blah blah

--
Roentarre, Melbourne, Australia
http://www.roentarre.com
http://www.roentarre.blogspot.com
http://facade-roentarre.blogspot.com
http://photoblog50mm.blogspot.com

 

[Weziah]

is to print (by yourself if you have a good printer, or in a lab) the photos (same size), and then compare them.

Would be nice to print them in different sizes to see at what point the K20D resolution advantage would become visible.

--
http://www.flickr.com/people/lazar

 

[dpthoughts]

I don't dare to do a comparison based on the given material, because it is just too flawed methodologically, i.e. having softening artifacts applied to K10D from scaling, but not to K20D; having granted K20D more sharpening than K10D from not normalising their default sharpening levels - exactly what amateurish review sites have been criticized so much for in the past. Therefore, I still better wait for the DPReview K20D review, because I admit that their testing always reflected reasonably sound knowledge about digital photography topics, IMHO.

However, when disregarding the K10D image for now and looking at the K20D's in isolation (even though flawed from JPEG sharpening), I'm really positively impressed (yet another time) from what it can deliver in conjunction with a good lens.

Thumbs up!

Is it worth the money? For just learning or for just teaching photography: No, the K20D would be a waste of money and time here, better to forget about it and stick to the K10D, esp. since it probably will become dirt-cheap in ebay (as so many K10D owners indicate they are planning for an upgrade).

For anyone striving for excellence in A3 (or bigger), regardless whether pro, amateur or hobbyist, the anser would be: yes, go for the K20D. In addition, there are the other factors, partly already mentioned: Does noise look appealing (fine-grained, film-like versus 'digital')? Dynamic range? Lens calibration? Advancements in focusing or shake reduction (where opinions differ, though)? These other factors seem to point towards the K20D, as well.

 

[Richard Smals]

I don't dare to do a comparison based on the given material, because
it is just too flawed methodologically, i.e. having softening
artifacts applied to K10D from scaling, but not to K20D; having
granted K20D more sharpening than K10D from not normalising their
default sharpening levels - exactly what amateurish review sites have
been criticized so much for in the past.

WoW what a lot of heat... I provided you with the original out of camera crops with EXACTLY THE SAME JPEG settings here: http://forums.dpreview.com/forums/read.asp?forum=1036&message=28278926

no resize just a crop. I explained that i use ACR and that ACR is not giving the same treatment to both cams. You came with nothing then complaints although i was straightforward with my mythology and settings use right from the start. Even DPReview uses the jpegs for comparison. It sound more like you act as a pit-bull, just the fun of biting in legs without reasoning or provocation.

Therefore, I still better
wait for the DPReview K20D review, because I admit that their testing
always reflected reasonably sound knowledge about digital photography
topics, IMHO.

Do that, that's no problem at all.

However, when disregarding the K10D image for now and looking at the
K20D's in isolation (even though flawed from JPEG sharpening), I'm
really positively impressed (yet another time) from what it can
deliver in conjunction with a good lens.

The K10D had the same level of in camera JPEG sharpening.

I don't know why you act like you do but for me you tear my test out of proportion without any strong foundation. I looks more for the fun of it.

But OK that can happen, but i don't have to agree with it.

--
Bye4now



http://www.indots.nl

I have the deepest respect for all those people who like me.

 

[Adam-T]

Everyone knows the JPGs from the K10 Suck, especially out of box - Both samples look like they're from P&S Digicams, only the second one has the sharpening turned up ..

Shoot the thing in RAW to tell any worthwhile rez increase, all those are showing which has the most Haloed JPG defaults .

--
Please ignore the Typos, I'm the world's worst Typist

 

[jamonation]

Everyone knows the JPGs from the K10 Suck, especially out of box -
Both samples look like they're from P&S Digicams, only the second one
has the sharpening turned up ..

Shoot the thing in RAW to tell any worthwhile rez increase, all those
are showing which has the most Haloed JPG defaults .

I'll second that. I could certainly have purchased a sweet P&S for the money and shot in jpgs, but the whole point of buying and dslr for me is capturing in RAW (admittedly my Canon P&S has the hacked firmware).

Obviously the K20 is going to produce a better looking image, why else would the money and time have been invested into building the thing in the first place? Why bother changing to a CMOS sensor if the K10's CCD was sufficient?

In general, I wouldn't call the test method used flawed so much as just proving the obvious fact that yes, indeed, 4 million more sensor elements to work with makes images look better. When you think about upscaling and interpolating that way, adding 29% (100 - 10/14mp) more information to an image than was originally there pretty much guarantees the results achieved. Surprise!

In fact, if interpolation (of a jpg no less) is such a good method to use for this test, then I suppose I should be able to upscale those old 3.1 megapixel images of mine to 16x20 prints with no noticeable sharpening or blurring right?

 

[Adam-T]

thing in the first place? Why bother changing to a CMOS sensor if the
K10's CCD was sufficient?

The K10's CCD is sufficent, it's very sharp and in RAW and high ISOs better than any of the 12Mp cams based on the Sony CMOS (Dunno about the K20's Sam Chip, not compared and I'd NEVER ask a K20 owner for an unbiased reply)..

The problem with the K10's CCD is that it's only 10Mp and consumers expect 12 or more in a top line camera - the Difference in Rez between a K10 and K20 is more than it is compared to a D300 or A700 but it's not night and day (Night and Day is K100 -> K20 ) ....

The K100D was the First Pentax with a Top JPG engine (the earlier cams were Vile in JPG) for some reason the K10 slipped back but they pulled it back with the K200 - the K20 there doesn't look very good, in fact it looks like my Canon A650 with the sharpness cranked to the Max (Ugly), A Fuji S6500FD JPG (6Mp Bridge cam with a 28-300 effective lens) looks like Foveon at 100% Compared to that K20 Crop ..

--
Please ignore the Typos, I'm the world's worst Typist

 

[jl_smith]

The only fair comparison is to compare either full prints , or compare 100% crops.

Upscaling the K10D's crop to match the K20 is a flawed test - You'r introducing a foreign element to the picture...that element is normally not present in a K10D photograph. Even considering the small difference in crop "zoom factors", I think the 20d is sharper.

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

I believe if one were considering between the two, it would be beneficial to compare body feature sets, ISO performance, etc.

Then, I would consider how much of a cropper I am (a birder, for instance). If one finds themselves constantly cropping 100% to make photos, the K20D would probably be worth the upgrade.

--
http://www.flickr.com/photos/jl_smith