Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
10-19-2012, 04:27 AM   #76
Pentaxian
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,862
QuoteOriginally posted by UnknownVT Quote
The point was even a cheapo not-so-great 24Mp budget entry level dSLR (Nikon D3200) shows a lot more detail than any of the class leading 16Mp such as the Nikon D7000 and even Pentax K-5 - regardless of sharpening level.
I won't argue with the statement.

But just a disclaimer. I found the DPReview (and FWIW the Imaging Resource) studio comparison test shots a complete waste when it comes to evaluate sharpness. Just don't do that.

The editor of IR once confessed to me that the shots aren't meant to be interpreted for sharpness anyway. They are meant to compare noise and color rendition.

There is a lack of control over choice of lens, selected aperture value (yes, it varies!), focus plane (it is a deep subject and they rely on AF which is not accurate enough for such comparisons (even if it is LV) and post-processing parameters (one parameter fits all camera/lenses is not true).

It is much easier to simply compare MP figures. And 24 is a 22% linear increase over 16.

10-19-2012, 07:50 AM   #77
Veteran Member




Join Date: Sep 2007
Location: USA
Posts: 1,812
QuoteOriginally posted by falconeye Quote
But just a disclaimer. I found the DPReview (and FWIW the Imaging Resource) studio comparison test shots a complete waste when it comes to evaluate sharpness. Just don't do that.

The editor of IR once confessed to me that the shots aren't meant to be interpreted for sharpness anyway. They are meant to compare noise and color rendition.
I realize you do know the difference between sharpness and resolution.

Quite simply - regardless of sharpening (within reason of course) those pictures show that a 24Mp sensor even on a cheapo budget entry-level dSLR will display more detail than the class leading 16Mp such as the Nikon D7000 and Pentax K-5 - that was the only purpose, not as a comparison of sharpness, or any other aspect of the cameras.

So to get back on topic - I very much doubt that even without an anti-aliasing filter the new 16Mp Pentax K-5IIs can show more details than a 24Mp....
is that fair? - no mention of sharpness.

Just as an illustration re: sharpening:
from: Nikon D3200 Review: Digital Photography Review
This is the JPG comparison -

QuoteQuote:
At low sensitivities and, crucially, with a high-quality lens (we use the Nikon 50mm F1.4 for our studio tests) the D3200’s 24MP CMOS sensor captures a lot of detail. That said, at pixel-level the Nikon’s output looks a little soft which is caused by Nikon's rather conservative approach to sharpening. Noise is not much of an issue at low ISOs.
The 100% crop from the 24Mp Nikon D3200 JPG is distinctly SOFT (less sharpened) especially compared to the other 24Mp Sony SLT-A65, yet I can see a LOT more detail than either of the 16Mp

QuoteOriginally posted by falconeye Quote
It is much easier to simply compare MP figures. And 24 is a 22% linear increase over 16.
No problems here seems very much like commonsense
- the only reason I even posted beyond my first in this thread was because of this:
QuoteOriginally posted by normhead Quote
Unfortunately I read on one forum that the 24 Mp APS-c sensors don't compare at all favourably to the 16 MP D7000, and they blamed that on the sensor at 16 MP already out-resolving the lenses. But I'd be happy to hear it if someone says different.

Last edited by UnknownVT; 10-19-2012 at 07:58 AM.
10-19-2012, 08:12 AM   #78
Pentaxian
normhead's Avatar

Join Date: Jun 2007
Location: Near Algonquin Park
Photos: Gallery | Albums
Posts: 23,765
QuoteQuote:
The point was even a cheapo not-so-great 24Mp budget entry level dSLR (Nikon D3200) shows a lot more detail than any of the class leading 16Mp such as the Nikon D7000 and even Pentax K-5 - regardless of sharpening level.
When Mona was up here and had her D3200 I uploaded both her and my pictures to my hard drive and examined them both. The D3200 has a decided advantage in jpeg rendition.. but not in raw files. As for "a lot more detail".. I'd say pretty much a saw off there. You definitely did not see a lot more detail if any. We didn't take a lot of shots but the ones we did of the same things, you wouldn't have been able to tell the difference. I just trashed her images yesterday unfortunately or I'd put up a couple 1:1 crops to show you.

I think the K-5 IIs will out perform it for detail despite being 16 MP against 24 MP. That's my prediction. It's exciting that a brand new 24 MP camera can compete with a 2 year old K-5. Although I think it's more comparable to a K-30 without the support for legacy lenses, which is worth at least $150, and would be a deal breaker for many of us. But being able to match the details captured is impressive. But if you look at the features of a k-5 as opposed to a D3200, there are lot of features there that are probably worth the extra 300 bucks to a lot of us. When I was explaining a few concepts and we were flipping through her menus to try and do on her camera what I was doing on mine, we didn't always come to a happy conclusion. You can do most everything you can do on my K-5 (except HDR which is pretty much worthless anyway) but it takes a lot more time.

I remain unconvinced you get much of anything for going to 24 MP with an APS-c camera. Now 24 MP FF, that's a totally different story.
10-19-2012, 08:44 AM   #79
Veteran Member




Join Date: Sep 2007
Location: USA
Posts: 1,812
QuoteOriginally posted by normhead Quote
When Mona was up here and had her D3200 I uploaded both her and my pictures to my hard drive and examined them both. The D3200 has a decided advantage in jpeg rendition.. but not in raw files. As for "a lot more detail".. I'd say pretty much a saw off there. You definitely did not see a lot more detail if any. We didn't take a lot of shots but the ones we did of the same things, you wouldn't have been able to tell the difference. I just trashed her images yesterday unfortunately or I'd put up a couple 1:1 crops to show you.
I think controlled conditions is a very important operative word here -
despite criticisms the comparison shots and measurements from dpReview.com do at least try to have some semblance of controlled conditions -
which, without any disrespect, casual shots by two different photographers with different grades of lenses cannot achieve.

Of course you may not be convinced by measurements from review sites,
nor the "controlled" comparative pictures at 100% crop -

So this is just another attempt including the newer Pentax K-30 which is not 2 years old -
from: Nikon D3200 Review: Digital Photography Review (for anyone else reading this, please click on the reference link to read the details.) -

I can honestly say I see more details in both the 24Mp APS-C Nikon D3200 and Sony SLT-A65 over the 16Mp K-30, and K-5 for that,
BUT of course I accept your opinion may differ.

I more than accept that for most 24Mp may not make that much practical difference over a 16Mp -
after all I shoot a K-x at 10Mp instead of its native 12Mp......

So getting back ON TOPIC - then how much difference is there to make using a dSLR withOUT an anti-aliasing filter worth it?


Last edited by UnknownVT; 10-19-2012 at 08:53 AM.
10-19-2012, 08:56 AM   #80
Pentaxian
normhead's Avatar

Join Date: Jun 2007
Location: Near Algonquin Park
Photos: Gallery | Albums
Posts: 23,765
I'm not only unconvinced by the images you posted, but I'd also suggest that you'd be unable to do so with unlabelled images. Exactly where is the detail that you are referring to that isn't present in the K-30 image. IN detail, that would be something that is in the D3200 image, that isn't in the K-30 image. In the images you've shown, the lost detail in the K-30 image is lost because they blew the exposure on the K-30 image and her cheek is blasted. I fail to understand how you didn't notice that given the fact that you're here trying to convince people of the varacity of your expertise. You can see detail, But you can't tell one exposure is blasted and the other isn't . Look at the collar of the image bottom right. The K-30 and K5 both look sharper than either of the other two images. But, the images are practically identical.

SO, can you see that the K-30 and K-5 images are blasted, that the K-30 /K-5 did a better job on the bottom right, that the testers didn't properly expose the K-5 /K-30 images and that practically, if you weren't looking at the other images, you'd have no problem with any of the other images, or are you and I living in different universes?
10-19-2012, 09:06 AM   #81
Veteran Member




Join Date: Sep 2007
Location: USA
Posts: 1,812
QuoteOriginally posted by normhead Quote
I'm not only unconvinced by the images you posted, but I'd also suggest that you'd be unable to do so with unlabelled images. Exactly where is the detail that you are referring to that isn't present in the K-30 image. IN detail, that would be something that is in the D3200 image, that isn't in the K-30 image. In the images you've shown, the lost detail in the K-30 image is lost because they blew the exposure on the K-30 image and her cheek is blasted. I fail to understand how you didn't notice that given the fact that you're here trying to convince people of the varacity of your expertise. You can see detail, But you can't tell one exposure is blasted and the other isn't . Look at the collar of the image bottom right. The K-30 and K5 both look sharper than either of the other two images. But, the images are practically identical.

SO, can you see that the K-30 and K-5 images are blasted, that the K-30 /K-5 did a better job on the bottom right, that the testers didn't properly expose the K-5 /K-30 images and that practically, if you weren't looking at the other images, you'd have no problem with any of the other images, or are you and I living in different universes?
I think the images are clearly labeled - this was simply a screen shot from the referenced/linked dpReview.com page - I did nothing special. They are the converted RAW test shots - as described on that page.

OK - there's the area under the eyebrows that has cross hatching - it is very clear in both the 24Mp APS-C images whereas they are hardly resolved either of the 16Mp K-30 or K-5.

Then there's the vertical watermark-like lines crossing the face - again both 24Mp APS-C images show quite clearly - whereas they are hardly visible on the 16Mp.

There is a difference between sharpness and resolution/detail as I tried to illustrate in Post #77 including the quote from dpReview, and I said so too that the D3200 image was soft. But one can see more detail as described above.

I live on earth, where civil discussion, and even disagreement
does not need to resort to rudeness.

Last edited by UnknownVT; 10-19-2012 at 09:11 AM.
10-19-2012, 09:40 AM   #82
Pentaxian
normhead's Avatar

Join Date: Jun 2007
Location: Near Algonquin Park
Photos: Gallery | Albums
Posts: 23,765
Whilst doing the dishes and pondering those shots it came to me, if I'd brought in an assignment like the one posted for the k-5 and k-30 and tried to excuse that kind of loss of detail when the camera is clearly capable of better, he would have laughed my sorry ass right out of class.

These guys do a fantastic service to photographers putting these images up the way they do. But they probably don't have the time to sit down and completely optimize every image they take. You have to know enough about photography to look at those images and know what you're seeing. If the camera can produce the quality shown in the bottom right of a K-5 and K-30 images, it can produce the same quality throughout the frame. The issue isn't with the camera, it's with the photographer.

Sorry if I was little harsh.. but this is basic photography. It's your job as photographer to get it right, not the cameras. And it's never the cameras fault. It's the photographers fault. If he's using the wrong equipment to start with that's his fault. If he's not getting everything he can out of an image, that's his fault. The camera can't do that for you everytime, sometime you have to know enogh to overide the deaults and get it done right. The first rule of photography.. maintain maximum possible detail in your original image. If you want to blow out your highlights or create dark shadows in PP that's fine, but in your original image, always shoot for the maximum detail.

The guys who handed in those K-5 and K-30 images, they get a failing grade in any course I ever took, or taught.

You always want to check and make sure you aren't getting sucked in by someone else's incompetence. Imagine if I went to an editor with one of those images and tried to say the camera made me do it that way...

honest, there are a lots of areas in technical photography where an image like that would get you fired, and it has nothing to do with the limitations of the K-5 or K-30.

Last edited by normhead; 10-19-2012 at 10:07 AM.
10-19-2012, 09:45 AM   #83
Veteran Member




Join Date: Sep 2007
Location: USA
Posts: 1,812
is this any better?


So if 24Mp APS-C may be unconvincing -
how convincing is no anti-aliasing filter K-5IIs going to be over the plain K-5II which does not have as much risk of aliasing and moire?

EDIT to ADD - I opened in my photo editor the first full faced comparison image with the K-30 in Post #79 - and did a quick analysis concentrating on the K-30 "blasted" cheek area - none of the RGB levels exceeded 230, mostly it was down around 200 (max level is 255) - so perhaps the pics were not over-exposed?



Last edited by UnknownVT; 10-19-2012 at 09:58 AM.
10-19-2012, 09:48 AM   #84
Veteran Member
westmill's Avatar

Join Date: Jan 2012
Location: Stoke on Trent
Photos: Gallery | Albums
Posts: 1,146
The way I see it is that quality is determined by the lens. What you get with more pixels is bigger file sizes. Prints would need to be huge before you could spot the difference here though. Its obvious that all things being equal a 24 milion pixel sensor will resolve more detail in a large pic than a 16 milion pixel, but it would need to be very big for there to be any genuine difference. Its a balancing act for most though. We are talking and looking at base ISO here. Things change considerably when the ISOs start to go up. As for the K52s not being capable of being as sharp as a 24m pixel APSC... I would not bet on it. The fuji pro 1 plays very close to the new cannon 5 which is a new 24m pixel full frame. Cameras like the D800 are capable of producing prints the size of billboards ! In truth, very few photographers need more than 12 milion pixels. Those that do are lucky indeed lol
10-19-2012, 09:50 AM   #85
Pentaxian




Join Date: Oct 2011
Location: Savannah, GA
Posts: 4,989
QuoteOriginally posted by UnknownVT Quote
I think the images are clearly labeled - this was simply a screen shot from the referenced/linked dpReview.com page - I did nothing special. They are the converted RAW test shots - as described on that page.

OK - there's the area under the eyebrows that has cross hatching - it is very clear in both the 24Mp APS-C images whereas they are hardly resolved either of the 16Mp K-30 or K-5.

Then there's the vertical watermark-like lines crossing the face - again both 24Mp APS-C images show quite clearly - whereas they are hardly visible on the 16Mp.

There is a difference between sharpness and resolution/detail as I tried to illustrate in Post #77 including the quote from dpReview, and I said so too that the D3200 image was soft. But one can see more detail as described above.

I live on earth, where civil discussion, and even disagreement
does not need to resort to rudeness.
If you download the RAW files from dpreview and examine them yourself and adjust the exposure/contrast etc you can see those details are all there, although there are more resolved in the 24mp sample. I think the difference of removing the AA filter makes up some ground there. It isn't just about more detail -- the AA filter makes everything mushy at the pixel level -- the AA-less samples I've seen (which is very few at this point, so I could be wrong) maintain crispness down the pixel level, so if moire isn't a problem I think large prints from an AA-less camera are going to look a little nicer even if the 24MP does have some more detail. Or, printed at same size they will be about the same, which is good enough. If I had to choose between a 24MP Pentax body with AA filter and a 16MP AA-less body that had all the same features otherwise and were the same price, well then I'd have some thinking to do...
10-19-2012, 10:05 AM   #86
Veteran Member




Join Date: Sep 2007
Location: USA
Posts: 1,812
QuoteOriginally posted by vonBaloney Quote
you can see those details are all there, although there are more resolved in the 24mp sample.
Thank you for the confirmation -
that was all I was really trying to say -
to get more detail it is probably easier to just go to a higher resolution sensor.

There might be more reasons for not having an anti-aliasing filter -
but I think for most it is for apparently more detail -
hence my suggestion - that's all it was -
not any direct comparison of cameras. not a suggestion that the Nikon D3200 or Sony SLT-A65 is better than the (name any camera), nor even a suggestion to buy a Nikon D3200 or Sony SLT-A65 over the K-5II or K-5IIs.
10-19-2012, 10:29 AM   #87
Pentaxian




Join Date: Oct 2011
Location: Savannah, GA
Posts: 4,989
QuoteOriginally posted by UnknownVT Quote
Thank you for the confirmation -
that was all I was really trying to say -
to get more detail it is probably easier to just go to a higher resolution sensor.

There might be more reasons for not having an anti-aliasing filter -
but I think for most it is for apparently more detail -
hence my suggestion - that's all it was -
not any direct comparison of cameras. not a suggestion that the Nikon D3200 or Sony SLT-A65 is better than the (name any camera), nor even a suggestion to buy a Nikon D3200 or Sony SLT-A65 over the K-5II or K-5IIs.
Except that breaks down your argument that it would be "easier to just go to a higher resolution sensor". It is not easy at all if the choices that would make it easy are not there. The easy thing for me to do is get a KIIs...
10-19-2012, 10:48 AM   #88
Pentaxian
normhead's Avatar

Join Date: Jun 2007
Location: Near Algonquin Park
Photos: Gallery | Albums
Posts: 23,765
If I was right and the K-5 images were blown, which is certainly to be expected looking that the exposure value in the EXIF of the various images, the K-5 and K-30 image are exposed at a wider aperture than the D3200 images were, you'd expect the Pentaxes to perform better in the darker areas of the image...

And viola...



The Pentax images shows the same amount of detail, but looks sharper. Why the authors chose over expose the Pentax images is unclear, but it's clear from examining the small images, examining the exif and examining the finale images that they did. You simply cannot compare detail at different exposure values and expect to make the kinds of conclusions you guys are making.
10-19-2012, 10:51 AM   #89
Pentaxian
kh1234567890's Avatar

Join Date: Sep 2010
Location: Manchester, UK
Photos: Gallery
Posts: 2,327
Of course all this assumes that you'll never want to crop. Probably fine if you shoot still life and can spend time framing your shots properly, otherwise having the extra pixels does help. Perhaps people do not crop anymore ...
10-19-2012, 11:02 AM   #90
Pentaxian
normhead's Avatar

Join Date: Jun 2007
Location: Near Algonquin Park
Photos: Gallery | Albums
Posts: 23,765
I can't believe I spent this amount of time arguing a point with folks who don't seem to understand the nature of a blown highlight. Time to go take some pictures, the dogs are waiting.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
camera, dslr, lightroom, model, photography, pp
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
K5IIs sample? pinholecam Pentax News and Rumors 349 12-20-2012 12:00 PM
Official K5II and K5IIs sample is out with DA * lens PentaxSX Pentax News and Rumors 187 11-21-2012 07:03 PM
K5II and K5IIs samples scunning14 Pentax K-5 5 10-02-2012 03:42 PM
10 things to know about the K5II/K5IIs jpzk Pentax K-5 16 09-24-2012 05:33 AM
Would you risk buying a brand new K5 without warranty? vinceloc Pentax K-5 17 08-15-2012 03:14 PM



All times are GMT -7. The time now is 04:23 PM. | See also: NikonForums.com, part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top