[stevebrot]

I'd like to see the test images from photozone so I can see for myself

I think you might be disappointed. He does not use a standard resolution target.

Photozone Lens Test FAQ


Steve

---------- Post added 10-10-14 at 01:17 PM ----------

16 megapixels is enough if i don't have to crop it to much. 24 is enough for me.

I usually frame pretty closely, but to get the combination of perspective and angle that I needed for the photo below would have involved a dive off a cliff (hard to explain). Instead, I ended up wasting about 35% of the original frame. Final pixel dimensions were 4858 x 3230




Steve

(...yeah, I know the foreground focus is off. AF was no help here and my viewfinder was fogging up...)

---

Last edited by stevebrot; 10-10-2014 at 01:29 PM.

[emalvick]

I am one of those that believe 16 MP is adequate only because I've never printed big enough and probably won't.

YET, I purchased a K-3 and will probably take advantage of the added cropping room assuming I'm satisfied with the resolution when cropping is done. The reality though is that I wanted to upgrade from my original K5 and felt better buying the K3 with the freebies and its other advantages over a K5iis.

Ideally for me, they would have released a 16MP K3 (perhaps a K5iii). I know someone said it is not likely, but I would assume that it could be as a mid-to-upper level dSLR rather than a flagship level camera. If I remember, they kind of did something like that with the K200D, which was a lesser version of the K20d but the specs were essentially a K10d with improvements. Oh well... I think I'll be happy with the K-3 and as per my usual, most of my photos will end up processed to 6 to 10 MP and look fantastic.

[stevebrot]

If I remember, they kind of did something like that with the K200D

Yes, and the K200D was soundly mourned when it passed. It truly hit the value sweet spot for the time. Now if it had only had both front and rear control dials.


Steve

[dadipentak]

my viewfinder was fogging up

That's a seriously fun shot.

[Nicolas06]

I just processed some shoot I took in Italy from Capri. Most shoot reframing is small or there no reframing at all. But out of the 15-16 final shoots I decided to keep, 2 are heavily cropped !

Following is a practical example of this. Please note my narrowest lense available I took this day was FA77 and I took the picture with it For sure the initial framing is not good and I'am the one to blame, but to really target the background like the crop would have required me to take the 50-135 and to shoot 135mm... My fault I didn't want to carry it...

So I came from 24MP to 9MP (4241*2121).

But sometime that much more conveniant to do it this way. I mean technology is there, why not use it?

[dadipentak]

But sometime that much more conveniant to do it this way. I mean technology is there, why not use it?

I couldn't agree more!

[RonHendriks1966]

I just processed some shoot I took in Italy from Capri. Most shoot reframing is small or there no reframing at all. But out of the 15-16 final shoots I decided to keep, 2 are heavily cropped !

Following is a practical example of this. Please note my narrowest lense available I took this day was FA77 and I took the picture with it For sure the initial framing is not good and I'am the one to blame, but to really target the background like the crop would have required me to take the 50-135 and to shoot 135mm... My fault I didn't want to carry it...

So I came from 24MP to 9MP (4241*2121).

But sometime that much more conveniant to do it this way. I mean technology is there, why not use it?

That is very nicely a little smaller then Olympus'ses micro4/3th sensorsize. Why not use that technology since it's there?

I know this made your image, but with K-01 you would have 3474*1737 pixels, wich is still a good size with 6 megapixel and you can still print a nice image (if you print at all).

[Nicolas06]

That is very nicely a little smaller then Olympus'ses micro4/3th sensorsize. Why not use that technology since it's there?

I know this made your image, but with K-01 you would have 3474*1737 pixels, wich is still a good size with 6 megapixel and you can still print a nice image (if you print at all).

I do happen to have a K3, so I use my K3 I have nothing against micro 4/3 but I do not want to buy another camera, an adapter and put my FA77, loose autofocus just for that.

If I replace K3 by a micro 4:3, then if it also 24MP, the situation in term of quality ect that some have is even worse. You loose 1EV, it doesn't cost that much less. Don't know for size...

As for it would have still 2474x1737 pixel available and that more than enough... In theory. If thoses pixels where perfect yes. But it is not, there low pass filter, there interpolation of colors...

The same kind of crop, many time was a bit too much with K5 and started to really show lack of sharpness where it is still ok with K3.

This come you know from the actual resolution you get in practice, not the theoretical resolution. Maybe DxO is not good, but it show it. A lens get 8-9MP effective on their tests with K5, 10-12 with K3. If you count this actual resolution, it not 6MP worth of detail you really get, but 2MP. While the K3 would get 2.5-3MP depending of the lens.

That not much in many case, but as such average resolution the difference will show in prints and from my practical experience on screen too. And I have only a 2MP full HD screen, 4K 8MP screen will soon become the standard.

If you count the effective resolution of 8MP a K5 or K-01 typically give you also found you don't need thoses A1 or 40" shoot to start to see the difference, an A2 20" will already clearly show it without crop, an A4 shoot or a high resolution screen will show it if you crop.

[RonHendriks1966]

As for it would have still 2474x1737 pixel available and that more than enough... In theory. If thoses pixels where perfect yes. But it is not, there low pass filter, there interpolation of colors...

The same kind of crop, many time was a bit too much with K5 and started to really show lack of sharpness where it is still ok with K3.

This come you know from the actual resolution you get in practice, not the theoretical resolution. Maybe DxO is not good, but it show it. A lens get 8-9MP effective on their tests with K5, 10-12 with K3. If you count this actual resolution, it not 6MP worth of detail you really get, but 2MP. While the K3 would get 2.5-3MP depending of the lens.

I don't know about those figures of only having 2 megapixels.

I have an A3 print made from K-5 and DA*300mm, shot at iso5000 and it comes from only 1762x2467 pixels. So yes that is pushing to the limits, but stil an amazing image. It was on my exhibition and earlier this year it even was on TV.



So there is a lot possible. But yes using a bigger part of the sensor would be great. In this it would be great if I had 600mm, or even the 1.4xTC.

[Nicolas06]

I don't know about those figures of only having 2 megapixels.

I have an A3 print made from K-5 and DA*300mm, shot at iso5000 and it comes from only 1762x2467 pixels. So yes that is pushing to the limits, but stil an amazing image. It was on my exhibition and earlier this year it even was on TV.



So there is a lot possible. But yes using a bigger part of the sensor would be great. In this it would be great if I had 600mm, or even the 1.4xTC.

This work well because the subject is contrasty and the noise processing did an awesome job. That a very nice shoot.

If I want to look a little more into it, there no much real detail in the photo. The border of subjects are oversharpened, dithering of color tones are very visible and it look a little like posterized.

It still look good, because you spent time trying to get the max out of it. But it would have looked nicer if it was possible to shoot it at ISO 100 and with more pixels. Obviouly. I doubt the K3 would have helped that much here, maybe far less need to sharpen it, but maybe more need for color/noise removal...

[Aku Ankka]

If they built the K5IIs with only 12Mp it would have outperformed the D800 in low light. I think 16Mp is the most I will go with APS-C. 24Mp on APS-C is stupid. Pentax was playing the "me-too" game.

A full frame at the same exposure settings (that is aperture number, shuter speed and light) will collect about 2.25 times more light than an APS-C sensor. Image noise is almost entirely a function on the noisy nature of light itself. If you capture twice the light, you have about 1.41 times better signal-to-noise ratio (SNR).

Pixel size is almost irrelevant.

[RonHendriks1966]

This work well because the subject is contrasty and the noise processing did an awesome job. That a very nice shoot.

If I want to look a little more into it, there no much real detail in the photo. The border of subjects are oversharpened, dithering of color tones are very visible and it look a little like posterized.

It still look good, because you spent time trying to get the max out of it. But it would have looked nicer if it was possible to shoot it at ISO 100 and with more pixels. Obviouly. I doubt the K3 would have helped that much here, maybe far less need to sharpen it, but maybe more need for color/noise removal...

Part of the oversharpening is from my export to the little jpg. In print this isn't so prominent. Ofcourse there is less detail, but there is a lot to see in the image. So no real fine detail, but all Essentials are in the image. I don't think that the K-3 would be much better in this situation. Having 600mm would bring more detail, as is with having more light to work with.

[altopiet]

If I want to look a little more into it, there no much real detail in the photo. The border of subjects are oversharpened, dithering of color tones are very visible and it look a little like posterized.

I'm really in awe of peoples abilities to see such fine detail....it inspires me to use more of my time studying images in more detail, to see the trees, rather than the forest.....Till now I’ve been looking at images much too superficially.

[normhead]

Pixel size is almost irrelevant.

Do you have a source for that? Because one would suspect that the size of pixels would help average out the spikes and valleys in the available light, thus reducing visible noise. So, I'm afraid for myself i wouldn't be able to accept a statement like that without seeing some kind of empirical data. I can see it might be true that larger pixels make a difference, I can also see it being true that larger sensor and more available light being the deciding factor. I'm nt even sure how you determine this without research. SO where's the research?

[Pioneer]

I'm really in awe of peoples abilities to see such fine detail....it inspires me to use more of my time studying images in more detail, to see the trees, rather than the forest.....Till now I’ve been looking at images much too superficially.

I agree. I spend far too much time looking at the actual photograph thinking how nice it is. Obviously I should be spending more time delving deep into it to see those oversharpened edges.

Extra pixels are always nice and the technology does permit cropping out little slices from larger photographs. But I still find some of my better digital photographs have come out of the outdated sensors in some of my older 6 and 10 mp Pentax cameras. Once I started putting my K-01 through the paces I realized I probably had found one the best cameras I could have dreamed up. It uses lenses that I am very familiar with, and does so in a form factor that feels almost perfect. The only things that would have made it better is if they could somehow have shoe-horned a viewfinder into the camera as it was, and included a full frame sensor.

I know that at some point in the future somebody will introduce another camera that triggers my YES button, but so far I am very happy with the K-01. Maybe it will end up being a long time keeper like my DL2 and K10D.