[dtmateojr]

But if you were given the choice: K30 @ 16MP or K30 @ 24MP, same price, maybe slightly lower FPS for the 24MP version, what would you choose?

Had they retained the 16mp sensor for the K3 I would have already bought it. That was my fear when they announced the K3; that Pentax might attempt to chase the stupid Nikon D5xxx...and they did The K5 IIs was almost perfect! For people who just want to take photos, improvements in AF tracking was all that was needed for its successor.

Anyway, what's done is done. If Pentax does go FF I hope they stick with 24mp so at least I might be interested. Anything more than that and I won't even test it. I have not touched a K3 yet.

[Rondec]

50MP on aps-c (!) would require higher shutter speeds than we're used to using for handheld if you want absolutely pixel-sharp results, but it will never look any worse than lower-MP in any case and may often look better even with some acuity lost to blur. Consider this comparison, taken from a dpreview thread on this:

I don't disagree, but I do know that pixel peepers can get awfully stirred up, when they see blurred pixels on a high density sensor. I just know that if you don't have good technique you get little to no benefit going from a K5 to a K3.

---------- Post added 02-28-14 at 04:58 PM ----------

Had they retained the 16mp sensor for the K3 I would have already bought it. That was my fear when they announced the K3; that Pentax might attempt to chase the stupid Nikon D5xxx...and they did The K5 IIs was almost perfect! For people who just want to take photos, improvements in AF tracking was all that was needed for its successor.

Anyway, what's done is done. If Pentax does go FF I hope they stick with 24mp so at least I might be interested. Anything more than that and I won't even test it. I have not touched a K3 yet.

There is no particular down side to the K3, though. It is faster in every respect than the K5, has better resolution, equivalent high iso shooting.

[jsherman999]

There is no particular down side to the K3, though. It is faster in every respect than the K5, has better resolution, equivalent high iso shooting.

+1, unless you're working off some incorrect assumptions about diffraction, or noise, or resolution, the K3 is the better camera in most ways. The K5IIs can be had for a song now, though, so I'd be hard-pressed to decide if I were in the market for an aps-c dslr right now.

[dtmateojr]

+1, unless you're working off some incorrect assumptions about diffraction, or noise, or resolution, the K3 is the better camera in most ways. The K5IIs can be had for a song now, though, so I'd be hard-pressed to decide if I were in the market for an aps-c dslr right now.

What incorrect assumptions about diffraction? What is it about diffraction that you hopelessly fail to understand?

[civiletti]

I am not sure that the extra resolution of the K-3 is worth losing the greater dynamic range of the K-5 series for landscape imaging. Remember though, that the K-5 16mp is equivalent to the D800 36mp. The 24mp K-3 would be equivalent to 50-something MP in full frame. In the 1:1 comparisons I have seen the extra detail from the K-3 is marginal. The difference from 24mp to 36 mp in full frame would not be marginal.


In any event my next body for landscape will be full frame, whether it is Pentax, Sony or Nikon - unless the new D645 is a killer at $10k or less.

[jsherman999]

What incorrect assumptions about diffraction? What is it about diffraction that you hopelessly fail to understand?

I think it's you who's still working on his homework assignment.

.

---------- Post added 02-28-14 at 04:46 PM ----------

In any event my next body for landscape will be full frame, whether it is Pentax, Sony or Nikon - unless the new D645 is a killer at $10k or less.

Always factor in the lens cost with MF. I don't know what the used market is like there, though, could be easy on the $.

[normhead]

I think the point is that 16 Mp on APS-c is probably the best compromise using current technology.Obviously, 16 Mp on FF can achieve amazing low light performance. Don't get me rong, I love my K-3 for what I bought it for, more resoltion for cropping and a better AF system, better in some ways than Canon or Nikon APS-c. But to date, for landscape I still see the k-5 !!s as the ultimate Pentax, until the 30 Mp FF is introduce.

As the pixel size goes down, the diffraction limit goes up. Theres no avoiding that. The K-3 to me may be the best compromise you can get with current technology. If Pentax could scale the k-1 back to 20 Mp, I wouldn't complain.

[dtmateojr]

I think it's you who's still working on his homework assignment.



.

---------- Post added 02-28-14 at 04:46 PM ----------





Always factor in the lens cost with MF. I don't know what the used market is like there, though, could be easy on the $.

You really are hopeless. Just because you don't believe in gravity does not mean it won't kill you. Anyway, carry on with your superstitious beliefs and may the megapickles bless you. LOL!

[jsherman999]

You really are hopeless. Just because you don't believe in gravity does not mean it won't kill you. Anyway, carry on with your superstitious beliefs and may the megapickles bless you. LOL!

Since you're responding, care to elaborate on your claim that at f/11 and smaller there's no resolution benefit on FF to the 36MP with any lens? Or how Roger Cicala obviously doesn't know how to test lenses, because his results prove you wrong? (did he get back to you on that?)

.

[dtmateojr]

Since you're responding, care to elaborate on your claim that at f/11 and smaller there's no resolution benefit on FF to the 36MP with any lens? Or how Roger Cicala obviously doesn't know how to test lenses, because his results prove you wrong? (did he get back to you on that?)

.

Let's make this really easy for you.

Supposing that your image has N lines to resolve. Unless you don't believe in the physics of light, diffraction will present only M lines to your sensor. M is always less than N. There is no way in hell you can resolve N when you only see M. No way.

At f5.6 for a 36mp FF, N = M. At apertures smaller than f5.6, you will see that N is reduced to M. At f11, M is only half of N...unless you have photoshop kung-fu like those jokers LOL!!! Sorry, I still can't get over that thing. It was just too funny.

And, no he hasn't replied to me. I still think his methodology was wrong. Let me explain:

f11 is capable of resolving say 50lpmm. Same as any aperture wider than that. So if all you have is 50lpmm then plotting the resolution of f5.6 to f11 will look flat. Very very basic. Don't tell me I still have to explain graphing to you. Anyway, his sample was only half of the max res of a 36mp sensor that's why he is not seeing the full potential of the lens. 1000 lp/ih for a 50mm lens is pathetic. It's either he had a very bad lens or a bad image sample.

---------- Post added 02-28-14 at 04:28 PM ----------

Here's another example: Car headlights. If the car is very very far, you won't see two headlights but only ONE!!! Diffraction is like that. Light spreads such that adjacent light sources appear to merge. Now tell me if your 36mp camera can see two separate headlights better than my iphone in that scenario.

ROFL!!!

[jsherman999]

Let's make this really easy for you.

Supposing that your image has N lines to resolve. Unless you don't believe in the physics of light, diffraction will present only M lines to your sensor. M is always less than N. There is no way in hell you can resolve N when you only see M. No way.

At f5.6 for a 36mp FF, N = M. At apertures smaller than f5.6, you will see that N is reduced to M. At f11, M is only half of N...unless you have photoshop kung-fu like those jokers LOL!!! Sorry, I still can't get over that thing. It was just too funny.

And, no he hasn't replied to me. I still think his methodology was wrong. Let me explain:

f11 is capable of resolving say 50lpmm. Same as any aperture wider than that. So if all you have is 50lpmm then plotting the resolution of f5.6 to f11 will look flat. Very very basic. Don't tell me I still have to explain graphing to you. Anyway, his sample was only half of the max res of a 36mp sensor that's why he is not seeing the full potential of the lens. 1000 lp/ih for a 50mm lens is pathetic. It's either he had a very bad lens or a bad image sample.

He keeps his scale at a point where he can compare lenses across formats easier, and it's easy enough to convert his line pairs (one black and one white) to line widths (the width of either the black or white) by simply doubling. Same with mm versus image height.. And the point is that the curves will not meet at f/11 in that comparison even if he changes the scale, because he would be changing the scale for both the lower-MP sensor and the 36MP sensor.

As you can see, right at f/11 the 36MP D800 is resolving about 870l, the 24MP D3x around 780l. It sounds like you're trying to say that if the scales below went to 2000LP, instead of 1300, the curves on the 36MP chart would magically change so that they would meet the 24MP curve at f/11:



You keep asserting that everyone's measured data on this is wrong, and your particular problem with lensrentals data seems to be concentrated on his LP/IH scale, now.

.... unless you have photoshop kung-fu like those jokers LOL!!!

Yes, let's get back to this point. Are you claiming now that sharpening does not affect MTF? Or are you claiming that it does, just not to the extent that they show?

.

[dtmateojr]

it affects MTF but can never exceed diffraction limit. No way.

I understand that he is using the same image sample for different formats but you are pushing me to believe that his graph is an accurate representation of *relative* resolution between apertures. It's not. No way near it.

[ElJamoquio]

There is no particular down side to the K3, though. It is faster in every respect than the K5, has better resolution, equivalent high iso shooting.

Exactly, the K-3 is better in every way (except a minor difference in size that might be better or worse).

There is no disadvantage to the K-3, and since we're still well below the point where diffraction becomes increasingly critical, there's room to explore on the top side, too. Newbies can safely ignore those who say otherwise.

If you don't crop or print large, there's not a ton of upside to 24MP on APS-C... but under no circumstances i there a downside to the K-3 over the K-5.

[mecrox]

Exactly, the K-3 is better in every way (except a minor difference in size that might be better or worse).

There is no disadvantage to the K-3, and since we're still well below the point where diffraction becomes increasingly critical, there's room to explore on the top side, too. Newbies can safely ignore those who say otherwise.

If you don't crop or print large, there's not a ton of upside to 24MP on APS-C... but under no circumstances i there a downside to the K-3 over the K-5.

Hmmn, my concern with 24 mpx on APS-C is street shooting. It's not focus precision, it's camera movement. Since shots are usually taken fast there often isn't time to get one's stance and bracing just right. One may even be moving while taking the shot. Now if I set the shutter fast enough, I don't have a problem with smearing on the K5 though I may well find that the focus is a little off. I normally disable IBIS anyway since the shutter speed is too high to make it necessary - typically 1/250th or 1/320th.

The question is whether the K3 will show the same characteristics or whether the K3's higher mpx will tend to produce loss of sharpness caused by poor technique (i.e., not being firmly held steady) and camera movement, and not only from imprecise focus (the camera is moving, perhaps, and so is the subject), at least compared to the K5.

Do you have any experience of this either way?

[dtmateojr]

The K5 already exhibits this blurring not just because of hand movement but even a simple mirror slap with the camera on a tripod. Ever since I got the K5 and now the K5 IIs, I have always used mirror lockup when shooting landscapes. That something that I never had to do with previous cameras. With a 24mp sensor it will only get worse. The K3 is not the best camera for precision landscape photography.