[JohnBee]

The 24mp is slightly worse even downsampled. It basically doesn't have any high-ISO advantage, although you'd be stretching things to say it would be any noticeably worse at normal viewing sizes at very high-ISO, and would be better (more resolution) at base ISO..

Actually....

I think an ISO 12800 may be hard to substantiate due to the level of IQ degradation. Though the advantages the K-5 holds over the NEX-7(even resized) would be in the grain quality and chroma noise performance.

In my own tests, I've found ISO6400 was the sweet spot for the K-5, whereas the NEX-7 seemed to top-out at/or around ISO4000(resized) and chroma noise on the NEX would rear its ugly head as early as ISO3200. However, its on the grain quality that the K-5 really shines. ie. where the RAW files begin to suffer from detail degradation and NR recovery.

Having said that, I don't think the K-3 is going to bring a usable 102400 to the sensitivity range this year, though I have good reason to believe that Sony's new sensor will allow Pentax to deliver yet another APS-C to beat in 2012.

---

Last edited by JohnBee; 01-08-2012 at 07:06 PM.

[Anvh]

In any case, the jump from 16MP to 24MP is NOT a big change.

How is a 50% increase not a big change?

About minium pixel size there is a good article on that on limunous landscape, not sure what the number was but it depends on difraction of the lens.

[Anvh]

Actually....

I think 12800 may be a little misleading based on noise artifacts . However, the main advantages of the K-over and NEX-7(even resized) will be in the grain quality and chroma noise performance. In my tests, I've found ISO6400 to be the sweet spot for the K-5, whereas the NEX-7 seems to top-out at/or around ISO4000(resized). Though I must say, that the chroma noise on the NEX rears its ugly head as early as ISO3200.

You do know that above ISO1600 the K5 aplies noise reduction on the photos, also the RAWs

[JohnBee]

You do know that above ISO1600 the K5 aplies noise reduction on the photos, also the RAWs

Absolutely!
And that's what makes it so darned good

Pentax engineering has proven to be unsurpassed in this area, and this is why I believe they will make this sensor soar like none other(higher dividends).

[jsherman999]

Absolutely!
And that's what makes it so darned good

Pentax engineering has proven to be unsurpassed in this area, and this is why I believe they will make this sensor soar like none other(higher dividends).

Agreed, even though there is NR, and that's 'cheating'... the proof is in the pudding. Pentax's implementation threads the needle almost perfectly, reducing noise while retaining enough detail.

If the K-3 adopts this 24MP sensor, it will most likely surpass the NEX-7 - problem is, it probably won't surpass the K-5 with regards to noise at high ISOs. That's probably OK, each generation can't expect to make the leap the K-5 did.

[JohnBee]

If the K-3 adopts this 24MP sensor, it will most likely surpass the NEX-7 - problem is, it probably won't surpass the K-5 with regards to noise at high ISOs. That's probably OK, each generation can't expect to make the leap the K-5 did.

You know, I too used to think Pentax high ISO NR was somewhat of a hack. But after some extensive testing, I came to the conclusion that it was an ingenuous piece of engineering. I also think this was the secret sauce that allowed Pentax to not only claim the APS-C high ISO crown, but that also allowed the K-5 to shoot in FF territory as well. Having said that, despite my greatest efforts, I was never able to match or exceed Pentax's low level noise reduction quality, with even the best of the tools available today. And so hats off to them for pulling that off, and I can't wait to see if they've made improvements to this technology in their upcoming body!

On the topic of succession, I think the K-3 could potentially surpass the K-5 even if it were ever so slightly. ie. by looking-up the DxOMark scores for both NEX-7 and K-5, you'll find both systems to have very close SNR performance(40.78 vs 41). However... since Pentax has better signal noise(p/pixel quality) we can also expect improvements across the bar(SEE: color, tonal and DR etc). And so the K-3 could end-up being a major upgrade if we factor in the higher resolution and IQ improvements as a whole.

Of course this is all just speculation, and I'm honestly not sure if this will make the K-3 appealing enough for existing K-5 owners to want to upgrade. But if Pentax is really smart about it, I suppose they could deliver a body that is the envy of every APS-C shooter.

Here's to hoping

[ProfHankD]

How is a 50% increase not a big change?

About minium pixel size there is a good article on that on limunous landscape, not sure what the number was but it depends on difraction of the lens.

50% is not a big change for something that matters as the log of the square root. 4X pixel count isn't even a huge change. If you want another example where 50% isn't a big deal, look at computer processing speeds -- the machines on the Top500 fastest supercomputers get 2X faster every year, and have for decades.

As I said earlier, the "diffraction limit" isn't directly relevant given photon-counting sensors and/or computational photography processing.

[drougge]

But a 50% change in pixel count is in fact a 50% change in pixel count, which means 50% change in storage and processing needed. I'd much rather not have my camera that much slower. (And it will be. Not compared to some other camera, but compared to the same camera with less pixels.)

(Also my computer, no matter how fast my computer is, it would be handle smaller images faster.)

[RonHendriks1966]

Well the new Canon G1X made a new entry in the pixel war. Not by largely increasing the number of pixels, but by enlarging the sensor giving larger pixels with better image quality output.

[Ubuntu_user]

You know, I too used to think Pentax high ISO NR was somewhat of a hack. But after some extensive testing, I came to the conclusion that it was an ingenuous piece of engineering. I also think this was the secret sauce that allowed Pentax to not only claim the APS-C high ISO crown, but that also allowed the K-5 to shoot in FF territory as well. Having said that, despite my greatest efforts, I was never able to match or exceed Pentax's low level noise reduction quality, with even the best of the tools available today. And so hats off to them for pulling that off, and I can't wait to see if they've made improvements to this technology in their upcoming body!

On the topic of succession, I think the K-3 could potentially surpass the K-5 even if it were ever so slightly. ie. by looking-up the DxOMark scores for both NEX-7 and K-5, you'll find both systems to have very close SNR performance(40.78 vs 41). However... since Pentax has better signal noise(p/pixel quality) we can also expect improvements across the bar(SEE: color, tonal and DR etc). And so the K-3 could end-up being a major upgrade if we factor in the higher resolution and IQ improvements as a whole.

Of course this is all just speculation, and I'm honestly not sure if this will make the K-3 appealing enough for existing K-5 owners to want to upgrade. But if Pentax is really smart about it, I suppose they could deliver a body that is the envy of every APS-C shooter.

Here's to hoping

I'm on it with you – my thoughts exactly!

[Anvh]

Absolutely!
And that's what makes it so darned good

Pentax engineering has proven to be unsurpassed in this area, and this is why I believe they will make this sensor soar like none other(higher dividends).

I would like they didn't do it, the software i've at home is more powerful and i've far more controls, so when I set the noise reduction OFF it should be OFF all the time

[Anvh]

As I said earlier, the "diffraction limit" isn't directly relevant given photon-counting sensors and/or computational photography processing.

But what the point of having more pixels then any lens can resolve, it's a waist...

But a 50% change in pixel count is in fact a 50% change in pixel count, which means 50% change in storage and processing needed. I'd much rather not have my camera that much slower. (And it will be. Not compared to some other camera, but compared to the same camera with less pixels.)

(Also my computer, no matter how fast my computer is, it would be handle smaller images faster.)

You forget the processing also gets faster, that one actually goes faster then pixels count goes up.
Also with tablets and smart-phone becoming very popular the increase of the ARM power will also show hopefully.

[JohnBee]

I would like they didn't do it, the software i've at home is more powerful and i've far more controls, so when I set the noise reduction OFF it should be OFF all the time

Don't take this the wrong way, but I've spent months researching the issue and so I'd encourage anyone to try and see for themselves.

You can bypass the Pentax low level NR by limiting your sensitivities to 1600 or less, and manually push processing your images in post. At which point you can use your own NR to recovery detail etc.

And though I don't want to sound like a know-it-all, I will go as far as saying that I think you'll end-up being genuinely surprised at the results you'll find in the end.

[Geoff H]

Well the new Canon G1X made a new entry in the pixel war. Not by largely increasing the number of pixels, but by enlarging the sensor giving larger pixels with better image quality output.

Ooooh! - Shiney....

[ElJamoquio]

But a 50% change in pixel count is in fact a 50% change in pixel count, which means 50% change in storage and processing needed. I'd much rather not have my camera that much slower. (And it will be. Not compared to some other camera, but compared to the same camera with less pixels.)

(Also my computer, no matter how fast my computer is, it would be handle smaller images faster.)

(Infinitesimal)*4 = Infinitesimal, though.

In practice my camera write speeds matter but they aren't super-objectionable. On my CFD computer image processing is a cakewalk, on my netbook/laptop it's not-really-possible to load something in lightroom.