[gazonk]

The K-3 does also bring other advantages - such as AF etc. But one could always resize the 24 mp images down to 16 mp to gain some noise reduction benefits over the K5II

If the K-3 gets the same DR as the D7100, it will score slightly lower than the K-5 IIs even when resized (the D7100 scores lower on the DxO "print" setting). For high ISO DR, I expect the K-3 to beat the D7100 just as the K-5 beats the D7000 (thanks to Pentax cooking the RAW?). And that means that the K-3 will have better high ISO DR than the D7100 (which is almost 1 Ev better than the D7000 at high ISOs).

[D1N0]

Only Full frames do better dat DR than the K-5 II The D600 does 14.2 and the D800 14.4 (as to 14.1 for the K-5 II). The D7000 does 13.9 so maybe Pentax can scrape the same DR out of a 24mp Toshiba.

[Caat]

If the K-3 gets the same DR as the D7100, it will score slightly lower than the K-5 IIs even when resized (the D7100 scores lower on the DxO "print" setting). .

But to be fair the difference is relatively minor

[BrianR]

I'm no expert on flash: I hardly ever use it. But I am puzzled to understand why a ND filter wouldn't have a similar effect. I'm obviously missing something owing to my ignorance of flash photography, but it is something I'd like to get more into.

It would have exactly the same effect on exposure assuming an appropriate ND filter. But given the choice of having to add a filter which will lower the quality (possibly imperceptible depending on the filter) or being able to select a lower ISO with a slightly higher quality (a little more DR) and not have to pull out a filter, it's not hard to choose between the two options. It's another case of more choices being nice to have with no drawbacks.

[dosdan]

People who do not need more mp but want greater dr will see a problem.

While I'm a fan of DR, more MP is handy when you need to crop. And particularly if you shoot action, you end up cropping a lot.

Just how much greater DR can you actually use in the real world with lens flare and veiling glare likely in very high contrast situations? Ever tried directly shooting a street lamp and nearby structures at night?

Of course, if the subject is static, you can get much greater DR with the raw HDR, while still having high MP.

Dan.

[banep]

There are some interesting marketing materials with tech explanations on this you tube channel but I don't understand japanese: pentaxplus - YouTube

[banep]

About my previous message: there is also english version now

[Rondec]

Honestly, the difference in dynamic range is about the same between the two cameras -- 0.4 EVs difference is certainly not enough that you wouldn't actually notice a difference in real life shooting. If you need wider dynamic range than 13 EVs, then you really need to take multiple exposures. Those who have shot with the D7100 have said that there is a real difference with regard to printing ability as compared to the K5 -- more details on the D7100. I think this will be a great landscape/wildlife camera and I don't see anything negative about it in comparison to the K5 II.

[TenZ.NL]

Just did a quick noise comparison (1600 and 3200) on DPR between the D7100, K5IIs and the K5 classic. The D7100 looked worse than the K5`s, I hope Pentax does a similair job as they did with the 16MP Sony sensor.

[Caat]

Just did a quick noise comparison (1600 and 3200) on DPR between the D7100, K5IIs and the K5 classic. The D7100 looked worse than the K5`s, I hope Pentax does a similair job as they did with the 16MP Sony sensor.

Worse at 100% or worse even when resized?

[TenZ.NL]

Worse at 100% or worse even when resized?

@ 100%

But have a look:
K5 IIs @ 1600
D7100 @ 1600

It is somewhat difficult to compare because of the bigger resolution. And also the K5IIs produces moire (look at the VW beetle) where the D7100 does not. But the image of the Nikon seems somewhat softer.

[zapp]

120g more brain mass. All the screws that I was missing in my K5. This is the most convincing argument for a K3.
Time to wait for real tests.

[Caat]

@ 100%

But have a look:
K5 IIs @ 1600
D7100 @ 1600

It is somewhat difficult to compare because of the bigger resolution. And also the K5IIs produces moire (look at the VW beetle) where the D7100 does not. But the image of the Nikon seems somewhat softer.

The colour noise seems a touch worse on the 7100 in general - although we will have to see what Pentax comes up with

[audiobomber]

I'll go by what DXO says; the K-5 IIs and D7100 are virtually equals for noise, DR and colour tone. Comparing at different resolutions is problematic, as is conversion with different softwares.

[dosdan]

Comparing at different resolutions is problematic

Yes. More MP on the same sized sensor means smaller sensels. For the same level of sensor technology, small sensels are individually noisier than large sensels. But if you were to print out the same sized output images, say A4 or 11"x8.5" sized prints, then more sensels reduce the overall noise level. So, if the fill factor is the same (how much of the sensor's surface is actually capable of generating a signal from the light, rather than "wasting" some of the sensor's area with connecting traces and other structural elements, both trends tend to cancel out. In the old days, when fill factors were low, more MP meant overall noisier images. But today, with very high fill factors due to efficient micro-lens designs, there's not much in it.

To test this assertion, compare the DxoMark SNR 18% graphs of the the D7100 vs the K-5II:

Compare cameras side by side - DxOMark

Look at the SNR 18% curves. These show the shot/photonic noise performance of the sensors, and are a good indicator of the sensor's QE (quantum efficiency). The "Print" tab shows the outputs normalised to the same printout size, and the two curves are very close. Now click on the "Screen" tab and you are comparing the same number of MPs (like 100% pixel peeping) and you'll see a difference in noise performance due to the smaller sensels: 3.9µm vs 4.75µm distance between the centre of each sensel.

It comes down to how you want to use the extra MPs. If you want to produce a proportionally larger printout with the the D7100, and look closely at it, then yes, it will appear noisier. But if you use the extra MPs to increase the fine detail in the same-sized printout, extra noise should not be an issue.

However, there is a flip side to this. While the extra MPs means you can crop an D7100 image more, for the same number of overall pixels in the resultant image, the D7100 image will look noisier than a K5II image.

Finally, there's an argument that, for the same overall image SNR, since smaller sensels produce smaller sized "grains" of noise, the noise is therefore more film-like and thus more aesthetically acceptable. This assumes that the two cameras have the same ratio of luminance noise to chroma noise. Many designers more heavily filter the chroma (colour) noise as it's more noticeable, and the loss of colour resolution caused by the filtering is unlikely to be noticed. The luminance (B&W) noise is less objectionable than chroma noise and heavily filtering it tends to produce a noticeable loss in resolution.

Dan.