Pentax K-1 II vs K-1 Noise Performance
A hands-on comparison in RAW and JPEG mode
By PF Staff in Hands-On Tests on Apr 28, 2018
In this article, we continue our hands-on comparison of the image quality of the Pentax K-1 with the new Pentax K-1 II, with a closer look at noise performance (earlier this week, we tested the new hand-held Dynamic Pixel Shift Resolution mode).
While both cameras share the same 36-megapixel sensor and PRIME IV engine, the K-1 II also incorporates Pentax's Image Accelerator Unit, which improves signal-to-noise ratio at the time of capture, and thus promises less noisy files in both JPEG and RAW mode. Let's take a look at some photos!
First test scene
Our first test scene was shot at ISO 6400, 2s, and F4.5 with the D FA* 70-200mm F2.8 lens. Both cameras were set to RAW+, with auto white balance, "bright" custom image with default settings, no in-camera noise reduction, and no lens corrections.
We then developed both cameras' RAW files with the same lighting settings, no luminance noise reduction, and only baseline (25%) color noise reduction in Adobe Camera Raw. 100% crops follow.
Here we observe that the K-1 II indeed produces files with less noise straight out of camera. The Image Accelerator clearly has an effect on RAW files. At the same time, though, fine edge details are slightly clearer in the K-1's more noisy file, and perhaps the shadows have a hint more detail.
In JPEG mode the K-1 II is a clearer winner, with hardly any pixel-level noise despite the rather high ISO that we used (and no in-camera noise reduction). However, as we saw in the previous comparison, the K-1 does render fine details more clearly.
To verify these findings, we returned to the studio and shot the same test setup that we used for our earlier DPSR comparison. Note that we repeated all of the tests shown a second time in order to ensure that the experimental setup itself was sound.
Studio Test Scene
As in the outdoor test, both cameras were configured with the same settings: no noise reduction, no lens corrections, and RAW+. We used the D FA 50mm F2.8 macro at an aperture setting of F11, focused manually using live view with magnification.
Let's start with the raw files as before. The comparisons below span the ISO range common to the two cameras (i.e. 100 to 204800).
As a supplement to these unedited comparisons, we also ran the ISO 3200 raw files through the latest version of Photoshop CC in an effort to get the best possible result from each camera (within a reasonably short amount of time, of course).
The two cameras are very close even here, but subjectively we would say that the K-1 II has the advantage after all is said and done. The K-1 II also appears to produce more natural colors, and shows better clarity, but this may be due to a slight variation in white balance at the time of capture (and subsequently during development).
Now let's take a look at out-of-camera files in JPEG mode:
In JPEG mode the K-1 II has an edge, especially above ISO 12800. At lower ISO settings, as we saw in the outdoor test, there is clearly less noise, but details are also slightly softer.
The Pentax K-1 and K-1 II are both phenomenal cameras capable of producing clear and detailed images at high ISO settings. We were expecting the K-1 II to blow the K-1 out of the water with its new image processing unit, but it turns out that the cameras are actually not far removed from each other in raw mode. On the other hand, if you like shooting in JPEG mode and want good results with minimal post-processing, then the K-1 II delivers cleaner files with less visible noise. The nice thing about the K-1 II is that all this noise reduction is an inherent component of its sensor data pre-processing; it is not tied to any in-camera settings (such as the JPEG noise reduction setting).
At the end of the day, once post-processing processing is accounted for, our preliminary verdict is that the K-1 II only inches ahead of the K-1 in terms of image quality. Purists might not appreciate the fact that with noise reduction comes some loss in detail, but pixel-peeping is certainly needed to appreciate the differences. We would therefore say that the K-1 II's DPSR feature is by far more exciting than the incorporation of the image accelerator noise reduction unit from a practical perspective.
This of course begs the question of why the difference between the K-1 and K-1 II is not as dramatic as we had expected. After all, when we compared the K-70 (with the image accelerator unit) to the K-3, the K-70 fared much better at higher ISO settings. Our hypothesis is a simple one: the K-70 used a newer sensor than the K-3, and the generational improvements contributed to better noise performance alongside the image accelerator unit. Since the K-1 and K-1 II share the same sensor, the difference is more subtle.
Stay tuned for our in-depth review of the K-1 II, which will include a more comprehensive look at noise performance across a wider variety of scenarios, including photos at ISO 409600 and 819200. We also plan to compare the practical impact of DPSR versus traditional tripod shooting.
In the coming week, we will be taking a look at the improvements to the K-1 II's autofocus.
What do you think? All original files used in this article are available for you to download and experiment with. Do you prefer the look of the K-1 II image over the K-1?