Heute kam der Test raus: https://www.dpreview.com/reviews/pentax-k-1-ii-review
Das schlimmste und überraschendste vorweg:
Mit der K-1 II kommt jetzt eine Zwangsentrauschung auf RAWs, auch bei tieferen ISOs, die sich nicht abschalten lässt (insbes. auch nicht von der Verschlusszeit abhängig ist).
Diese Zwangsentrauschung ist offenbar relativ simpel, sie matscht leider Details weg, auch bei tieferen ISOs.
Hier einige der Key-Messages in Kürze als Zitate:
(Accelerator-Chip)
"Unfortunately, our testing reveals this additional processor applies full-time noise reduction to Raw files resulting in inferior image quality to that of its predecessor."
"it's actually just forcing noise reduction in Raw. It can't be turned off."
"Baked-in noise reduction results in Raw files with progressively less detail than the original K-1 as the ISO increases. This can not be turned off."
"as ISO climbs, the K-1 II's detail capture begins to slip as a result of progressively stronger noise reduction being applied to Raw files by the pre-processor. To make matters worse, this forced noise reduction cannot be turned off."
"Shadows of higher ISO files can have some artifacts and loss of detail due to noise reduction and filtering of the Raw."
"The K-1 II does not have a dual-gain architecture for increased high ISO dynamic range, unlike some of its peers."
Schattenanhebung:
"some artifacts - visible as broken up text in the pushed crop - due to some filtering of deep shadows. This is not evident in the original K-1."
"possibly some form of noise reduction"
"The filtering is irreversible and can leave behind artifacts and reduce contrast and detail"
"It's unfortunate that this filtering cannot be turned off, as destructive processes should be saved for post-processing.
"Even the pushed ISO 100 shot from the K-1 II appears to have slightly less contrast than a similar shot from the K-1, likely due to some filtering even at base ISO."
"Importantly, the K-1 II does not have dual gain architecture, like most recent cameras from Nikon and Sony, to boost high ISO performance. Instead, Pentax appears to have chosen mandated noise reduction to give the appearance of better high ISO performance. We feel that noise reduction should be a choice left up to Raw shooters."
(Handheld Pixel-Shift)
"Pixel Shift mode [...] causes unwelcome artifacts"
"results in some jagged artifacts"
"Instead of using this mode, we recommend sticking to traditional pixel shift or simply creating your own super resolution photo in post using four image files."
"prone to subtle artifacts"
Wurde auch schon in deren Vorabartikel zu gesagt.
https://www.dpreview.com/articles/7242776989/pentax-k-1-mark-ii-dynamic-pixel-shift-mode
"Calling this mode 'Pixel Shift' is a bit of a stretch"
"is more like the multi-frame image stacking methods smartphones use"
"comes at the cost of having random artifacts"
"dotted artifacts in (seemingly) random portions of the image"
"the potential for artifacts limits its appeal"
(Bildqualität allgemein)
"the K-1 actually offers better image quality than the K-1 II"
(Autofokus)
"Continuous AF performance still unreliable overall"
"Limited AF point coverage"
"Tracking still not as reliable as competition"
"AF subject tracking shows improvements over the K-1, but still lags in reliability behind similar models from Sony and Nikon."
"uses the same 33 point AF system as the K-1. AF Points are concentrated toward the center of the frame. By 2018 standards AF point coverage is quite limited."
(Autofokus: Tiefen-Tracking)
"In short, we're not seeing a ton of improvement when it comes to depth tracking compared to the Pentax K-1."
"struggles to keep up with the movement of the cyclist."
(Autofokus: Breiten-Tracking)
"the hit rate still lags behind what we've come to expect from cameras in this class."
"missed shots again seem to be a result of focus not being able to keep up"
(Autofokus: Nahporträts)
"If you click into the gallery you'll notice four or five images with missed focus, or focus that is somewhat close to correct but not bang-on. It's also worth noting that initial acquisition and confirmation is relatively slow compared to the competition."
(Stabi)
"between 2.3 and 3.6 stops of added stability depending on focal length."
(Lahm)
"Slow max burst of 4.4 fps"
(Peoplefoto: garnicht erst Test-würdig)
"we have opted not to specifically address "Candid & Street," "Weddings & Events" and "Video".
(Overall Conclusion)
"In many aspects we regard as crucial when testing, like autofocus performance and video capabilities, it lags behind the competition."
"Unfortunately, the K-1 II has worse image quality than its predecessor.
This is due to forced noise reduction to Raw files resulting in less overall detail. At base ISO, the difference is negligible, but as the ISO increases detail loss becomes increasingly noticeable."
"For now, astrophotographers in particular may want to avoid this camera since the baked-in Raw noise reduction is likely to affect the rendition of starry skies at higher ISOs."
"our expectations [...] coupled with a disappointing drop in Raw image quality has resulted in the K-1 II having a lower score than its predecessor."
"for those wishing to point their camera at moving subjects, or wanting to capture decent video, a lot of other cameras would serve you better."
(versus A7 III)
"Sony Raw files [...] don't progressively lose detail as the ISO increases. Sony also has much nicer JPEGs. Performance-wise, the a7 III absolutely smokes the K-1 II, with far superior AF reliability and coverage, a faster burst rate and better battery life. The Sony also shoots excellent quality 4K video, compared to mediocre 1080p on the Pentax."
"Taken as a whole though, the a7 III is a much more well-rounded camera."
Score
Herabstufung von 84 auf 79 Punkte, das wäre früher glaub ich Silver Award gewesen.
Als K-1 I User würde ich also bei dieser bleiben. Und das Firmware-Update-Angebot nicht anrühren. Lieber auf eine K-1 III hoffen und warten, oder andere Konsequenzen ziehen, falls man die Nase voll und die Geduld verloren hat.
Das schlimmste und überraschendste vorweg:
Mit der K-1 II kommt jetzt eine Zwangsentrauschung auf RAWs, auch bei tieferen ISOs, die sich nicht abschalten lässt (insbes. auch nicht von der Verschlusszeit abhängig ist).
Diese Zwangsentrauschung ist offenbar relativ simpel, sie matscht leider Details weg, auch bei tieferen ISOs.
Hier einige der Key-Messages in Kürze als Zitate:
(Accelerator-Chip)
"Unfortunately, our testing reveals this additional processor applies full-time noise reduction to Raw files resulting in inferior image quality to that of its predecessor."
"it's actually just forcing noise reduction in Raw. It can't be turned off."
"Baked-in noise reduction results in Raw files with progressively less detail than the original K-1 as the ISO increases. This can not be turned off."
"as ISO climbs, the K-1 II's detail capture begins to slip as a result of progressively stronger noise reduction being applied to Raw files by the pre-processor. To make matters worse, this forced noise reduction cannot be turned off."
"Shadows of higher ISO files can have some artifacts and loss of detail due to noise reduction and filtering of the Raw."
"The K-1 II does not have a dual-gain architecture for increased high ISO dynamic range, unlike some of its peers."
Schattenanhebung:
"some artifacts - visible as broken up text in the pushed crop - due to some filtering of deep shadows. This is not evident in the original K-1."
"possibly some form of noise reduction"
"The filtering is irreversible and can leave behind artifacts and reduce contrast and detail"
"It's unfortunate that this filtering cannot be turned off, as destructive processes should be saved for post-processing.
"Even the pushed ISO 100 shot from the K-1 II appears to have slightly less contrast than a similar shot from the K-1, likely due to some filtering even at base ISO."
"Importantly, the K-1 II does not have dual gain architecture, like most recent cameras from Nikon and Sony, to boost high ISO performance. Instead, Pentax appears to have chosen mandated noise reduction to give the appearance of better high ISO performance. We feel that noise reduction should be a choice left up to Raw shooters."
(Handheld Pixel-Shift)
"Pixel Shift mode [...] causes unwelcome artifacts"
"results in some jagged artifacts"
"Instead of using this mode, we recommend sticking to traditional pixel shift or simply creating your own super resolution photo in post using four image files."
"prone to subtle artifacts"
Wurde auch schon in deren Vorabartikel zu gesagt.
https://www.dpreview.com/articles/7242776989/pentax-k-1-mark-ii-dynamic-pixel-shift-mode
"Calling this mode 'Pixel Shift' is a bit of a stretch"
"is more like the multi-frame image stacking methods smartphones use"
"comes at the cost of having random artifacts"
"dotted artifacts in (seemingly) random portions of the image"
"the potential for artifacts limits its appeal"
(Bildqualität allgemein)
"the K-1 actually offers better image quality than the K-1 II"
(Autofokus)
"Continuous AF performance still unreliable overall"
"Limited AF point coverage"
"Tracking still not as reliable as competition"
"AF subject tracking shows improvements over the K-1, but still lags in reliability behind similar models from Sony and Nikon."
"uses the same 33 point AF system as the K-1. AF Points are concentrated toward the center of the frame. By 2018 standards AF point coverage is quite limited."
(Autofokus: Tiefen-Tracking)
"In short, we're not seeing a ton of improvement when it comes to depth tracking compared to the Pentax K-1."
"struggles to keep up with the movement of the cyclist."
(Autofokus: Breiten-Tracking)
"the hit rate still lags behind what we've come to expect from cameras in this class."
"missed shots again seem to be a result of focus not being able to keep up"
(Autofokus: Nahporträts)
"If you click into the gallery you'll notice four or five images with missed focus, or focus that is somewhat close to correct but not bang-on. It's also worth noting that initial acquisition and confirmation is relatively slow compared to the competition."
(Stabi)
"between 2.3 and 3.6 stops of added stability depending on focal length."
(Lahm)
"Slow max burst of 4.4 fps"
(Peoplefoto: garnicht erst Test-würdig)
"we have opted not to specifically address "Candid & Street," "Weddings & Events" and "Video".
(Overall Conclusion)
"In many aspects we regard as crucial when testing, like autofocus performance and video capabilities, it lags behind the competition."
"Unfortunately, the K-1 II has worse image quality than its predecessor.
This is due to forced noise reduction to Raw files resulting in less overall detail. At base ISO, the difference is negligible, but as the ISO increases detail loss becomes increasingly noticeable."
"For now, astrophotographers in particular may want to avoid this camera since the baked-in Raw noise reduction is likely to affect the rendition of starry skies at higher ISOs."
"our expectations [...] coupled with a disappointing drop in Raw image quality has resulted in the K-1 II having a lower score than its predecessor."
"for those wishing to point their camera at moving subjects, or wanting to capture decent video, a lot of other cameras would serve you better."
(versus A7 III)
"Sony Raw files [...] don't progressively lose detail as the ISO increases. Sony also has much nicer JPEGs. Performance-wise, the a7 III absolutely smokes the K-1 II, with far superior AF reliability and coverage, a faster burst rate and better battery life. The Sony also shoots excellent quality 4K video, compared to mediocre 1080p on the Pentax."
"Taken as a whole though, the a7 III is a much more well-rounded camera."
Score
Herabstufung von 84 auf 79 Punkte, das wäre früher glaub ich Silver Award gewesen.
Als K-1 I User würde ich also bei dieser bleiben. Und das Firmware-Update-Angebot nicht anrühren. Lieber auf eine K-1 III hoffen und warten, oder andere Konsequenzen ziehen, falls man die Nase voll und die Geduld verloren hat.
Zuletzt bearbeitet:
DXO ist bei mir schon seit Jahren untendurch. Wie gesagt, die Pentax 645Z hat den besten Sensor bis vor kurzem und wurde nicht in deren Listing aufgenommen. Hätten sich ja welche dafür interessieren können. Und jetzt ist eben die Fuji GFX ganz groß. Abartig dieses Vorgehen. Dpreview verweist ständig auf Tests und Messungen von DXO. Wer die regelmäßig liest, weiß das.