Web Analytics RiceHigh's Pentax Blog: Official K-5II/IIs Full Samples

Tuesday, October 02, 2012

Official K-5II/IIs Full Samples



http://www.pentax.jp/japan/products/k-5-2/ex/

Measurebate yourself but I don't think they are obviously better than the original K-5's. In particular, the IIs samples are not that "sharp" against what was expected.
Time (UTC +0800):
標籤： , , ,

Comments (29)

Loading... Logging you in...
Dashboard | Edit profile | Logout
  • Logged in as
-2 Vote up Vote down
Moritz Schwertner's avatar - Go to profile

Moritz Schwertner · 649 weeks ago

My K20D is much, much sharper then the K5 II s - as is my 5D Mark II. Especially how there are no fine details preserved is shocking in my opinion...
4 replies · active 649 weeks ago
+7 Vote up Vote down
Cynog's avatar

Cynog · 649 weeks ago

Are you blind? There's wonderful detail in the IIs shots
7 replies · active 649 weeks ago
0 Vote up Vote down
Jaz Johal's avatar

Jaz Johal · 649 weeks ago

Blinded by self
+2 Vote up Vote down
Leon's avatar

Leon · 649 weeks ago

As k5 shooter I can say that those samples are very detailed and sharp. Do not confuse sharpening with amount of details.
6 replies · active 648 weeks ago
0 Vote up Vote down
unoriginal_pics's avatar

unoriginal_pics · 649 weeks ago

"You know what would be cool? Let's fly 1/2 way across the world to Banff. No one's seen Banff photos before. I heard it's beautiful. These will be the most original test photos ever! ". I'd love to see that expense report.
+1 Vote up Vote down
Victor's avatar

Victor · 649 weeks ago

It makes no much sense to compare K5-II and K5-IIs based on different images. BTW, is it clear from the description how these jpeg samples were created? Are they provided by the in-camera image processor, or is it a raw converter that did the job?
+2 Vote up Vote down
Frank's avatar

Frank · 649 weeks ago

I don't think a landscape shot is a good reference, since the atmosphere can have a lot of effects epecially if we talk about sharpness, resolution and details... so that's not scientific, BTW: How can we expcet much improvement in these categories by using the same image sensor and image processor? (for the K5-II wihout S)
Even for the K5-IIs there should not be a WHOW effect if the anti-aliasing fiter of the K5 is well designed. i'm not a image sensor/optics expert but EE Engineer,, and you design any AA Filter so that it cuts off those stuff that your sensor can't handle. Of course no filter is a perfect one and so it has effect to the part your sensor can handle, but if it is well made the positive effects shall superimpose the negative ones.
0 Vote up Vote down
6X7's avatar

6X7 · 649 weeks ago

Hi Frank,

I'm an EE engineer too and as surely you know the Nyquist_frequency. The filter is a minimal a factor 2 stronger than the sensor resolution, So you really through away resolution to overcome sampling problems!

6X7
0 Vote up Vote down
jacok-x's avatar

jacok-x · 649 weeks ago

not even close to the latest CSC (NEX-7, OM-D, X1-pro) WITH KIT LENSES.
There some softness in these pictures....
0 Vote up Vote down
Frank's avatar

Frank · 649 weeks ago

6x7 you are right, the sensor can only represent "frequenies" which are half of his resolution. If we just simplifiy and think of one line of the picture and black and white and have a sensor resolution of 10px per mm we can only resolve a pattern of 5px per mm. Which is half of the sample frequency. If the pattern now increases to 6px per mm there will be some low frequency pattern visible at the sensor output. Now the AA filter shall cut off at half of the sensors resolution 'frequency' to avoid the slow frequency pattern which would be a light / dark modulation of the original pattern. Of course it will supress hard contrasty transitions cause they consist of high frequencies too. But the sensor can not represent more resolution if you take off the aa filter as long as it is not synchronized with the source pattern, which isnt the case usually. So there cant be any advantage by signals that are higher than fs/2 or we cant resolve better at patterns whith more pixels per mm than half of the sensors resolution. I dont know how optical filters work, but i think they will not be perfect filters, so they will have effecet on structures below fs/2 and decrease contrast/detail resolution. These effects will be removed by removing the AA filter. Image shall be more contrasy then, and slightly sharper cause hard edges will look better due to the removed aa. But if there are harsh transitions or high frequency patterns , they will look more sharper because of contrast behaviour. Also i think you can not remove them with sw i think. But i admit i have no idea about aa filters work. A commn technique is to put sampling frequency much higher. Tbc
2 replies · active 649 weeks ago

Post a new comment

Comments by

Subscribe to: Post Comments (Atom)