Jump to content

New Hasselblad


Sandy Vongries

Recommended Posts

400 MP may be over the top, even for landscape photography, but in principle, you get more detail if you downsample than shoot at the final resolution. This is particularly true for video, which is low resolution to begin with. I shoot at 1080p but rarely publish at that resolution. I downsample to 720p, which produces smaller file sizes and the sampling process nearly eliminates aliasing and staircase effects in diagonal lines. Violin bows are particularly subject to stair casing. For stills, downsampling seems to preserve more details too, even to PNET size, which is only slightly less than 720p.
Link to comment
Share on other sites

The main advantage of pixel shifting is the elimination of color aliasing (due to color interpolation). At least four exposures must be merged, one for each color in the Bayer matrix - RGGB. The Sony A7Riii does this, but with a 42 MP output. Color aliasing is most pronounced with patterns of saturated color, which are fortunately rather rare in nature.

 

Science notwithstanding, Hasselblad can output an image with any number of pixels they choose. It could be that for selected subjects 100x6 images is equivalent to 400 MP. From experience, an 18 MP Leica M9, sans AA filter, equals or exceeds the acuity of a 24 MP sensor with an AA filter. Sigma invariably triples the MP specification for Fovix sensors, since each cell is stacked three deep. Spinning is not the exclusive property of politicians.

Link to comment
Share on other sites

Photoshop, Lightroom and related products do "smart" resampling, where blocks of pixels are analyzed to assess luminosity, color and details, then reduced to a single pixel at the target resolution. Continuity is maintained because adjacent pixels are always considered. Compression (e.g., JPG) uses similar techniques to cull redundant data while preserving as much detail as possible. In practice, downsampled images retain more detail than expected because of the intelligent process. Guitar and violin strings, for example, usually come out clearly, even though they may only be one or two pixels wide in the original. Faces remain sharp, even though other details may be lost.

 

Examples of "dumb" resampling can be found in DSLR video, at least in the early years, where pixels were "binned" or combined into blocks and averaged, or lines were skipped. Other techniques include cropping the sensor to APS-C size and using only that portion for video. Cameras like the latest Sony A7 and A9 use the entire sensor and resample to 2K (HD) or 4K intelligently.

Edited by Ed_Ingold
  • Like 1
Link to comment
Share on other sites

Same idea is used in the Olympus (EM1 mk II) and Panasonic (G9) m4/3 cameras. It presumably has some use for someone, but not all that useful for anything that is liable to move while it takes the shots. Any slight movement of foliage, trees, grass etc will remove much of the resolution advantage. But for studio shots, cities, still lives, dead things, and if the moving items are not a big part of shot (or it is a very still day) then I suppose it could have a purpose.
Robin Smith
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...