What am I doing wrong? I only get 21.5 MB

[Jeff S]

Advertisement (gone after registration)

Easy for me. Helped to do homework in advance to avoid common mistakes and to simplify the implementation.

Jeff

Edited August 16, 2021 by Jeff S

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[Jean-Michel]

3 hours ago, mmradman said:

Which current full frame camera is 16bit?

Another poorly understood aspect in digital photography commonly encountered here on LUF, there is no quality loss between uncompressed and lossless-compressed RAW file, benefit is smaller digital file.  I hear argument digital memory is cheap, so is misuse of earth resources.

Don't know of any 16 bit per pixel in the 35 mm format –the M(240), and SL2 are 14 bits per pixel. PhaseOne does have 16 bits per pixel sensors, at a price!

My note of 24 mb or so DNGs are with raw compression ON (uncompressed makes for   48.2 mb fies)

As for the 8 or 16 bit: that is bits per channel (not per pixels) for when going from LR (or ACR, or …) to PS. There are good reason for using one or the other as need be. 16 bits per channel makes for smoother gradations in PS. For many cmyk workflows (printing on presses) the printers cannot handle 16 bits and need the files in 8 bits. And many PS filters , if you use them, need the image to be in 8 bits.

Lightroom Classic happens to be my workflow of choice — used to be Bridge/ACR/PS — I am sure that there are many other excellent ways to process and catalog images.

And, after doing all the processing, cataloguing, etc., if one does not print the image, it does not exist.

[beewee]

I recall the Hasselblad X1D also has a 16bit sensor output.

That being said, you don’t need a 16bit sensor to have 16 bit files. 16bit/colour channel is only a container format and not the input. You can store untouched/raw 14bit data in a 16bit container and it means you won’t lose any information as opposed to storing 14bit data into an 8 bit container which would require some downsampling in the colour/gradient domain. Also, you can still benefit from editing in 16+bit space with a 14bit sensor in post because, as you make colour/tonal changes (i.e. exposure, colour balance, etc…), some changes will result in in-between values that would not otherwise be able to fit cleanly in a 14bit container but a 16bit container can help capture those subtle changes. Lightroom and other image editing software will perform mathematical operations at much higher bit depth than 14bit even if the input is only 14bit. Think 64bit or 128bit resolution when the edits are made the algorithm is generating the resulting image but the result is then quantized back into whatever output format you need. For example, 8bit for JPEG/TIFF, 16bit for TIFF, 10bit output to monitor (typically), etc…

Edited August 16, 2021 by beewee

[FlashGordonPhotography]

3 hours ago, beewee said:

I recall the Hasselblad X1D also has a 16bit sensor output.

That being said, you don’t need a 16bit sensor to have 16 bit files. 16bit/colour channel is only a container format and not the input. You can store untouched/raw 14bit data in a 16bit container and it means you won’t lose any information as opposed to storing 14bit data into an 8 bit container which would require some downsampling in the colour/gradient domain. Also, you can still benefit from editing in 16+bit space with a 14bit sensor in post because, as you make colour/tonal changes (i.e. exposure, colour balance, etc…), some changes will result in in-between values that would not otherwise be able to fit cleanly in a 14bit container but a 16bit container can help capture those subtle changes. Lightroom and other image editing software will perform mathematical operations at much higher bit depth than 14bit even if the input is only 14bit. Think 64bit or 128bit resolution when the edits are made the algorithm is generating the resulting image but the result is then quantized back into whatever output format you need. For example, 8bit for JPEG/TIFF, 16bit for TIFF, 10bit output to monitor (typically), etc…

The X1D/GFX50 sensor is 14 bit (in a 16 bit wrapper). The newer 102MP sensor in the GFX100s but not yet used by Hasselblad is 16 bit. The larger 50+ MP sensors used by HB and Phase are 16 bit.

Gordon

[Jeff S]

3 hours ago, FlashGordonPhotography said:

The X1D/GFX50 sensor is 14 bit (in a 16 bit wrapper). 

Just as the sensors in the various Leica S iterations.

Jeff