Suggested Capture Settings

because fusion requires the capture of two images it’s doubly impactful to adjust camera settings in a way that will maximize the speed of capture pretrigger / flash mode the ixh series of cameras has a dedicated mode for using continous lights like the stellar led in this mode a pretrigger is allowed, and should be set to 80ms for use with fusion otherwise blue tint may be observed in the white capture, lowering color accuracy black reference for example, on a phase one ixh you can set "black ref" to "precorded" in camera settings which will save a fraction of a second per capture this setting/feature is sometimes called dark frame, or dark current calibration shutter mode on some cameras such as the phase one ixh you can set the camera to use a sensor based electronic shutter (es) rather than a mechanical shutter in some cases this will increase capture speed bit depth on some cameras you can lower the bit depth of capture for a faster capture speed this would not be recommended for extremely high contrast content such as dense glass plate negatives but may be an acceptable option for reflective materials – you should test in both modes and compare final results for yourself ixh where possible the phase one ixh has a unique feature that allows dt nexus to know when the camera has completed the first exposure, even before the file from that capture has been saved, so that the second capture can be triggered immediately this makes the phase one ixh faster for fusion than other cameras, despite its very high resolution sharpening when fusing images their existing sharpening settings are used generally for best results we would suggest reducing or eliminating sharpening during the fusing process and then adding additional sharpening to derivatives as desired adjustments fusion pairs should have identical adjustments for example, if the sapphire capture has its white balance changed the result will be erroneous in the future we plan to add more checks to ensure that is the case; until then the onus is on the user capture iso and compression while fusion is supported for any iso and raw compression setting, we suggest using the lowest real iso of the camera and the highest quality raw compression setting for example with an ixh 150mp camera we would suggest iso 50 and iiq 16l output tone note that fusion uses linear scientific tonescale where available (phase one cameras) this is because fusion is intended to produce exacting color and tone and linear response will reduce the absolute accuracy of the highlights however, this does mean that there is no "grace" given to tones that are above the white point set during calibration that means blown highlights will not be handled gracefully we plan to explore various ways of handling this issue in future versions of fusion; for example, via hdr encoding that would allow tones brighter than scene white transmissive capture, additional diffusion while the two channels of light (white and sapphire) are interspersed within the panel (so that we avoid issues of colored shadows or textural rendering issues) shooting directly into the light will reveal an extremely subtle pattern in the sapphire capture this pattern is sufficiently subtle that you may not notice it without careful scrutiny, but for extremely high quality use cases it may unacceptable a thin secondary diffuser such as the "dt secondary diffuser for dt atom" positioned between the light and the film will eliminate that pattern higher deltae numbers for fusion do not be alarmed if the deltae results for fusion are not as low as the white profile on the target which was used to make the profile the white only profiles in dt nexus are generated using a look up table profile (lut) such profiles aggressively match actual input color to desired output color using a table of corresponding values; this drives the deltae results on the target being used to profile very low on the other hand, fusion profiles in dt nexus are generated using a 6 channel matrix profiling method this method builds an overall response map rather than a patch specific look up table, which leads to a more robust and holistic result because of their nature the numerical results (deltae values) of the target being used to profile will almost always look better for a lut however, a 6 channel matrix profile (if executed properly) will produce far better overall color accuracy this can and should be verified using an independent target – one of a different make and model than the one used to make the color profile learn more here https //heritage digitaltransitions com/the importance of evaluating color quality using an independent target/