I decided to write a few lines in continuation of my previous article about the Nikon’s next move towards improving its future APS-C sensor(s); well it turned out that it’s a full frame called Z7 this time! I asked again Bill Claff if he could give me the permission to use his measured data on Nikon Z7 sensor parameters to add to the data-set I already analysed on my previous work , and he kindly did. My focus is on the figure 4 of that post where I speculate where the next Nikon sensor would surface by making a shaded box in that figure. Well it is not a big surprise that Nikon’s Z7 sensor is almost falling close to that region where 7RM II, 5D Mark IV and 850D are. Of course this is not an APS-C sensor but sure is the right direction to go towards higher DR. Once again PCA model was able to allocate a new sensor (Z7) quite nicely with regards to the existing sensor data. I am looking forward to see the next Nikon’s APS-C in my data.
For those who may be interested I will soon upload another post using supervised multivariate modeling to get deeper into the sensor data and highlight areas where most of the reviews on the Internet were unable to explain or simply overlooked.
I have removed DSNU and PRNU variables from this PCA model and as you might have noticed the PCA score of the first two components has improved by 12.7% to 82.7% comparing to the previous model (R2x= 0.485 and R2x=0.342, Figure 1) . This means that the current model was able to explain 82.7% of the data variability which is very good. In fact those two variables were making the model little unstable although you may just notice minute offsets in sensor scores.