This is an update on using the digital zoom in my camera (phone). And yes, I am now even more convinced that the digital zoom is, within reason, the way to go!
The piece that spawned this update was a totally random test in the last part of April and reported elsewhere here. This second, more controlled and, frankly, maybe more realistic test, was done recently and the results were more than gratifying as they confirmed the initial findings.
The camera (phone) was put on a tripod to assure the view of the scene was absolutely controlled and a base exposure was made. The camera increased its zoom in 1/10th unit increments and an exposure was made at each incremental increase through a zoom factor of x2.9. The upper range of the test was based completely on the scene presented in the viewfinder. Some may say this was completely arbitrary and didn't test the "full-range" of the zoom, and they would be absolutely right. Maybe a further test is in the offing... who knows at this point. But, in terms of 35mm equivalence, this gave a zoom range of roughly 28mm through 75mm.
The results of the test are reported here using 700px x 525px images and here using 1024px x 768px images. (For those who are picky, the images all have a 1px black border which adds 2px to the width and height.) In addition to making the digital presentation, all of the images were printed at 8"x6" at 300dpi on coated matte presentation paper using an Epson printer and the results were just as positive. With the exception of resizing and bumping the dpi up to 300 for making the prints, these photographs are exactly as they came out of the camera (phone). They are the full frame with absolutely no adjustment.
Both links will open a new window. It's recommended that you view either presentation in full screen mode.
And finally... in the previous piece on this subject, it was noted that images above a zoom factor of 2.3 started to get "soft"... but that doesn't seem to be the case here. Everything seemingly holds together pretty well all the way through. This is one of those things that just can't be explained. Since this testing is far more controlled and far less off-the-cuff than the previous effort, it would seem these results are a more accurate representation of what happens in the zooming process.