Serif Photo Zoom

[JohnRostron]

As a Serif customer, I was offered Serif Photo Zoom, their photo-enlarging program at £23, (80% off). I had a look at the reviews and these were generally positive, so I thought I would give it a try. I chose a photo of a male Chaffinch taken in the cafe gardens on Tresco, Isles of Scilly. I reduced it first to 50%, then to 25% and then used PhotoZoom, Serif Affinity and Photoshop CC2017 to resize back to the original. There was no subsequent post-processing.

#1 Here is the original at 100%, cropped from the original image.


#2 Reduced to 50%, then enlarged back to the original with Photoshop


#3 Reduced to 50%, then enlarged back to the original with Affinity Photo


#4 Reduced to 50%, then enlarged back to the original with Photo Zoom


#5 Reduced to 25%, then enlarged back to the original with Photoshop


#6 Reduced to 25%, then enlarged back to the original with Affinity Photo


#7 Reduced to 25%, then enlarged back to the original with Photo Zoom


One thing is clear. None of them make a good job of the detail in the 25% reduction/enlargement. However Photo Zoom does make a better job of the not-so-fine detail. Look at the white feathers on the wing. They do seem clearer in the Photo Zoom image.

Looking at the 50% images, there is less to choose between them, but I still think that Photo Zoom has done the better job.

Overall I think it is a good buy. I have a number of images scanned from slides where I saved them as too small. Photo Zoom may well rescue these into a usable form.

Edit: In my original posting, I had sharpened the final images. I have now reverted to unsharpened versions.

Comment welcome.

John

[DanK]

Interesting. Thanks for posting. I agree that the Photo Zoom images look better. However, I think a cleaner test would be to do this with no sharpening or other processing after enlargement.


Sent from my iPad using Tapatalk

[JohnRostron]

Interesting. Thanks for posting. I agree that the Photo Zoom images look better. However, I think a cleaner test would be to do this with no sharpening or other processing after enlargement.

Fair comment. I have replaced the original images posted with unsharpened versions.

John

[Manfred M]

Interesting. Thanks for posting. I agree that the Photo Zoom images look better. However, I think a cleaner test would be to do this with no sharpening or other processing after enlargement.


Sent from my iPad using Tapatalk

I would agree and using just the base (raw data) image. The downsizing and upsizing is not a fair test as that brings the downsampling algorithms into play as well. A parametric editor would definitely have an advantage over a pixel based editor in that type of exercise as the original data would be retained and the data the re-upsize would use the original data, not the downsized data.

[JohnRostron]

True, but my ultimate target images were, as I said, images that had been resampled down and where the originals were no longer available to me (the slides had gone back to the original owner).

In this exercise, I took the original Tiff, and converted to jpeg (for the 100%), then to 50% and 25% as Jpegs in Affinity. I used maximum quality for each of these. Once the various re-scalings were complete, I stacked the seven images and cropped to just the Chaffinch, then exported each as a 80% quality jpeg for this posting.

John