Again, this is a subjective judgement and I can’t present any timings.īut the big question is, are the results better? For natural subjects I would say no, they are on a par with the previous version but for manmade subjects the improvement is noticeable. Processing performance appears to be on a par with the old Gigapixel AI version 4. I did though have the impression of it rendering the preview slightly faster. This is hard to judge with any certainty because after the upgrade I don’t have access to the previous version. The preview speed of Gigapixel AI 5 seems slightly faster in the previous version. When I switch to the Natural image type, the edges are slightly soft but fine grain or noise in the image seems to be better preserved. With the Man-made option selected, edges become clearer and appear more heavily sharpened. Whilst the settings for “Suppress Noise” and “Remove Blur” don’t change, the results are quite different. Whilst experimenting with the latest version and with the preview zoomed to 100%, I could see the difference. All you need to do is pick the type of subject that you’re enlarging, being either natural or man-made. They therefore tuned their routines to handle these two types of enlargement. Topaz Labs has realised that there is a large difference between natural subjects such as landscapes and those include man-made objects.

The big news in Topaz Gigapixel AI 5 is the special enhancement for images of man-made objects. But there’s also a new “Image Type” option. This was also in the previous version of Gigapixel.

There are though a couple of special features to handle the enlargement of images with facial recognition. After that you can pick the automatic options and it will create a great enlargement. In fact, the most difficult thing is probably selecting the size and resolution of the enlarged image. Gigapixel is extremely user friendly and easy to use. There’s some valuable information in here, but I’m not sure they are necessary.

Topaz labs gigapixel ai