Inspired by the recent Sky Competition – May 2020 which for various reasons I was unable to submit any entries, I decided to check all my images where the sky was the main subject or an important part of the image.
I recovered the original raw files and reprocessed them with concentration on the sky. the raws were re processed with Qimage with concentration on the sky which mainly involved reducing the Blue and cyan to make the clouds more prominent. The blue sky areas were excluded from sharpening to reduce sharpening of any noise or chromatic aberrations.
I noticed in the competition there was some comments about lack of cloud sharpening. I have found many clouds (eg lenticular clouds) do not have enough edge factor to take sharpening well, increase the sharpen factor over sharpens other areas and any noise, you take you choice, at least I do and say sod it.
The original raw files could not be found in a few cases, so jpegs have been recovered from various sources.
A Click on image opens up full screen, Esc or top left/right arrows opens up the image with title and comments.
Tif to Jpeg conversion at 100%
Tif to Jpeg conversion at 80%
This image is much sharper in WebP 'format' When checked against the submitted image.
Test chart to test color shift in WebP 'format'
RB Wix hosted WebSite images now in WebP format
I have recently noticed that Wix now convert images on site from submitted jpg to WebB
I have done a few tests on my comp./screen
I find there is a small difference between the uploaded image and the right clicked downloaded image.
Surprisingly, there appears to be a discernable improvement in image quality over the submitted, mainly as far as I can see an increase in sharpening. This may not be to everybody’s liking.
The whole business is about image file size. But I find this is only achieved on large file size jpegs
A jpeg at 1400x1050 at 100% 1.4 MB is compressed to about 276K
A jpeg at 1400x1050 at 80% 267K is compressed to about 255K
Bearing in mind the downloaded image is smaller than the 1400x1050 submitted.
There is a small colour shift. The second in top left on colour chart is #F1956C on submitted jpeg image and #F1966C on the right click downloaded WebP
To view WebP most image progs appear to require a plug in. For viewing I have found paint net 4.2.5 which supports WebP is quite good, allowing jumping from one image to another for comparisons.
These pictures were taken on a recent visit to the Guggenheim Museum, Bilbao. I gave up trying to straighten the verticles.
Pay Millions for an Architect and construction then spoil it with advert.