Preferred PNG optimizer?
Posted: 04 Aug 2018, 21:53
Dear cardfloppers and eagle eyes,
been working in prepress for the last 22 years, using Photoshop daily and from time to time I add a new set (mostly old school sets) of descreened scans to my tablet.
For each set, there's a PS actionlist to produce very agreeable descreening results, with perfect transparent corners, from the base XLHQ pics set.
So I save as .png and optimise the output with pinga atm ( https://css-ig.net/pinga ) which works like a charm (output > mark all > send to pinga > optimize. The output will be re-/overwritten). It uses all available cores, so this will sloooow things down on your machine(!) And I don't think it uses dithering, only palette customisation and other tricks, I added an example.
(I didn't destroy every last bit of screening, mind you Descreening, for me, has served its purpose as soon as there's no chance for a moiré on my tablet anymore. And as you all know, every descreening routine eats fine detail, so I try to keep slightly more detail. Except for those brutal screenings from the very first sets
Oh, and if you're seeing intransparent corners, that might be due to any kind of factors, it definately works here!)
My question to you is: What are your preferred ways of png optimisation, focusing on best color reproduction (perceptual)?
Size differences of a few kbytes (eg. 315kb or 400kb for the same image) wouldn't make much difference to me.
I tried most of these already: https://css-ig.net/png-tools-overview#overview but might have missed the best settings regarding (perceptual) color accuracy on some.
Ironically, png optimisation via "save for web" using photoshop wasn't "that" bad, but it's a pain to implement with actionlists (at least with CS6 here).
So, hit me with your preferred and proven to be very accurate color reproduction png optimisations, command line tools wouldn't be a problem
Thanks in advance!
Cpt. Kird
been working in prepress for the last 22 years, using Photoshop daily and from time to time I add a new set (mostly old school sets) of descreened scans to my tablet.
For each set, there's a PS actionlist to produce very agreeable descreening results, with perfect transparent corners, from the base XLHQ pics set.
So I save as .png and optimise the output with pinga atm ( https://css-ig.net/pinga ) which works like a charm (output > mark all > send to pinga > optimize. The output will be re-/overwritten). It uses all available cores, so this will sloooow things down on your machine(!) And I don't think it uses dithering, only palette customisation and other tricks, I added an example.
(I didn't destroy every last bit of screening, mind you Descreening, for me, has served its purpose as soon as there's no chance for a moiré on my tablet anymore. And as you all know, every descreening routine eats fine detail, so I try to keep slightly more detail. Except for those brutal screenings from the very first sets
Oh, and if you're seeing intransparent corners, that might be due to any kind of factors, it definately works here!)
My question to you is: What are your preferred ways of png optimisation, focusing on best color reproduction (perceptual)?
Size differences of a few kbytes (eg. 315kb or 400kb for the same image) wouldn't make much difference to me.
I tried most of these already: https://css-ig.net/png-tools-overview#overview but might have missed the best settings regarding (perceptual) color accuracy on some.
Ironically, png optimisation via "save for web" using photoshop wasn't "that" bad, but it's a pain to implement with actionlists (at least with CS6 here).
So, hit me with your preferred and proven to be very accurate color reproduction png optimisations, command line tools wouldn't be a problem
Thanks in advance!
Cpt. Kird