Neue Testversion 19.40b11

Hier diskutieren die Betatester von PhotoLine untereinander und mit den Entwicklern
User avatar
Herbert123
Mitglied
Posts: 2041
Joined: Sat 12 May 2012 21:38

Re: Neue Testversion 19.40b11

Post by Herbert123 »

photoken wrote: PhotoLine needs to first implement the one feature it lacks (compared to the competition): the ability to create complex, non-symmetrical vector shading. Then we would finally be able to do complete illustrations exclusively as vector graphics. The ideal way to implement this would be with Diffusion Curves, but implementing "gradient meshes" would be a decent enough alternative.
.
I agree wholeheartedly! BUT PLEASE NO GRADIENT MESHES! The gradient mesh tool in Illustrator is an abomination. That approach is hardly artist-friendly, and it was, I feel, originally designed with more simplistic drawings in mind. I suppose gradient meshes can be powerful, but if it is implemented, the GUI should be completely redesigned.

And as an alternative approach, I would ask the Hubers to check out diffusion curves instead:
http://maverick.inria.fr/Publications/2 ... /index.php
Implementing diffusion curves would also be very beneficial for bitmap to vector conversions.
/*---------------------------------------------*/
System: Win10 64bit - i7 920@3.6Ghz, p6t Deluxe v1, 48gb (6x8gb RipjawsX), Nvidia GTX1080 8GB, Revodrive X2 240gb, e-mu 1820, 2XSamsung SA850 (2560*1440) and 1XHP2408H 1920*1200 portrait
User avatar
photoken
Mitglied
Posts: 2162
Joined: Sat 28 Sep 2013 01:25

Re: Neue Testversion 19.40b11

Post by photoken »

Herbert123 wrote:BUT PLEASE NO GRADIENT MESHES!
Well, if the choice is between having no vector shading and using gradient meshes, gradient meshes is better than nothing.

Even the W3C recognizes that reality in their SVG specification: they "prefer" DiffyCurves, but will "accept" gradient meshes.

Quite frankly, I'm surprised that DiffyCurves haven't been implemented in any programs, especially in AI. After all, that research paper is seven years old and all but one of the researchers belonged to an Adobe Research Group. Some Inkscape programmers contributed an analysis, but so far, nada. Maybe the mathematics is too troublesome for practical implementations...I don't know.
Ken
Yes, I think it can be eeeeeasily done....
Just take everything out on Highway 61.
bkh
Betatester
Posts: 3661
Joined: Thu 26 Nov 2009 22:59

Re: Neue Testversion 19.40b11

Post by bkh »

Herbert123 wrote:And as an alternative approach, I would ask the Hubers to check out diffusion curves instead:
http://maverick.inria.fr/Publications/2 ... /index.php
Implementing diffusion curves would also be very beneficial for bitmap to vector conversions.
Probably too expensive, computationally. They need GPU acceleration to do 512x512 images in realtime with no blurring. They can't do blurring in real time. (Blurring speed would be similar to PL's variable blur with no GPU acceleration.)

Btw., it seems that it's not possible to add new colours inside a curve, all you get is gradients from edge to edge.

Cheers

Burkhard.
User avatar
Herbert123
Mitglied
Posts: 2041
Joined: Sat 12 May 2012 21:38

Re: Neue Testversion 19.40b11

Post by Herbert123 »

bkh wrote:
Herbert123 wrote:And as an alternative approach, I would ask the Hubers to check out diffusion curves instead:
http://maverick.inria.fr/Publications/2 ... /index.php
Implementing diffusion curves would also be very beneficial for bitmap to vector conversions.
Probably too expensive, computationally. They need GPU acceleration to do 512x512 images in realtime with no blurring. They can't do blurring in real time. (Blurring speed would be similar to PL's variable blur with no GPU acceleration.)

Btw., it seems that it's not possible to add new colours inside a curve, all you get is gradients from edge to edge.

Cheers

Burkhard.
No, I do not think so: this paper https://www.cg.tuwien.ac.at/research/pu ... 09-solver/ describes a solver method to implement diffusion curves which is easier and faster (or at least just as fast) to implement than traditional gradient mesh solvers.

So definitely doable.
/*---------------------------------------------*/
System: Win10 64bit - i7 920@3.6Ghz, p6t Deluxe v1, 48gb (6x8gb RipjawsX), Nvidia GTX1080 8GB, Revodrive X2 240gb, e-mu 1820, 2XSamsung SA850 (2560*1440) and 1XHP2408H 1920*1200 portrait
User avatar
Martin Stricker
Mitglied
Posts: 872
Joined: Tue 14 Oct 2003 08:19
Location: BW

Pipette, Farbaufnehmer funktioniert nicht richtig

Post by Martin Stricker »

Hallo,

bei mir funktionieren die Weiß-, Grau- und Schwarzpipetten in den Gradationskurven nicht mehr richtig, wenn man sie innerhalb des Vorschaufensters verwendet. Beim klicken zeigt sich zunächst die erwartete Änderung, aber sobald man die Maustaste loslässt, springt das Ergebnis wieder auf das ursprüngliche Bild zurück. Klickt man direkt ins Bild funktioniert alles wie erwartet.
Das gleiche Verhalten habe ich bei Weißpunkt, Farbkorrektur und dem Kanalmixer beobachtet.
Ich bitte um Bestätigung.

LG Martin
bkh
Betatester
Posts: 3661
Joined: Thu 26 Nov 2009 22:59

Re: Pipette, Farbaufnehmer funktioniert nicht richtig

Post by bkh »

Martin Stricker wrote:bei mir funktionieren die Weiß-, Grau- und Schwarzpipetten in den Gradationskurven nicht mehr richtig, wenn man sie innerhalb des Vorschaufensters verwendet. Beim klicken zeigt sich zunächst die erwartete Änderung, aber sobald man die Maustaste loslässt, springt das Ergebnis wieder auf das ursprüngliche Bild zurück. Klickt man direkt ins Bild funktioniert alles wie erwartet.
Das gleiche Verhalten habe ich bei Weißpunkt, Farbkorrektur und dem Kanalmixer beobachtet.
Ich bitte um Bestätigung.
Passiert bei mir auch, und zwar nur in dem Teil der Vorschau, der die Änderungen zeigt. Wenn ich die Maustaste länger drücke, springt der Wert hin und her. Sieht aus, als würde PL den korrigierten Wert als neuen Wert nehmen.

L.G.

Burkhard.
User avatar
Martin Stricker
Mitglied
Posts: 872
Joined: Tue 14 Oct 2003 08:19
Location: BW

Re: Pipette, Farbaufnehmer funktioniert nicht richtig

Post by Martin Stricker »

Passiert bei mir auch, und zwar nur in dem Teil der Vorschau, der die Änderungen zeigt.
Stimmt, ich habe bei mir keine geteilte Vorschau, weshalb ich das auch nicht bemerkt habe.

LG Martin
User avatar
Herbert123
Mitglied
Posts: 2041
Joined: Sat 12 May 2012 21:38

Re: Neue Testversion 19.40b11

Post by Herbert123 »

Bug layer effects and antialiasing:

When global antialiasing is turned off, and an emboss layer effect is applied to an object (for example, let's say a text object), the emboss effect without any softness creates artifacts.

This makes sense: antialiasing is turned off. However, when antialiasing is turned on, Photoline fails to update & recalculate the layer effects. To remove the artifacts, the layer effect dialog must be opened, and a small change must be made to force the antialiasing to take effect.

Expected result: turning antialiasing on and off should update all layer effects, and recalculate them.

By the way, it works as expected when the layer properties are used to control the antialiasing.

That is the most number of times I have used the word "antialiasing" in a couple of sentences :D
/*---------------------------------------------*/
System: Win10 64bit - i7 920@3.6Ghz, p6t Deluxe v1, 48gb (6x8gb RipjawsX), Nvidia GTX1080 8GB, Revodrive X2 240gb, e-mu 1820, 2XSamsung SA850 (2560*1440) and 1XHP2408H 1920*1200 portrait
Martin Huber
Entwickler
Entwickler
Posts: 3999
Joined: Tue 19 Nov 2002 15:49

Re: Pipette, Farbaufnehmer funktioniert nicht richtig

Post by Martin Huber »

Martin Stricker wrote:bei mir funktionieren die Weiß-, Grau- und Schwarzpipetten in den Gradationskurven nicht mehr richtig, wenn man sie innerhalb des Vorschaufensters verwendet. Beim klicken zeigt sich zunächst die erwartete Änderung, aber sobald man die Maustaste loslässt, springt das Ergebnis wieder auf das ursprüngliche Bild zurück. Klickt man direkt ins Bild funktioniert alles wie erwartet.
Das gleiche Verhalten habe ich bei Weißpunkt, Farbkorrektur und dem Kanalmixer beobachtet.
Ich bitte um Bestätigung.
Ich werde das reparieren.

Martin
Martin Huber
Entwickler
Entwickler
Posts: 3999
Joined: Tue 19 Nov 2002 15:49

Re: Neue Testversion 19.40b11

Post by Martin Huber »

Herbert123 wrote:Bug layer effects and antialiasing:

When global antialiasing is turned off, and an emboss layer effect is applied to an object (for example, let's say a text object), the emboss effect without any softness creates artifacts.

This makes sense: antialiasing is turned off. However, when antialiasing is turned on, Photoline fails to update & recalculate the layer effects. To remove the artifacts, the layer effect dialog must be opened, and a small change must be made to force the antialiasing to take effect.
I will fix that.

Martin