HOME | DD | Gallery | Favourites | RSS
| zy0rg
# Statistics
Favourites: 202; Deviations: 337; Watchers: 2303
Watching: 42; Pageviews: 150362; Comments Made: 3431; Friends: 42
# Interests
Favorite movies: PrestigeFavorite bands / musical artists: Oceansize
Favorite games: Machinarium
Other Interests: music, art
# Comments
Comments: 557
Senzune In reply to ??? [2017-08-07 13:19:09 +0000 UTC]
Makes sense, though of course new features will be meaningless if the old and essential ones aren't there Hence the comparison.
As for the new thing (though it's being developed for Apophysis and/or Chaotica atm): Xaos node editor.
Hell, you might even make this a very important main screen (with sidebars for variations, color settings, etc) in parallel with an Apophysis-like regular editor display.
Basically a node screen that displays the connections between transforms, obviously, with the weight value logarithmic-ally proportional to arrow width (such that very high values don't instantly take up the entire screen).
The nodes themselves contain transform name, names of its variations, and a rectangular display of its color and associated location (e.g. 0.4 on the gradient). Perhaps even all other settings in an expandable manner.
Selecting a node will bring the sidepanel to the associated settings, similar to how Apophysis' regular editor works. Selection a transform in the regular editor of the new program would do the same thing.
Editing values directly from within a node is a maybe - it can be useful for quick alterations but can also be an annoyance when you simply wanted to just select the node instead of a value to change.
(Selecting the wrong things is a major disadvantage of Chaotica's editor, btw.)
For user friendlyness, display all main windows for fractal editing at once, instead of requiring alternative windows for the editor, color/gamma settings, etc.
So a 4-window view. Top left: xaos node editor. Bottom left: transform editor. Top right: re-arrangable panels for all the transform settings. Bottom right: Render panel with sidebar (or top bar or whatever, just make it rearrangeable) for render settings.
The panels/bars within each of the 4 windows can be rearranged by the user freely, as can the 4 windows. (Note that I used 'windows' rather freely as well: what I meant is that all displays are to be done in a single screen without having to open any secondary Microsoft Windows Window (like clicking the triangle on Apophysis or opening the World Editor in Chaotica).
For support of smaller displays and to avoid clutter, allow the user to combine any number of these 4 main views behind tabs (similar to Chaotica combining sidebars).
I realize it's not entirely a new feature per se but it's still a new way of displaying.
A truly new and likely impossible feature: adaptive/dynamic final render weighting. A mouth full and there's probably a better name for it. The renderer recognizes areas that are being calculated less compared to other areas, and adjusts the final weight of calculations (not the relative weights/xaos) to attempt a more evenly distributed rendering.
I haven't thought of something else truly new yet (asides from the impossible reversal one) but as I become more familiar with structured fractals, I'll probably think of something.
👍: 0 ⏩: 1
zy0rg In reply to Senzune [2017-08-07 14:32:01 +0000 UTC]
It looks like we have similar ideas.
Xaos node editor - I wanted to do visualization of xaos using nodes with arrows too. It would be useful for better understanding of complex flames.
I turned down this idea after all and instead I'm going to remove xaos at all. 95% of use cases for xaos are chaining and/or layering. For these purposes I'm going to add new virtual transformations called "chain", "group" and "iterate". In such case you will clearly see what fills what without the need for any additional visualizations
Display all main windows for fractal editing at once - I'm actually going to do these windows similar to Photoshop, where you can make them standalone, combine them into tabs or collapse into buttons. I'm also going to add a possibility to separate them into standalone window, so that you can move them to another screen if you have multiple displays. There is also one more option, but I don't want to spoil it for now...
Adaptive/dynamic final render weighting - If I understood you correctly you want to make image brightness evenly distributed, so that there were no black areas because of low density. I have a solution for this.
👍: 0 ⏩: 1
Senzune In reply to zy0rg [2017-08-07 14:56:18 +0000 UTC]
You're going to remove Xaos? I assume only the name, because you can't have these chains/groups without relative weights.
I mean it's basically the most important feature of Apophysis for creating structured fractals so that paragraph scared me a bit.
As for the adaptive rendering thing, what I meant is that since fractals are iterated based on chance, pixels are calculated more towards the 'centers' of calculations for each transform and less the further away they are (or the weaker the results of each transform). An easy example is a simple linear tiled square. The further away from the center, the lower the chances each pixel has of getting rendered. In the render, this is visible by more pixels in the center and less at the periphery.
Could you share some details about this new rendering solution? Or do you want to keep that a surprise as well?
👍: 0 ⏩: 1
zy0rg In reply to Senzune [2017-08-07 15:19:37 +0000 UTC]
I'm going to remove the concept of xaos, i.e. some global numbers controlling rendering flow based on probabilities. The solution with "chain", "group" and "iterate" covers all use cases that I could think of and makes it in much more transparent way.
Regarding the solution for adaptive rendering - gamma in Apophysis actually does what you want but in a very limited way. My solution is to give control over "alpha" channel in "curves" editor, in addition to usual rgb.
👍: 0 ⏩: 1
Senzune In reply to zy0rg [2017-08-07 15:31:44 +0000 UTC]
Are you sure that's the same thing regarding the rendering thing?
What I'm talking about is pixels actually being calculated (regardless of color or even transparency). E.g. even on a high sampling level, there are still small patches where no pixels were calculated at all. Gamma is different in that it decides how pixels should look when many calculations are done on them, not what pixels actually get calculated.
It's in the same way as rendering at very low density/sampling level: a lot of unconnected rendered pixels and inbetween just areas that weren't calculated at all. Those are the areas I speak of.
👍: 0 ⏩: 1
zy0rg In reply to Senzune [2017-08-07 16:26:22 +0000 UTC]
If you want me to solve this issue so that renderer chooses to render some areas better by throwing more pixels there, then I can not. It will break probabilities and result in unpredictable images.
What I can do is to increase transparency on areas where less pixels were drawn. And this is what gamma does. Curves on alpha channel will do the same but with more control compared to gamma.
👍: 0 ⏩: 0
moonhayze In reply to ??? [2017-07-23 12:03:10 +0000 UTC]
Hey Zyorg, i love your work. Do you have a hi-res version of "AGAIN" so that i may create a print for my entire living room wall?
👍: 0 ⏩: 1
zy0rg In reply to moonhayze [2017-07-23 15:19:04 +0000 UTC]
Hi and thanks for support.
I have a 3840x2160px rendered version of the picture, which clearly isn't enough for an entire wall print.
It is possible to render it in almost any resolution, but it will take some time and effort so I'll expect payment for it.
We can discuss pricing separately if you're still interested. Send me a note in such case.
Regards
👍: 0 ⏩: 0
poca2hontas In reply to ??? [2017-07-13 18:50:29 +0000 UTC]
welcome to and thanks for joining us
👍: 0 ⏩: 0
eeksray In reply to ??? [2017-06-12 18:17:34 +0000 UTC]
Damn... so.... you figured out how to control apophysis so well that you can create an eye and even a reflection in that eye.... Nice. eyelashes? next?
Thanks
👍: 0 ⏩: 1
zy0rg In reply to eeksray [2017-06-14 10:16:57 +0000 UTC]
Thanks for support
It's not as hard as it may seem. Just some clicking and dragging, and a few random numbers.
👍: 0 ⏩: 0
sharperofthefew In reply to zy0rg [2016-10-04 17:11:53 +0000 UTC]
NP! I'm running a webcomic you can check out in the link in my signature!
👍: 0 ⏩: 0
sharperofthefew In reply to ??? [2015-10-03 22:17:39 +0000 UTC]
IVE BEEN IN LOVE WITH YOUR DESKTOPS FOR YEARS
👍: 0 ⏩: 1
EmeryPhotographics In reply to ??? [2015-10-03 07:57:31 +0000 UTC]
Happy Birthday! I hope your birthday as a wonderful as your work!!!
👍: 0 ⏩: 1
zy0rg In reply to EmeryPhotographics [2015-10-06 11:20:57 +0000 UTC]
Thank you! Probably it was even better.
👍: 0 ⏩: 0
Antares2 In reply to ??? [2015-10-01 20:50:19 +0000 UTC]
wishing you a happy birthday soon !
👍: 0 ⏩: 1
<= Prev | | Next =>