Being a feature that is hardly every used, is there going to be any speed optimization updates for this very useful node?
As you note, this is a feature that is not frequently used. Consequently we can't really justify putting much dev time into it. There are better dedicated heightfield editors out there with more robust erosion functions so that's what we'd recommend if you need something beyond the basic capability TG2 currently provides.
- Oshyan
I figured as such. At least make the alpine fractal terrain more accessible. Someone else had suggested that already here and it needs it.
I'm not sure what you mean by "accessible"...
- Oshyan
More inputs for the node to drive additional effects. That would be fun. ;D Each set of alpine settings can be pluggable.
Hmm, I think I see what you mean, but given virtually no other similar (generator) node works that way, it's not likely to be possible any time soon. We've contemplated some kind of more general "drive any setting by the output from a node" type of system for the future some time, but it'd be a pretty major undertaking.
- Oshyan
Not quite seeing how that would work. Perhaps it is going over my head.
Another example is how you guys added additional inputs to the cloud layer node so all sorts of effects never before possible are now a fact. I can see the same being done to that alpine node so for example a power fractal can be used as an input for the late disposition effect as with the rest of the remaining effects. I would imagine certain erosion effects being made possible that were not before. It is not changing the alpine fractal code itself, only opening it up per each setting to make it plugable from other nodes.
Hmm, well something to think about.
- Oshyan
Certainly, the more play a user has, especially those who love to go real deep into these nodes, the more enticing. I know this software is capable of some extraordinary stuff. However, if more node access is opened up, then even better. I like exploring the power fractals but i want to explore the alpine fractal as well. I am sure others would agree. ;)
Quote from: ChrisC on November 28, 2011, 12:53:09 AM
Certainly, the more play a user has, especially those who love to go real deep into these nodes, the more enticing. I know this software is capable of some extraordinary stuff. However, if more node access is opened up, then even better. I like exploring the power fractals but i want to explore the alpine fractal as well. I am sure others would agree. ;)
Definitely agree ;D
Although I know it's not relevant at this point I think it is good to consider for TG3 that the architecture will be set up to be able to drive many settings by a function input. Something which not only is already done in the last couple of years by TG's sole competitor, but also like you have mentioned it will create a wealth of possibilities.
Another great example would be to be able to plug powerfractals into the strata shader's individual settings.
Or into fake stones to control size with a powerfractal, or pancake effect, or tallness, or... ;D ;)
Cheers,
Martin
Please, do not get me drooling here. ;D
Indeed, in that area we need to compete with that sole competitor. It makes for good business to.
Of course we have no end of things we can add. ;D
I'm curious if you know of any good examples from said (or unsaid, hehe) competitor's software that relies on this ability to drive parameters by node output. I've always assumed it would be tremendously useful and cool, but I frankly don't see a lot of interesting (i.e. unusual, creative, couldn't be created from obvious and known traditional methods) stuff from... Vue (there, I said it, hehe).
- Oshyan
Here i found one example but i am certain there is more. I think all perimeters in each node is linkable.
http://youtu.be/jYuYu3IN8UY
On another note, a voronoi noise could be of help to be added to the power fractal node. Voronoi is everywhere in nature. ;D
Quote from: Oshyan on November 29, 2011, 01:11:11 AM
Of course we have no end of things we can add. ;D
I'm curious if you know of any good examples from said (or unsaid, hehe) competitor's software that relies on this ability to drive parameters by node output. I've always assumed it would be tremendously useful and cool, but I frankly don't see a lot of interesting (i.e. unusual, creative, couldn't be created from obvious and known traditional methods) stuff from... Vue (there, I said it, hehe).
- Oshyan
If you look at Quadspinner's latest Masterclass LA you'll see some of the uses of it, besides that you will see that virtually every parameter in the software has that particular symbol for shader input next to it, like we do have for animation :)
I think you'd be surprised Oshyan about the myriad of effects/results you can achieve when using a powerfractal (for instance) as input for a parameter.
Even powerfractal driven powerfractal parameters ;) And the examples Chris and I pointed out plus many more, likely.
I'll try some more, since you can't think of many examples:
Fake stone size based on greyscale input, allowing for altitude based sizes as well or distance driven.
You could do the same for the scale parameters in the powerfractal or PF controlled voronoi/perlin scale
PF controlled water roughness/wavesize/wind gusts/transparency (though you could fake transparency by using multiple shaders, but that would be slower).
PF controlled object sizes in a population.
Anyway, best proof so far is the cloud node which has these specialized inputs. I can't speak for everybody, but I do use those frequently :)
Cheers,
Martin
Quote from: ChrisC on November 29, 2011, 02:39:30 AM
Here i found one example but i am certain there is more. I think all perimeters in each node is linkable.
http://youtu.be/jYuYu3IN8UY
On another note, a voronoi noise could be of help to be added to the power fractal node. Voronoi is everywhere in nature. ;D
Yeah I like that suggestion, saves time and networkspace.
You're misunderstanding what I'm asking, I think. I know Vue *has* the inputs everywhere, and people may even use them sometimes. My question is whether this produces *unique* and desirable effects that would be difficult or impossible to achieve otherwise. Note that I have been a proponent of adding this feature (when possible) for a long time, so I'm not against it at all. It just occurred to me that, since we have a real example of the capability "in the wild", it would be interesting to see just how often and to how noticeable an effect it is used.
So far I honestly haven't seen anything that surprised or impressed me. All the examples you mentioned do seem good, I just wonder if the end result is a huge improvement, or just a minor one.
- Oshyan
No I'm not misunderstanding, you're selectively in responding/reading Oshyan ;) :P
I know you're not against it, the least I know.
I'm pretty sure it will raise possibilites which are either impossible to do or extremely hard to create or very expensive to render when trying to design it by using existing functions/tools only.
With only function control input for clouds only we'll never know whether it will be a huge improvement or just a minor one ;)
I see your point about "noticeable" or not and using the cloud control input is a step further in advancement than creating nice clouds themselves.
The same thing may account for allowing function input into a variety of other nodes/parameters.
Being it very complex already it might even just add more complexity.
No selective reading/responding here, at least not intentionally. Perhaps just a good old fashioned misunderstanding. ;D
Yes, the clouds are the best example, and they were in TG2 before Vue. That's kind of my point, if this capability is/was already in Vue, then what of the new "drive height/depth by input" parameter they added? I guess it's not universal in their system?
Anyway, it would be a much bigger task to add this kind of capability generally to all nodes. The cloud controls are of course a great addition, we already add that kind of capability selectively where there's a clear need for it. The greater question is the time/cost vs. reward of this particular "global parameter driven by input" kind of system. One look at the "feature wish" thread will show there are plenty of things for us to spend our time on. ;)
- Oshyan
http://www.facebook.com/photo.php?fbid=208687082505355&set=a.131897166851014.11543.126709077369823&type=3
http://www.facebook.com/photo.php?fbid=208384332535630&set=a.131897166851014.11543.126709077369823&type=3
http://www.facebook.com/photo.php?fbid=148112385229492&set=a.131897166851014.11543.126709077369823&type=3
Someday, i would like to see this inside of terragen.
Yeah, QuadSpinner's rock formations are quite nice. Is that how they're done in Vue?
- Oshyan
Indeed, i am still grasping at creating such remote effects. For example, trying to make Goblins using the fake stone shader is still not possible, at least not that am aware of.
Quote from: Oshyan on November 29, 2011, 05:47:24 PM
No selective reading/responding here, at least not intentionally. Perhaps just a good old fashioned misunderstanding. ;D
Yes, the clouds are the best example, and they were in TG2 before Vue. That's kind of my point, if this capability is/was already in Vue, then what of the new "drive height/depth by input" parameter they added? I guess it's not universal in their system?
Anyway, it would be a much bigger task to add this kind of capability generally to all nodes. The cloud controls are of course a great addition, we already add that kind of capability selectively where there's a clear need for it. The greater question is the time/cost vs. reward of this particular "global parameter driven by input" kind of system. One look at the "feature wish" thread will show there are plenty of things for us to spend our time on. ;)
- Oshyan
Yes Oshyan, like I mentioned in one of my previous replies it IS universal in their system.
At least, all the shots I've seen it that video it seemed that virtually every parameter allowed for linking it with another input.
Quote from: ChrisC on November 29, 2011, 08:12:45 PM
Indeed, i am still grasping at creating such remote effects. For example, trying to make Goblins using the fake stone shader is still not possible, at least not that am aware of.
The best idea to get those would be separate rock objects. If I understand correctly by what you mean with Goblins? I guess you mean kind of spires with loose rocks stacked?
Quote from: Oshyan on November 29, 2011, 05:47:24 PM
No selective reading/responding here, at least not intentionally. Perhaps just a good old fashioned misunderstanding. ;D
Anyway, it would be a much bigger task to add this kind of capability generally to all nodes. The cloud controls are of course a great addition, we already add that kind of capability selectively where there's a clear need for it. The greater question is the time/cost vs. reward of this particular "global parameter driven by input" kind of system. One look at the "feature wish" thread will show there are plenty of things for us to spend our time on. ;)
- Oshyan
;D
I know about the scope of the task, very big. It will be quite a change in the shader-architecture which is best to be build from the ground up I think, therefore I said it's likely not relevant now and perhaps only for TG3. If something like that will be considered.
There's indeed a lot more being requested. Some, if not most, are more desperately needed.
Quote from: Tangled-Universe on November 30, 2011, 05:07:59 AM
Quote from: ChrisC on November 29, 2011, 08:12:45 PM
Indeed, i am still grasping at creating such remote effects. For example, trying to make Goblins using the fake stone shader is still not possible, at least not that am aware of.
The best idea to get those would be separate rock objects. If I understand correctly by what you mean with Goblins? I guess you mean kind of spires with loose rocks stacked?
http://www.utah.com/stateparks/goblin_valley.htm
I am not certain there are other types of formations on other parts of the earth but these are the best examples of what i am after.
Martin, you're absolutely right that there are many uses to controlling all sorts of parameters with functions or other shaders, and we have this marked for future development. There are some things that won't give the results you expect, though. If you've tried to use the 'scale' input on the Perlin 3D Scalar function or the voronoi functions, you might have noticed that changing the scale has weird effects because it distorts the domain rather than changing the size of individual features. Either we just accept those limitations, or these cases need some special care on our end to make them usable. Another side of this is that there are plenty of things where we would have to reengineer the shaders to make every parameter pluggable, in a way that would make them less efficient. These things are possible but we're not attempting them just yet. In certain cases, such as the Cloud Layer, we identified particular needs/requests and came up with a solution for them, and we'll probably keep doing that on certain nodes as part of ongoing development. The specific case of modulating the size of the stones in the Fake Stones Shader is also something I've thought about. I'm also interested in making it produce a range of different sizes within one node, and after something like that is working it would be relatively easy to make the size controllable with another input. Sadly that's not something we're planning to do soon, because we have some pretty big projects we need to concentrate on in 2012.
Chris, the idea of modulating the settings on the Alpine Fractal Shader came up before and I'm keen on making that more pluggable in future.
Hi,
I have been following this thread. I remember reading (I had thought it was this thread, now I see that it is not) that people are achieving a global erosion effect by using another program with TG2, was it world machine or something like that? Can someone point me to the information Im looking for? I would like to look into this subject.
Yep that's correct.
There are 2 softwares which do this:
World-Machine 2
Geocontrol 2
Google will help you find them.
Some people favor one over the other.
I think WM2's erosion is more realistic, but if I have to believe others GC2 offers more easy control over basic terrain shapes and layout functions for things like river and roads.
I use World-Machine 2 and actually never investigated GC2 to see the differences.
Perhaps Oshyan has more experience/information as he's pretty familiar with tons of 3rd party apps in regard to terrain and satelite stuff and such.
Cheers,
Martin
Quote from: TheBadger on December 06, 2011, 04:49:47 AM
Hi,
I have been following this thread. I remember reading (I had thought it was this thread, now I see that it is not) that people are achieving a global erosion effect by using another program with TG2, was it world machine or something like that? Can someone point me to the information Im looking for? I would like to look into this subject.
Unfortunately only local and not global. ;)
In my assessment over the years of using both world machine and geo control, both have there perks.
World Machine:
Nodes which means doing wacky geomorphology effects that are impossible without a node system. Macro nodes which means making your own terrain effects. Rotating 3d viewer after rendering is done. Many great masking tools as well within the node system. The layout feature is very nice for shaping terrain with drawing them and adding shape primitives. The advanced perlin noise is great. A plugin system which has been touched very little in my honest opinion. Some nifty color shading nodes and lighting nodes.
Geo Control:
Erosion filters that are unmatched by anything i know of. It acts more like in nature, you can mix many types of erosions which ebb and flow into one another just as in nature and there is a lot of types with some erosions that were only cracked within this software itself from it's developer. Rivers tools that conform to the terrain and even fill in lakes. There are road vector tools that can be drawn with a spline that conforms to the terrain. A selection system for easy way to make masks for external use of mapping shaders. A gradient map shader with the option to use any color gradient to shade a selection and so on. Iso Lines for using a spline for drawing terrain shapes.
My advice over the years. Use both. Make some neat stuff in world machine and erode it in geo control. Both are important to have in one's arsenal. I tend to think folks underestimate as to how powerful geo controls erosion really is. The only real lacking in the erosion area in both applications is deep and long talus/sedimentary fans. This effect has not been cracked yet.
Thanks for that info Chris :)
Hasn't Geo Control developement ground to a halt? I read that very small updates are released now and again but nothing truly big has been added for a long time. I think GC2 came out like 3 or 4 years ago now?
Quote from: reck on December 06, 2011, 06:05:12 PM
Hasn't Geo Control developement ground to a halt? I read that very small updates are released now and again but nothing truly big has been added for a long time. I think GC2 came out like 3 or 4 years ago now?
Not really, just slow development and lack of communication from the developer. (No development blogs or hints of new stuff) There are bug fix updates but no new significant features that i am aware of. That is all i know thus far.
Quote from: ChrisC on December 06, 2011, 08:02:12 PM
just slow development and lack of communication from the developer. (No development blogs or hints of new stuff) There are bug fix updates but no new significant features that i am aware of.
I think this is what I meant actually. Chris maybe he has been working on a mega update for the last few years in secret and will release it to the public soon :)
I know Planetside development is frustratingly slow but at least it feels as if the product is moving forward. When updates are released there are some big additions and not just the odd bug fix and terragen remains an incredibly stable application.
Quote from: reck on December 07, 2011, 07:31:00 AM
I think this is what I meant actually. Chris maybe he has been working on a mega update for the last few years in secret and will release it to the public soon :)
I know Planetside development is frustratingly slow but at least it feels as if the product is moving forward. When updates are released there are some big additions and not just the odd bug fix and terragen remains an incredibly stable application.
Sometimes in development that does happen. Just when the user base thinks all is stagnant, a significant update is released so you never know.
I think for myself i never had any frustration with the slow development on the part of Planetside. It is what it is and i understand why it is.
Thank you T-U and chrisC
I'm going to be looking into these things. Seems to be a must know subject.
Quote from: Oshyan on November 23, 2011, 06:12:43 PM
As you note, this is a feature that is not frequently used. Consequently we can't really justify putting much dev time into it. There are better dedicated heightfield editors out there with more robust erosion functions so that's what we'd recommend if you need something beyond the basic capability TG2 currently provides.
- Oshyan
Hello, what are these better dedicated heightfield editors? Do you mean another program all together or a plugin?
Thanks
Quote from: Vortex on December 09, 2011, 02:31:00 PM
Quote from: Oshyan on November 23, 2011, 06:12:43 PM
As you note, this is a feature that is not frequently used. Consequently we can't really justify putting much dev time into it. There are better dedicated heightfield editors out there with more robust erosion functions so that's what we'd recommend if you need something beyond the basic capability TG2 currently provides.
- Oshyan
Hello, what are these better dedicated heightfield editors? Do you mean another program all together or a plugin?
Thanks
Previous page, bottom 3rd message ;)