Why Mojo Users Have Not Adopted TG2 (and some solutions)

[efflux]

Hi mhaze.

It's Mad, this thread. Starting out as critical and then a solution. The key is in ignoring the 0-1 colour settings in the distorting fractal. You set up big scales for that fractal but then use the colour to set a range of distorting scales. This is crucial because the perlin alone doesn't give you the power of a fractal and the fractal warp effect is best disabled. I hate it. I can't believe yet another problem is solved. It could be tricky to work this out but any problems and I can help.

The clamp multiplies the altitude several times to stretch out the vertical displacement then clamps this stretching, Say at 0 to whatever altitude. In my example the higher clamp value is not actually set at an altitude that has any effect - it's higher than most terrain. I don't want flat tops on the hills.

[efflux]

Another point is that I heavily use the spike limit on fractals to get sweeping shapes with no nasty spikes. Set it right down to a very low level. You'll need to increase displacement amplitude when you do this.

[efflux]

Next stage. Demonstrating my distorted perlin strata function which supplies colour and displacement. There is a tgc file on the Planet Pig thread with this. I've no time to continue now but at least some things are sorted.

[mhaze]

Thanks

I was clamping max far to low - still some way to go but at last I think I'm beginning to make some sense of it all

[Oshyan]

I think what it comes down to almost anything is possible (possibly given the right basis functions), it's just a matter of knowing (or figuring out) how. I would agree that some of Mojo's functions are useful in achieving certain results easier, but I question whether those results were immediately obvious or achievable (with good quality and controllability) in Mojo either. I remember the early days of Mojo and how much people stuck with many of the defaults and the same type of planets. I have seen how really few people in the Mojo community have ever even gotten very good with Mojograph. So I think at least a part of this is really just trying to come at TG2 from what you're used to in Mojo, some of which really does make sense and which we'd like to incorporate into TG2. Some of it is also just obscurity or lack of documentation, but those problems will also be solved in time.

Ultimately I think there is a remarkable amount of power and flexibility here already, especially considering the limited number of basis functions and the lack of externalized/networkable parameters. It will only get better from here, building on what we have, and I think it's safe to say that TG2 will grow beyond Mojo in most respects over time. As has been said it's already ahead in several important areas.

So while I see Mojo'ers point and respect their needs as users and that they may not be met by TG2 at this time, I also think it would be valuable to them as artists to experiment with TG2 and the *other* ways of creating the effects they want. After all for the most part it's the end result (the image, the art piece) that matters.

- Oshyan

[efflux]

The same thing seems to happen every time. I think it can't be done then there is a way but this is a cool direction because it's very related to things you'd immediately find in Mojo. I have already linked these pictured to a Mojo user who is open to trying TG2. Not one of the TG2 belongs in the recycle bin users. It will take a lot more to get them interested 

However, I understand totally why they believe this.

[efflux]

There is still a problem though. Even on terrain scale distortions like this, the shadows issue rears its head but that perlin billows terrain before the first colour picture I posted, is a really nice terrain and a basis for others I have worked on. I'm not actually so keen on ridged terrains. I might still try to tweak this a bit further, say try the perlin mix basis but I can't spend long on it.

[efflux]

Here's another test to show where I'm trying to head. There are an absolute minimum of nodes here. The strata function also plays a big role in providing that kind of flowing detail to give variety and scale. This terrain has endless sweeping variety and even appears so from high up but retains the fractal detail at close range. Maybe a bit too warped to be truly realistic but it has a hint of that Mojo lyricism.

[efflux]

I have this all sorted. Complete Mojo to TG2 crossover (at least every aspect that can be crossed over) but TG2 has the better atmospherics and GI which obviously shines through in the renders. Other than that, this image uses the same techniques as can be used in both apps. All the basics of how you'd do things in Mojo for all the components - distorted fractals, stratas and colour gradients. The core ingredients of a Mojo planet as long as you don't count Mojo's isosurfaces but that's not a necessity. The actual graph is not so simple in TG2 but it has that simplicity of function yet complexity of result as in Mojo. This is the finale image. A true Mojoworld in TG2 but it's better than in Mojoworld.

For those of you wanting to try this. In my previous graph shot, set up the power fractal attached to the warp shader's shader input as you would for the terrain normally. The clamp influences things here but just forget the clamp function to begin with. Then set up the other power fractal in a similar fashion but enter a huge value into the high colour setting. Maybe 10 times that of your largest feature scale. Then enter another value in the low colour a bit below the high colour setting. Set the warp lead in to nothing on both fractals. We don't want that.

[dhavalmistry]

awesome result efluxx

[efflux]

You'll have to play around with the distorting fractals settings a bit. Make the smallest feature size quite big or reduce the colour roughness for example. I'm making you experiment rather than giving a tgd. Once again, if it's a problem I can tweak some files.

[dhavalmistry]

no I prefer this way rather than TGD...

[child@play]

  amazing efflux

apart from the colours i used ( so 70's ), am i on the right track here? i should try adding some smaller scaled displacement too, that's looking really awesome in your image

[rcallicotte]

efflux, a TGC is always nice to see what values are used.  I learn better this way, but it's up to whoever feels they want to share this.  I'm learning, nevertheless.  Matt's TGC helped me to see the values he used to better understand the setup I had already created by his drawing and explanation.  This is the best way to learn, I think.

[efflux]

You're on the right track, child@play.

That is the result you should be getting. It's pretty extreme but you've got it working. It's just a question of tweaking the settings and you'll see what happens whether you want it really warped up or more natural.

Calico. I've supplied a tgc of the terrain from that screen shot with the graph, since the settings for this distortion are not so easy to understand but it's the same principle I used with the other distortion techniques. child@play got it working and by tweaking things wrongly you learn. Also, many of my files have a lot of time spent simply getting a nice result rather than just testing a technique. This part should be up to everyone to experiment with but I understand this is a bit tricky so here is a file.