TG2 Technology Preview issues

Started by commorancy, February 20, 2008, 07:27:09 AM

Previous topic - Next topic

Sethren

#15
You can fake fluid dynamics with displacement to a certain degree using function tricks that may or may not be available yet. I don't think anyone wants to go to the extent of full of fluid dynamics by means of science simulations inside TG 2 nor micro-scale phenomena. If you look at what old TG 0.9 did for foam and wave ripples it was a good start and i know that can be taken a bit further.

Bubbles, just add a new shader and have it placed in the right locations by means of masking, image maps and even terrain effectors.

I am telling you you can get away with a heck of a lot of shader tricks to fake water. Were not talking about putting Realflow inside TG 2.

That's right, illusion. We don't need hard core precise simulation to fake water features.

I have create crashing waves and a wet beach with foam in Genetica, a freaking texture program using nodes. Now why can't we take that into a 3d context. All i did was layer materials, noises and placed them accordingly.

Cyber-Angel

Well, I tried and thats that. If you want to do it properly (For use in motion pictures, say) I have stated the kind of things you need to do or it won't look correct, and if there is some thing that the human brain is vary good at its spotting some thing thats not correct. I am going to stop now as I really am getting far to old for this stuff, and going round in circles, having to repeat my self and having stuff thrown back in my face; when all I try to do is help in the politest way I know is really no fun any more.

I maybe one of the only people here (and there may well be others) that is of the opinion that you can have hard science and fun at the same time; both can exist in the same universe and they will in my life time, if that proves to be the case then i will say "I told you so" and will point back to what said here just for Ha Ha's.

For the record I see the potential of this software for what it can be down the road not for what it is, what I do today I do for future history and like so many in history I may well be proved right in the end, just as so many visionaries and thinkers before have been.

Its not that I don't respect other opinions, I do. Any way I am done with this subject any way, we'll see how things work out weather they be right, wrong or indifferent.  End Of Line.

Regards to you.

Cyber-Angel  ::)

                 

Sethren

#17
When i fake plate tectonics in World Machine or when very good fluvial erosion is done in GeoControl it looks pretty correct to me and i have been studying geo-morphology for a very long time now and i know what my eye sees as horridly fake and as fairly convincing.

I hope you don't think i am throwing anything in your face because i am not. I respect your and anyone else's opinion but i am not about to give up and say it can only be done with advanced full-on scientific simulations. I have gotten around that with using the terrain software that was mentioned. Water is a bit more difficult but not to the extream of what we tend to think it is. It's all about fakery, a good enough illusion with cleaver use of nodes, shaders and some added code could trick the eye into thinking what looks real and what does not. Look at the maxwell renderer for example, the most highly photo-realistic renders i have ever seen and yet with some cleaver coding things such as that can be done.

Also look at what Cajomi had done for his river tool in GeoControl. All he did was take just a little bit of science, nothing hard-core and yet through cleaver coding he was able to create the most convincing river creation tool i have ever seen that surpassed Bardens river plug-in for MojoWorld and that guy is a physicist. Rivers are a very hard process to simulate and yet he did it.

I have nothing against hard-science at all but honestly it is not necessary in the case for artists like myself creating life-like terrain and water features. I am an artist though and not a physicist so i don't need to use nor think about hard science to create realistic terrains and the like. If that is what you want then i respect that.

I can see new nodes and new ways of faking terrain effects and water effects but i can't imagine fairclough straying away from using good old math arithmetic and cleaver coding with having to be a physicist in order to do it.

I am a thinker to but also i tend to be open minded and a dreamer, a silly imagineer and say it can be done many ways and nothing is just limited to just deep text book ideologies.


Oshyan

Sethren, I would tend to agree with you in concept. The vast majority of CG work these days is accomplished through "cheats" and other non-simulated approaches to realism. The best shore waves I've seen done - essentially totally convincing to the untrained eye - were in Flags of Our Fathers and done by Digital Domain and while they used some degree of "simulation", there was also a lot of "fakery" involved. So I think that's a valid approach. But it's still extremely difficult to get anything that is even close to convincing and it's really not as simple as your proposed solution, at least not in practice. What you suggest would certainly work as a basic implementation, but it simply would not be very convincing in most situations. The versatility of that solution is also questionable.

I do maintain some hope that people will be able to figure something out using the function nodes. Lord knows the hurricane stuff is impressive and nearing some really amazing and realistic results. But waves will be harder...

- Oshyan