Moving clouds ?

[old_blaggard]

Honestly, I'm not sure if that would work.  If you randomized the seed every frame, I'm sure it wouldn't, but if you incremented it by 1 or 2 every frame... anyone want to do a test?

[BPauba]

Just a thought here since i dont have the animation version of tg2...but would it be possible to get that "variance" that others on this thread have mentioned by animating the "seed" of the cloud density shader from say frame 1 to frame 50. so over those fifty frames the cloud density morphs and we get variance. Just an idea anyway 

From what I know the seed does not have a "progressive randomness" that follows the number line. What I mean by that is seed#1 is going to be just as different as seed#550 when compared to seed#2. I do not believe any seed is more similar to one seed then to another. So what I believe will happen if you manipulate the seed number as the frames move forward is just completely new cloud formations that will not morph into eachother. I have never tested this hypothesis out though, so please do not take my word! Just adding to the conversation .

[Tangled-Universe]

I second that, the seed number is a "random" input which results in totally different cloud formations.

[Njen]

Just a thought here since i dont have the animation version of tg2...but would it be possible to get that "variance" that others on this thread have mentioned by animating the "seed" of the cloud density shader from say frame 1 to frame 50. so over those fifty frames the cloud density morphs and we get variance. Just an idea anyway 

Seed provides (seemingly) random values with each integer. A Seed value of 1 will look totally different to a seed value of 2. Seed values in any 3D program are rarely animatable.

As I mentioned earlier, a new 'phase' value that will solve this trick is needed (similar to what other 3D programs have to do the same kind thing).

[Dark Fire]

I think the seed values are put into the internal algorithms to help to generate (apparently) random clouds while providing the user with a way of keeping a specific cloud formation.

[RealUser]

Rrrrrrrr! Now i am aroused!  My english is too limited to express how interesting all this weather stuff is to me. IMHO, weather "simulation" will be a plug-in task. I don't think it will ever be a regular part of a software like TG2, which is in my eyes an artistic tool and no "Earth Simulator". Isn't in Japan something like that?
I would love to see shaders that simulate atmospheric behaviour like Cyber-Angel and stevehmeyer described, with not too many values. For example: A particle animation system for clouds and other gaseous matter would be extremly usefull for faking tornados and other weather phenomenons. Forces like friction, velocity, wind, pressure and others should be part of it.
In the far distance i can see real weather rendering, but this is a matter of understanding the atmosphere and the weather system, as many meteorogolist claim they don't fully understand it, and number crunching power of our desktop computers. Maybe in ten to twenty years we are able to release all the horsepower for tasks like that. Until that time, we can fake everything with our artistic skills. 

[Cyber-Angel]

As far as I am concerned on this issue it maybe possible to do these things sooner than we may imagine, it is likely that the trend for multi-core CPU's will continue into the foreseeable future, and it can't be long given the law  of averages before we see CPU's with 100 or even more cores on the same chip, this is going to affect the way software of the future is designed and programmed.

It is a fact that even today's dual and quad core chips are in many instances not been used to the full extent of there capabilities and further more, many of the multi-threaded software titles out there only use a small amount of the capabilities of the modern chips.

Can it really be that long before we see a CPU with Petaflop or even exoflop performance? Once we see this kind of performance on home computers and providing power and chip heating can be kept to economic and sustainable levels it maybe at that time to start thinking about parallel computing on the desktop.

With computers this fast with enough memory realistic weather simulation is Terragen should not be a problem on a system with 100+ cores; maybe at that stage of home computing development you maybe able to dispense with many of the tricks and workarounds prevalent in computer graphics today.

Whats next in computing maybe by the end of this century or the middle of the twenty second century we may see the first quantum computers since these are still hypothetical at best, it is the considered opinion of this author that the author has no idea of what kind of software or indeed what software would look like for such a computer.

Regards to you.

Cyber-Angel     

[old_blaggard]

Whats next in computing maybe by the end of this century or the middle of the twenty second century we may see the first quantum computers since these are still hypothetical at best, it is the considered opinion of this author that the author has no idea of what kind of software or indeed what software would look like for such a computer.

For the record, quantum computers are already here: http://news.com.com/2100-1008-6159152.html

[RealUser]

For the record, quantum computers are already here: http://news.com.com/2100-1008-6159152.html

Really, where can i buy it? 

@Cyber-Angel: Computer power isn't always a matter of sheer physical presence of multi core CPU's. When i think about the old days of the C=64, Atari and Amiga computers: What did the programmers achieve with the, compared to nowadays computers, limited power? Much of the computational power is eaten up by the system (XP or even worst Vista). Many programs are multi megabyte monsters compared to software of the early multimedia computers in the 80's and 90's. I remember a DTP software for the Amiga comes on one 720kb diskette and one diskette with extras. What i want to say with that is: Computational power ist often waisted for the system or uneconomically programmed software. The best example is Vista. XP runs on many (even older) laptops perfectly, but no go with Vista. If i could run Amigas Intuition/Workbench natively on a PC, it would run as hell! If the "inventors" of the OS's would programm in a more economic way, we would have much more power for the software we want to use. Take a look at TG2TP. It is really small compared to other landscape software and runs pretty smooth after the bug-fixes and look what is possible with this limited version. I would say, good and economic programming is an artistry which many of the programmers out there don't pay much attention to anymore. TG2TP is a positiv example.

But i am getting way too off topic here. Let's return to the clouds and how we can get them look as we want them.

[Cyber-Angel]

He He,

I had an Amiga 500 with the standard ram configuration that came with that system and before that it was a Vic-20 so that's old, old school so yes Old_Balggard I understand what you mean. Look I happen to know more about computers then I care to admit to, but add lest your not having to type Load:/ and press play on the cassette tape drive any more.

What I was trying to get at in the way of truly dynamic, physically accurate clouds in the way that has been discussed previously in this thread would require more powerful computers then are currently available on the home consumer market, or even at professional workstation level computing which is where your more likely then not to see the kinds of system I was talking about implemented first before the technology filters down to the home market.

Yes, in the old days you had to wright clean code due to the physical hardware constraints imposed by what was state of the art technology back in the late 70's and early eighties, and more over the small percentage of the population who had and could afford to buy a computer, with the main buyers back then for what we call a PC today (To user the parlance of the day, an IBM Compatible, in reference to the Open Hardware Architecture pioneered by IBM), with business been the main buyers at this stage in computing history.

I would say IMHO that there are many demands placed on programmers today, call it a product of the commercial globalization of the software industry if you will; with the high pressure, tight release deadlines of the industry we have today maybe with the pressure to perform they just don't have the time for the artistry any more, in my view it is not that they don't want to its just there is not enough time, but I am pulling no punches about software that dose not do what its end user would expect.

I look forward to seeing what the future of both Terrgen and computing have to offer, and just for the record the feeling I had when I switched on a computer for the first time, I still have today; in other words keep the magic never lose it.

Regards to you.

Cyber-Angel 

[RealUser]

Hehe, it was me RealUser not old_blaggard. But i take that as a compliment. My english seems to get better. 

[Cyber-Angel]

Many apologies RealUser, not like me to make slip ups like that

Regards to you.

Cyber-Angel

[Dark Fire]

Whats next in computing maybe by the end of this century or the middle of the twenty second century we may see the first quantum computers since these are still hypothetical at best, it is the considered opinion of this author that the author has no idea of what kind of software or indeed what software would look like for such a computer.

For the record, quantum computers are already here: http://news.com.com/2100-1008-6159152.html

I can't wait until biological computers become powerful and widespread - then we'd be able to literally kill processes.

[RealUser]

@DarkFire: Hehe, like in "Existenz"! http://www.imdb.com/title/tt0120907/