How can I position a population at the side of my planet?

cyphyr · February 17, 2009, 12:11:11 PM

Hi Guys
I want to add a population at the "side" of my planet, a little north of the equator actually. As far as I can tell the population node is limited in its vertical "y" positioning. When I move the bounding box of the population node above the position I want its at nearly 90 degrees to the curvature of the planet at that point. This means that the population lands in completely the wrong place, its dimensions become distorted and the objects remain at 90 degrees to the plane of the ecliptic, not the globe. Solutions, suggestions or am I once again barking up an impossible tree ?

Thanks
richard

rcallicotte · February 17, 2009, 12:16:52 PM

I haven't tried this. Sounds like something to check out.

Mohawk20 · February 17, 2009, 06:00:43 PM

You could always rotate the object itself, but I have no idea how to rotate the population plane.

Oshyan · February 20, 2009, 03:45:40 AM

Unfortunately you cannot currently populate correctly on the side of the planet. The populations are projected from a flat plane relative to the planet so they become less correct as you get further from the coordinate origin. If possible I would recommend repositioning your terrain to be nearer to 0,0 where your populations should work correctly. This is also something we do plan to improve in the future.

- Oshyan

cyphyr · February 20, 2009, 08:54:33 AM

Thanks for explaining that one Oshyan

This dose mean that for the time being I cant do an animation that involves populations anyone other that near the 0,0 origin. I hope this is one of the issues to be resolved before the final, seems kinda important to me. Global scale forests, solar flares really big cities etc would all benifit from this working properly.
thanks again
richard

Oshyan · February 20, 2009, 10:30:25 PM

Solar flares? Global scale forests are already not possible because of memory limitations in a 32 bit application. Even in a 64 bit application TG2 can only use a max of 4GB of RAM, which won't be enough to have more than maybe 50 million instances of a simple object at max. Until there is a 64 bit version, or we make a breakthrough in populator optimization, it's not the population projection issue really holding things back on the global scale. I do agree it needs to be fixed of course, but I don't think it will happen before the final at this point. Generally speaking you can populate some pretty large areas without noticing issues, especially if you start at the coordinate origin. It's surprising in fact that this is an issue for you, unless you're planning to animate across 100's of kilometers, which would either be an extremely fast camera move, or a very long animation...

- Oshyan

cyphyr · February 21, 2009, 02:57:55 AM

yep a very long animation was the plan,

Thanks
Richard

mt_sabao · February 22, 2009, 07:00:24 AM

hi,sorry to highjack the thread, i'm also trying to place a crater very far from the origin (0,0,0), and i'm getting very weird results ~(750km,-130km,-104km)!
Is it possible to do? And how?

As a side thought, isn't it weird that we have rectilinear coordinates to place things on spherical worlds? Wouldn't it be easier to have spherical coordinates also (radius, angle1, angle2)?

mt_sabao · February 22, 2009, 07:09:03 AM

hmm, just noticed also that very far from the origin, altough my terrain is flat, it has a slope (as regards to the rectilinear coordinate system on the origin. this can be extremelly confusing for constrains and whatnot $:-\$

edit: here's what i mean:

mt_sabao · February 24, 2009, 07:41:56 AM

bump.

no one? should i then conclude it's not possible to place a crater far away from the origin?

edit:
and also, if i add a twist and shear, which is shearing on the Z axis, does this shearing go with the planet curvature, or on the planet's equator (where z is up, regarding the usual terragen axis on the origin --say, the north pole. ) we'll have no shearing?
It's a bit confusing and illogical, or am i making it more confusing than what it actually is?

Thanks in advance

rcallicotte · February 24, 2009, 08:36:09 AM

This has been answered in the forum before. Do a search. I don't have time to do it for you.

cyphyr · February 24, 2009, 08:45:04 AM

No its not illogical, I think but it is confusing. I tried your crater experiment and yes it stopped behaving itself the further from the origin it was moved. I've also noticed that the preview cursor height info (how high is the mountain I'm pointing at) is wrong or misleading the further from 0,0,0 one is. I've seen many posts having problems about the water level appearing in the wrong place, could this be related?

I don't think its illogical but I still (after 2 years?) cant get my head round TG's coordinate system properly. Basically for the time being if there's any location specific landform modeling your performing then do that's as close to the zero point as you can.

hope this helps

Richard

ps stuff like this needs to be addressed either in the software, so it works as the AVERAGE user expects, or in the documentation so these limits can be properly understood before the upcomming GOLD release.

mt_sabao · February 24, 2009, 09:12:28 AM

Quote from: cyphyr on February 24, 2009, 08:45:04 AM
No its not illogical, I think but it is confusing. I tried your crater experiment and yes it stopped behaving itself the further from the origin it was moved. I've also noticed that the preview cursor height info (how high is the mountain I'm pointing at) is wrong or misleading the further from 0,0,0 one is. I've seen many posts having problems about the water level appearing in the wrong place, could this be related?

I don't think its illogical but I still (after 2 years?) cant get my head round TG's coordinate system properly. Basically for the time being if there's any location specific landform modeling your performing then do that's as close to the zero point as you can.

hope this helps

Richard

ps stuff like this needs to be addressed either in the software, so it works as the AVERAGE user expects, or in the documentation so these limits can be properly understood before the upcomming GOLD release.

i mean it's (a bit) illogical to have a x y z coordinate system, when we are dealing with spherical worlds. why not just have a (r, lat, long) as well for the placement of objects?

rcallicotte · February 24, 2009, 09:23:55 AM

I agree. The coordinate system will need clarification, if I understand your need appropriately.

Oshyan · February 25, 2009, 12:18:12 AM

Yes, we're aware of the coordinate system incompatibilities and do plan to address them in the future. It's a bit trickier than it seems though, and that's why most other software hasn't managed a global scene basis.

- Oshyan