Integrating a larger terrain in TGE project

#21 Stephen Zepp - I appreciate the information that you've imparted if not it's delivery.

There are good reasons from my stand point for having larger terrains but my game and game play will not be directly affected if I don't have it. I've already been working on the problem and have a creative solution that might just work for me. It doesn't stop me from exploring other opportunities and there seemed a good lead in this thread over the last few days.

It turns out with your information that this is not true. There are real and negative aspects to consider in regards to implementing the Gorpe resource or contemplating extending the existing TGE terrain engine. Fine. Not a problem for me.

But what I can't abide...

You know, I just deleted a whole paragraph here because you've made me so angry with your reply.

I can't express how angry you've made me with the tone of your responses.

I accepted the information you've imparted and am grateful and enlightened BUT what I do not accepted is the tone of your delivery.

I especially do not appreciated this from an employee of a company that I purchase product from.

Good Day Sir!

(Okay, that last part -"Good Day Sir!"- was kinda funny and pathetic but... - piss on ya all the same :) )

#22

Quote:
I guess what I'm wondering is Stefan is can you give one reason why you would use outdated single threaded paged terrain hacks over a system design to solve the problem in a more robust and performant manner?

It is just that it currently is more useful to me. In a year it probably won't be, but struggling without the ingame editors was a pain and we went down that path a few times and always switched back, massive or not.

#23 @James: Out of curiosity, how in any way did you take what I said personally? You posted exactly twice in the forum, and they were "me too" posts.

My comment was directed at Stefan for the most part, and in some part the original poster.

If you wish to take offense at the truth, regardless of the delivery, then that's your choice--especially if the specific comment wasn't even directed at you.

Quote:
In a year it probably won't be, but struggling without the ingame editors was a pain and we went down that path a few times and always switched back

@Stefan: I guess that's what confused me--the Gorpe build doesn't have (cross-terrain block) built in editors either. In fact, you have to stitch your terrains together by hand, have to share terrain textures across all of the paged terrain blocks, and a variety of other things that make it even worse of a hassle than Atlas-without-editors.

#24 No need to implement it then, thanks for the info.

#25 If TGEA was close to useful/stable maybe this would not have been an issue. But at it's current state it is really just a drag. I piad for TSE a year ago or so, and have yet to find ANY use for it, as it is EA, it is given that it can't be used for production, but it has not gone anywhere in over a year.

Larger terrain CAN be tackled via paging. I encourage anyone interseted to look into paging techniques a little further. For larger maps, especially MMO's, paging is the best bet and has been used in many many commercial large scale MMO's, mostly due to its ability to easily integrate with server zoning algorithms/resource allocation.

That is not to put down ATLAS as it is a very powerful terrain system and I will really come to enjoy it more once GG starts to give TGEA a little more attention (apart from changing the name).

#26 Curious, what is the platform goals for TGEA? Will it just be a Windows only?

#27 I'm verry happy to see this thread coming up to life.

I may repeat myself at some point: but we have to admit GG has no interest in supporting a larger terrain in TGE. It would compete too much with TGE-A. With recent agreement between GG and Microsoft (which is a good point for GG and Windows developpers), I'm fearing Open GL port of TGE-A is already a dead horse.

I have been playing around exented TGE terrain with a squaresize set to 32 for Darkwind: war on wheels, as I'm the terrain maker for that game. The size is becoming quite interesting. The only relevant problems are loss of ground details, problems with water. Though our own maps are mostly desert located, water problems weren't really relevant.
If we could achieve such size as the default terrain with a proper ground detail, it could cover most developpers need. Specially if we can then extend it to a large scale, as we are currently able to.
That's not a very big step. Just an important step.

STef

#28 TGEA is also not an option for those of us who need cross platform capability.

#29

Quote:
The only relevant problems are loss of ground details, problems with water.

I was under the impression this was 'fixed' in TGE 1.5. Have you checked it out?

#30 Anthony, with all respect I don't think you have any idea what you're talking about. TSE is already very stable, has most of the shader features you'd want, and is set up in a logical way. Now I no longer even use TGE just because TSE is that much more optimized.

#31 @Stefan: no I did not, because Darwkind: war on wheels is built upon TGE 1.3 and won't be upgraded to another TGE version, due to complex changes in source code.

I own TGE 1.5, but haven't played a lot with it. I like it alot aside the terrain size stuff. 8-)

STef

#32

Quote:
Anthony, with all respect I don't think you have any idea what you're talking about. TSE is already very stable, has most of the shader features you'd want, and is set up in a logical way. Now I no longer even use TGE just because TSE is that much more optimized.

We have used TGEA on quite a few computers and none show any problems. If they do, it is usually a matter of upgrading drivers. However, dismissing what Anthony says (especially that last paragraph) is ignorant.

We are using legacy in our project because so far it has been so much easier to use than Atlas. Atlas is great and the terrains we had were awesome, but to get where we could use them was a pretty bumpy road with dedicated server builds not working, RGB values being reverted, command line tools and lack of ingame editors. It just was not worth the time.

Other than that, I agree.

#33 As a qualifier for my posting a comment: I have no opinion on any of what's been thrown out as hurdles or bigger concerns; I'm an artist first and foremost, I sort of understand the technical issues, and I don't really have an active 3D project at this point anyway.

The comment that really got my attention in this was:

Quote:
If we could achieve such size as the default terrain with a proper ground detail, it could cover most developpers need.

From an artist's perspective, if I'm understanding what Stephan is saying correctly, it's not that I'd like to have terrain that covers *more area* so much as I'd like to have more detail *in that area*.

When I make a test level to place content in, or work with a group of ground textures, I set the terrain square size to 4 so that I can get crisper texture resolution and nicer terrain features. I can accept a zone at that size - that's still a ton of space to be developing content and gameplay for. However, if I could have the area coverage of a square size of 8 with a resolution equivilant to square size of 4 (dare I dream for something as small as 2) I would be *really* happy as that would be a ton of detail in a nice large space. In short - greater mesh resolution with the same area coverage.

The other option that I think of first as an artist is higher resolution terrain textures. I've seen the reasons for that not being an option though, and while I am a TGEA licensee I haven't actually looked into Atlas options more than briefly. (I was initally intrigued by the Huge Terrain Creator but got turned off by the notion of paying for an early adopter tool for an early adopter engine. Perhaps later)

Anyway, I thought I'd bring a different perspective to the conversation.

- Don

#34

Quote:
If TGEA was close to useful/stable maybe this would not have been an issue. But at it's current state it is really just a drag. I piad for TSE a year ago or so, and have yet to find ANY use for it, as it is EA, it is given that it can't be used for production, but it has not gone anywhere in over a year.

3 AAA, 4 multi-million dollar commercial/simulation, and 5 console targetted AA game developers disagree with you.

Quote:
Larger terrain CAN be tackled via paging. I encourage anyone interseted to look into paging techniques a little further. For larger maps, especially MMO's, paging is the best bet and has been used in many many commercial large scale MMO's, mostly due to its ability to easily integrate with server zoning algorithms/resource allocation.

Which is exactly what Atlas does--multi-threaded, using background loading and preprocessed level of detail chunking as well as clipmapping combined with shaders for a terrain engine that is generations beyond legacy TGE terrain, at 15-20% faster processing.

I am however at a loss to see how you are linking terrain paging within a client executable with server back end architecture instance mangement and resource allocation, since the two are as unrelated as can be.

Quote:
That is not to put down ATLAS as it is a very powerful terrain system and I will really come to enjoy it more once GG starts to give TGEA a little more attention (apart from changing the name).

Full dynamic lighting, XBox360 as a platform layer, fully abstracted graphics layer, restructing of Atlas into Atlas 2 for further optimization as well as full 3D poly editor import/export, and literally hundreds of other updates doesn't count as attention?

Quote:
With recent agreement between GG and Microsoft (which is a good point for GG and Windows developpers), I'm fearing Open GL port of TGE-A is already a dead horse.

The two are absolutely, unquestionably, without a single doubt completely unrelated. The partnership with Microsoft is for Torque X, which is written in C#, and has nothing to do with TGE-A (or even TGE) whatsoever.

Quote:
TGEA is also not an option for those of us who need cross platform capability.

At this time. In fact, the abstraction of the graphics layer, which in turn allows you to implement any rendering device you wish makes TGE-A more cross platform capable then any GG offering.

#35 ... Maybe I should be more specific next time. Sorry for the confusion. I was not trying to put down GG.

#36 Or TGEA, if I didn't see amazing potential I wouldn't be upset. I am a customer for a reason :)

#37 So, what is the latest word on OpenGL (Linux/Mac) for TGEA ? If I felt sure it would be done, I'd whip out the credit card right now ( well, friday anyhow ).

#38 Hold on your horses people. It's not the right place to start a flamewar.

Could we study possibilities of improving heightmap details for more than a second? Currently having a two times more detailed ones (512x512) would be much appreciated.
What are the requested changes?
What are problems awaiting us?

Thanks.

STef

#39 It's not about TGE vs TGEA to me because I own em all and intend to use them all. To me it's about being able to use the cross-platform technology (TGE) in an environment which I have modified extensively and am used to working in and find works well. I know that given some more time TGEA is going to be my primary dev platform, it would just be nice to have the choice to use some larger terrains in the older technology is all.

Especially now that ArcaneFX is out for TGE, it's particularly inviting to use...

#40 Yet another post I come across, where someone brings up the topic of a better terrain manager in TGE... only to be shot down, with half the post going adamantly off topic. I wish you guys (GG) would just make a frickin' announcement telling us to lay off and buy TGEA, or shut the hell up. You do it enough in more subtle words as it is, why not make it official?

Stephen- I've seen multiple posts where you list the oh-so wonderful terrain features of TGEA, but not once have I heard ANYONE talk about the lovely new terrain features coming to TGE. Actually, haven't seen one post related to awesome new features coming to TGE at all. Would it be right to assume that you guys have had meetings and come to the conclusion that you won't be supporting TGE anymore? Seems about obvious to me. Unless you count a dinky exotic bugfix release as a feature update...