Dynamic Infestation?
Koruyo
AUT Join Date: 2009-06-06 Member: 67724Members, Reinforced - Shadow
AUT Join Date: 2009-06-06 Member: 67724Members, Reinforced - Shadow
When was the last time you heard something about it?
Footage from <u><b>Dez. 2006 </b></u>but in another engine? Ya me too.
Quote from 06:
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->This is a bit unusual as game developers typically keep everything about the game (even it's existence) secret until pretty far along in the development. Although there are some good reasons for this, we've decided to take a different approach for NS2 and really "open the kimono" to show off what we're doing, as we do it.<!--QuoteEnd--></div><!--QuoteEEnd-->
So whats up atm? Open the kimono again...
Footage from <u><b>Dez. 2006 </b></u>but in another engine? Ya me too.
Quote from 06:
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->This is a bit unusual as game developers typically keep everything about the game (even it's existence) secret until pretty far along in the development. Although there are some good reasons for this, we've decided to take a different approach for NS2 and really "open the kimono" to show off what we're doing, as we do it.<!--QuoteEnd--></div><!--QuoteEEnd-->
So whats up atm? Open the kimono again...
Comments
Btw that was in Max's engine running a HL2 bsp map. I think we can safely say that Spark is now the name of Max's engine
If you please, how do you know that Kouji?
@Arcadian the video says its an HL2 bsp map, and somewhere in the forums back then they told us it was maxs own engine.
<a href="http://www.unknownworlds.com/blog/2006/12/dynamic_infestation_1.html" target="_blank">http://www.unknownworlds.com/blog/2006/12/...estation_1.html</a>
The only thing that confuses me about the own engine stuff:
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->At first we though the lighting would also be a tough problem, since Source is based on pre-computed light maps for the static geometry. Trying to illuminate something that dynamically changes at run-time doesn't immediately seem feasible in this scenario, but the facilities built into the BSP for doing normal mapping actually made it pretty simple.<!--QuoteEnd--></div><!--QuoteEEnd-->
edit:
They need to implement it late in alpha because there is no beta if its not feature complete...
I'm sure it was stated somewhere around the time of the engine reveal.
you are just throwing out tech terms that you don't understand - it's actually got nothing to do with this conversation.
Because I am the local community Hινε UÏкεεÏεг <img src="http://members.home.nl/m.borgman/ns-forum/smileys/biggrin.gif" border="0" class="linked-image" />
Observe:
<!--quoteo(post=1683098:date=Jul 11 2008, 07:05 PM:name=Max)--><div class='quotetop'>QUOTE (Max @ Jul 11 2008, 07:05 PM) <a href="index.php?act=findpost&pid=1683098"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Thanks to everyone for the encouragement.
I know some people are wondering how we could possibly build an engine in only a few months. The truth is that we didn't; I've actually been working on the engine in some form since 2005. Originally the engine was just a way for me to try out different ideas (mostly related to 3D graphics) quickly and easily. <b><!--coloro:#FF0000--><span style="color:#FF0000"><!--/coloro-->And if you've been following the blog you've seen it a few times in different forms, including the dynamic infestation prototype video and the map viewer<!--colorc--></span><!--/colorc--></b>. So we didn't really decide to write a new engine, it's more like we decided we already had an engine, we just needed to add in a few more pieces. Adding in those pieces is what we really started doing 5 months ago.
So what are those pieces? The big one was networking which we talked about in the podcast a bit. We also added in collision support for character movement, character animation and physics / ragdolls and a sound system. All of the scripting glue code we wrote for Source just came right over. And since all of gameplay code is written in Lua script it was already very independent of the Source engine and we were able to directly use that with our new engine.
Right now we have every major piece in place in the engine, which is really great for development since it means everyone on the team can be productive without waiting on anything. Those pieces aren't all 100% yet, but I'm going to be continuing to work on that while the rest of the team progresses with the game.
We've been closed kimono about the engine for a while because things were still quite up in the air, but now that we're through that we hope to be talking about it quite a bit more on this blog.<!--QuoteEnd--></div><!--QuoteEEnd-->
I think it's safe to say no work has been done on dynamic infestation since the 2006 tech demo, given the current state of the game, since it's pretty obvious that the tech demo implementation will not work, has probably been scrapped, and will almost certainly need to be re-designed from the ground up.
So maybe using this collision geometry group with the old DI system isn't so difficult to implement.
Can't wait to see some DI in game eventually.
That bug implies that collision geometry is not related to the actual geometry of the level art - i.e. there are gaps between objects in the level art, but no gaps in the collision mesh that can't be shot thru. Collision geometry can be simple planes and cubes, since it isn't rendered, and it only has to be precise enough to keep players from intersecting level geometry. If it's more complex, it could hypothetically be useful for procedural DI to "grow" on, but it isn't particularly relevant to the vertex normal displacements shown in the tech demo.
The bugfix does indicate that shooting between pieces of impassable prop-geometry was not working. This collision mesh would work very well with the DI growth because the DI would fill in these gaps and make barriers much like the overhead pipe area of DI growth in the old video. This would fit the description of DI creating more cover for kharaa lifeforms as it fills in these gaps and nooks.
I'm just saying a workable version of DI in Spark isn't going to be that difficult unless it gets bogged down in trying to be overly detailed. The idea is that an infested area will look almost unrecognizable compared to the same area without infestation. Losing some of the prop-geometry details under the DI will be a feature.
And I'm just saying the version shown in the tech demo (vertex normal-based displacements) isn't workable. I'm not saying it's impossible for UW to create a workable version, just that it's more likely than not that such a version doesn't currently exist.
Agreed.
Players are supposed to recognize an area despite DI-spread.
It would also look silly, if DI covers all the light sources.
I dont not know what the plans are and if they have changed any or been discussed any further, its all a mystery.
<img src="http://www.ns2res.com/wp-content/gallery/Artwork/envconcept06.jpg" border="0" class="linked-image" />
Different lighting, but still a few things sticking out and helping to recognize the area as new player.
DI growth is going to be very very slow, if I remeber rightly Charlie saying so some time ago.. If its going to creep very slowly then the code for it does not need to run checks very quickly.
It will be totally appropriate that leap becomes an upgrade if I can run on any surface in the level without thinking about it. It will take skulk tactics to a new level imo.
The culling was fully done, it just had a very major bug.
--Cory
what does this mean?
As i understand it, the beta will be a stage of the game where most things are implemented and the game just has to be balanced. This would mean that DI doesn't have much impact on the game-play or maybe even none at all