WOW... I am amazed. Especially the sunlight (even though a bit hardcore visible) was stunning. <img src="style_emoticons/<#EMO_DIR#>/smile-fix.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile-fix.gif" /> I absolutely look forward to this game. It just jumped to #2 of my waiting-for-list. (number 1 being diablo3 <img src="style_emoticons/<#EMO_DIR#>/biggrin-fix.gif" style="vertical-align:middle" emoid=":D" border="0" alt="biggrin-fix.gif" /> )
<!--quoteo(post=1689983:date=Oct 11 2008, 05:18 AM:name=lunchtimemama)--><div class='quotetop'>QUOTE(lunchtimemama @ Oct 11 2008, 05:18 AM) <a href="index.php?act=findpost&pid=1689983"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Remarkable work, guys. I was skeptical when you first announced your own engine. A small team doing a graphically competitive engine in a short amount of time? Well, I am very impressed. It really is a testament to the incredible talent on your team. You seem to have both the looks and the smarts: an attractive engine which is also easy to work with. Not to dump on Valve, but I do feel this highlights the limitations of their emphasis on pre-compilation, both in terms of game logic and graphics. (Side note: while I know you've settled on Lua, I recommend you have a chat with the <a href="http://www.mono-project.com" target="_blank">Mono</a> guys at the next GDC - they're doing really interesting stuff with embedded virtual machines in games.)
I agree with the earlier post that radiosity is the crown jewel of Source's lighting, both as it applies to light maps and model lighting. Their "radiosity normal-mapping" technique is really quite clever and it give the impression that models "fit in" to their surroundings. It is a subtle but compelling effect. As mentioned, real-time radiosity calculations are not feasible and dynamically generated "bounce lights" have problems of their own (some of which Flayra mentioned). I'm not enough of a graphics guy to offer another solution, but I urge you to be on the lookout for a good way to approximate radiosity. Another thing you might want to cheat on is ambient occlusion; it really helps models "fit in."
The God Rays are another effect whose key is subtlety. The whirling columns of light slipping into the Hive chamber from a fan 20 meters away are a little distracting. The rays should fall off based on distance and view angle. They are there to establish atmosphere, not remind everyone how cool your engine is.
Which brings me to the larger purpose of graphics. As I have said, this tech demo is very impressive. I am confident that you guys will produce a terrific looking game. But lest we loose ourselves in the spectacle of window shutters, it servers everyone to remember the core strengths of Natural Selection; the reasons I return to NS time and again: mood and gameplay. These technical achievements are perfectly poised to enhance both, and I am sure you've kept these goals foremost in your development process, but allow me to explore them from my point of view (feel free to have thought of all of this years ago).
The Word for NS2 has been "dynamic." Infestation, lighting, environment. All aspire to dynamism. Much has been made of the game's variability, and rightly so. NS2 promises to be one of the most "dynamic" multi-paradigm multiplayer games on the block. But if this promise is to be realized, these individually expressive elements must be successfully combined into a living, breathing, dynamic whole. It must be more than window shutters and light switches (it's a tech demo, I know, and a really cool one). It is not enough to have procedural infestation over here and spinning fans over there. The endgame is a synthesis of all of these dynamic systems into an environment that feels real. That has character. That evokes emotion.
The two hallmark "dynamic lighting" schemes to date are time-of-day, in which the motions of heavenly objects are simulated "in real time!", and the ever-popular "dynamic lights!" which are like regular lights, only you can turn them off / shoot them out. The former offers no enhancement to gameplay (other than visualizing how much time one has spent in isolation from the real sun), and the latter involves deliberate (and tedious) intervention by the player to either find the light switch or take careful aim. Both are little more than curiosities. We are still waiting for the "dynamic lighting" system that means something.
And here I think you are blessed with your setting. Future industrial technology is replete with glowies, blinkies, auras, emissions, and indirect light of all sorts. There's "regular" mode, "power fluctuation" mode, "power failure" mode, "emergency lighting" mode, "proximity alert" mode, "red alert" mode, "lethal radiation alert" mode, "violent decompression alert" mode, and "power restored" mode, just to name a few. The procession of infestation through the map is a ripe excuse to fiddle with the interior lighting design. To excerpt a post from <a href="http://www.unknownworlds.com/ns2/forums/index.php?showtopic=103379&view=findpost&p=1663285" target="_blank">another thread</a>, This touches on one of the three roles of lighting: sensory encumberance. The other two are information conveyance and emotional queue. Sensory encumbering light shifts give one team a tactical advantage by impairing the other team's vision. By forcing marines to use flashlights or night vision (or just suffer dim or sporadic illumination), the Kharaa stand a better chance in toe-to-toe encounters. Information conveying light shifts alert players to some change in game state. The approach of enemies, the completion of hives/structures, the progress of time-dependent variables, the occurrence of major events, &c.. Lighting can also convey information as it changes through space as well as through time: areas of the map near infestation may have different lighting characteristics, communicating the direction of the alien territory. Also, lighting effects can be used to provide specific tactical information. If marines are using flashlights, the light beam will signal their approach and position to aliens in ambush. Hive sight might also include any number of additional lighting effects to convey information on friendly and hostile activities. Valve discusses the use of lighting to subtly convey information in the TF2 developer commentaries - from the direction of the sun, the red side of the map is cast in warm light, while the blue base sits in cool shadows, subconsciously orienting players. Lighting is a remarkable high-bandwidth channel and human beings are given to pick up on patterns whether you point them out or not. Finally, lighting can serve as an emotional queue. Putting players on edge, plunging them into panic, bathing them in relief. Lighting also colors the larger emotional pallet of an environment. The "mood" of a place is in its lighting.
These three are the important functions of light and they are intertwined (e.g., when you impair senses, you elicit emotion). Dynamic lighting is well and good by it self, but its full potential lies in the service of these contexts. Spinning fans and fluttering shutters have their place, but I encourage you to design your lighting around its function rather than its effect. Most of all, leverage these fully dynamic lights to bring your environments to life! Let the lights change with the game. Make them mean something.
And you've thought of all of that before. Great. Can't wait for the next vidcast.<!--QuoteEnd--></div><!--QuoteEEnd-->
Quoted For Truth (QFT).
This post is dead on for how I feel about light as well - like say, if there was infestation near those fans, would they stop? I know, its just a lil tech demo, just trying to emphasize my agreement with that well written post by lunchtimemama.
Well, thank you for sharing and may your work towards deadlines be blessed Unknown Worlds <img src="style_emoticons/<#EMO_DIR#>/smile-fix.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile-fix.gif" />
The lighting of fans and such in ns_hera was very skilled, but definitely lacking in terms of atmosphere - this is a feature I've wanted to see in a *popular* engine for a long time (cough, doom).
<!--coloro:lime--><span style="color:lime"><!--/coloro-->As to performance<!--colorc--></span><!--/colorc-->, what FPS drop did you guys see (total) going from static to dynamic lighting, and what kind of system was that example running on?
Awesome work guys, dynamic lighting will surely help build the atmosphere of NS2. Radiosity is nice, but i dont think il miss it <i>that</i> much <img src="style_emoticons/<#EMO_DIR#>/smile-fix.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile-fix.gif" />
Very good job there, now that's something we can look at while we wait. Don't let us wait that long again. Even if you're only showing us what you had to eat today, I'd still watch it!
I really like the video and am impressed with the progress. I still feel my main concerns however like I said earlier, I still think leaving source is going to delay the game past it's prime release period. If this game takes another year to come out, just don't think it would have the buzz as if it was coming out this year. Anyways you guys have done good work, best of luck in your future.
<!--quoteo(post=1689983:date=Oct 11 2008, 10:18 PM:name=lunchtimemama)--><div class='quotetop'>QUOTE(lunchtimemama @ Oct 11 2008, 10:18 PM) <a href="index.php?act=findpost&pid=1689983"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->...<!--QuoteEnd--></div><!--QuoteEEnd--> Man, what an amazing post. As CanadianWolverine said, QFT!
I would just like to reiterate that I am very impressed with the work so far on engine. Please try to see that when we give these constructive comments, it is because to us new doors of opportunity have opened. We are not trying to criticise your current implementation of dynamic lighting, we are just getting excited over what this might evolve into.
The new engine is looking awesome. I hope the editor is awesome too <img src="style_emoticons/<#EMO_DIR#>/tounge.gif" style="vertical-align:middle" emoid=":p" border="0" alt="tounge.gif" />
Looks great, but the shadows are too sharp IMHO. Seeing the perfectly defined shadows of the fan spinning around in mid-air or the small tendrils on the sens chamber is a little excessive. If it's possible to soften them a bit or at least blur the edges that would help.
Had a *dropjaw* moment when I first saw the video, it's great stuff, exactly how I envisioned how great NS1 would be if it had the same engine that NS2 has. It's gonna be a thrill to move around the map if there's dynamic lighting <img src="style_emoticons/<#EMO_DIR#>/biggrin-fix.gif" style="vertical-align:middle" emoid=":D" border="0" alt="biggrin-fix.gif" /> Make sure you make some fun use of it now, this game has plenty of fun opportunities to have nice light-effects that actually affect the gameplay, hopefully in a positive way.
Also, curious about the Hive, it's glowing by itself but is it really emitting any light? It shows such a sharp line of shadow instead of softly illuminating the walls. Some light entity that is located in a weird spot or?
I am shocked! I was concerned at the beginning when you guys said you're writing your own engine, but now when i saw this... I think this was great idea to drop the Source. It looks amazing!
This looks pretty freaking astounding. Mad props to Max and the team. Looks like we're finally getting closer to the ideal situation of truly dynamic lighting. No more semi-transparent faked light shafts! Yay!
One thing I'd like to see is a limitation of the engine that you've found. Somewhere that just looks wrong. Showing all the good sides of something is good, leaving limitations to the viewer's imaginations is generally a bad thing. I'd rather know exactly where this engine fails rather than leaving it as a vague niggle in the back of my mind.
I don't know about other people, but I always try to find the breaking point of any game I play. Find a situation that the engine or the level design just can't handle. While the rest of the game might be stellar, knowing where it breaks is somehow comforting.
ShockehIf a packet drops on the web and nobody's near to see it...Join Date: 2002-11-19Member: 9336NS1 Playtester, Forum Moderators, Constellation
Lovely stuff Max. You need to stop Charlie trying to look buff next to you though. <img src="style_emoticons/<#EMO_DIR#>/wink-fix.gif" style="vertical-align:middle" emoid=";)" border="0" alt="wink-fix.gif" />
Side topic: We see in the hive room a Sensory Chamber plonked down on the floor. Have you guys got any plans to have the dynamic growth we've already seen 'sprout' into the terrain from the point the Chamber starts? It'd be great to see it look more like it'd grown there, rather than being a distinct, unintegrated entity. Oh, and I'm going to make a random (and probably near impossible at a code level) request. It would be awesomely cool if Cloaking (should it still exist) cause some sort of interaction/disturbance with light effects. Aside from sheer 'sweet' factor, could also have some value in things like areas you can light, and flashlight use, etc etc.
Otherwise, great to see work is still progressing nicely, keep up the good work!
ive posted before on this thread but ive been linking this movie to some friends that played NS years ago and there really impressed with the engine, i really cant wait to try it out! and neither can they, the link is on my vent servers comment and on my forum, hope this brings more hits to ya!
Great job guys I really like what you've done with the engine. Can't wait for the release.
<!--quoteo(post=1690000:date=Oct 11 2008, 01:06 PM:name=Ride)--><div class='quotetop'>QUOTE(Ride @ Oct 11 2008, 01:06 PM) <a href="index.php?act=findpost&pid=1690000"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Very good job there, now that's something we can look at while we wait. Don't let us wait that long again. Even if you're only showing us what you had to eat today, I'd still watch it!<!--QuoteEnd--></div><!--QuoteEEnd-->
I agree <img src="style_emoticons/<#EMO_DIR#>/tounge.gif" style="vertical-align:middle" emoid=":p" border="0" alt="tounge.gif" />
<!--quoteo(post=1689983:date=Oct 11 2008, 05:18 AM:name=lunchtimemama)--><div class='quotetop'>QUOTE(lunchtimemama @ Oct 11 2008, 05:18 AM) <a href="index.php?act=findpost&pid=1689983"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Remarkable work, guys. I was skeptical when you first announced your own engine. A small team doing a graphically competitive engine in a short amount of time? Well, I am very impressed. It really is a testament to the incredible talent on your team. You seem to have both the looks and the smarts: an attractive engine which is also easy to work with. Not to dump on Valve, but I do feel this highlights the limitations of their emphasis on pre-compilation, both in terms of game logic and graphics. (Side note: while I know you've settled on Lua, I recommend you have a chat with the <a href="http://www.mono-project.com" target="_blank">Mono</a> guys at the next GDC - they're doing really interesting stuff with embedded virtual machines in games.)
I agree with the earlier post that radiosity is the crown jewel of Source's lighting, both as it applies to light maps and model lighting. Their "radiosity normal-mapping" technique is really quite clever and it give the impression that models "fit in" to their surroundings. It is a subtle but compelling effect. As mentioned, real-time radiosity calculations are not feasible and dynamically generated "bounce lights" have problems of their own (some of which Flayra mentioned). I'm not enough of a graphics guy to offer another solution, but I urge you to be on the lookout for a good way to approximate radiosity. Another thing you might want to cheat on is ambient occlusion; it really helps models "fit in."
The God Rays are another effect whose key is subtlety. The whirling columns of light slipping into the Hive chamber from a fan 20 meters away are a little distracting. The rays should fall off based on distance and view angle. They are there to establish atmosphere, not remind everyone how cool your engine is.
Which brings me to the larger purpose of graphics. As I have said, this tech demo is very impressive. I am confident that you guys will produce a terrific looking game. But lest we loose ourselves in the spectacle of window shutters, it servers everyone to remember the core strengths of Natural Selection; the reasons I return to NS time and again: mood and gameplay. These technical achievements are perfectly poised to enhance both, and I am sure you've kept these goals foremost in your development process, but allow me to explore them from my point of view (feel free to have thought of all of this years ago).
The Word for NS2 has been "dynamic." Infestation, lighting, environment. All aspire to dynamism. Much has been made of the game's variability, and rightly so. NS2 promises to be one of the most "dynamic" multi-paradigm multiplayer games on the block. But if this promise is to be realized, these individually expressive elements must be successfully combined into a living, breathing, dynamic whole. It must be more than window shutters and light switches (it's a tech demo, I know, and a really cool one). It is not enough to have procedural infestation over here and spinning fans over there. The endgame is a synthesis of all of these dynamic systems into an environment that feels real. That has character. That evokes emotion.
The two hallmark "dynamic lighting" schemes to date are time-of-day, in which the motions of heavenly objects are simulated "in real time!", and the ever-popular "dynamic lights!" which are like regular lights, only you can turn them off / shoot them out. The former offers no enhancement to gameplay (other than visualizing how much time one has spent in isolation from the real sun), and the latter involves deliberate (and tedious) intervention by the player to either find the light switch or take careful aim. Both are little more than curiosities. We are still waiting for the "dynamic lighting" system that means something.
And here I think you are blessed with your setting. Future industrial technology is replete with glowies, blinkies, auras, emissions, and indirect light of all sorts. There's "regular" mode, "power fluctuation" mode, "power failure" mode, "emergency lighting" mode, "proximity alert" mode, "red alert" mode, "lethal radiation alert" mode, "violent decompression alert" mode, and "power restored" mode, just to name a few. The procession of infestation through the map is a ripe excuse to fiddle with the interior lighting design. To excerpt a post from <a href="http://www.unknownworlds.com/ns2/forums/index.php?showtopic=103379&view=findpost&p=1663285" target="_blank">another thread</a>, This touches on one of the three roles of lighting: sensory encumberance. The other two are information conveyance and emotional queue. Sensory encumbering light shifts give one team a tactical advantage by impairing the other team's vision. By forcing marines to use flashlights or night vision (or just suffer dim or sporadic illumination), the Kharaa stand a better chance in toe-to-toe encounters. Information conveying light shifts alert players to some change in game state. The approach of enemies, the completion of hives/structures, the progress of time-dependent variables, the occurrence of major events, &c.. Lighting can also convey information as it changes through space as well as through time: areas of the map near infestation may have different lighting characteristics, communicating the direction of the alien territory. Also, lighting effects can be used to provide specific tactical information. If marines are using flashlights, the light beam will signal their approach and position to aliens in ambush. Hive sight might also include any number of additional lighting effects to convey information on friendly and hostile activities. Valve discusses the use of lighting to subtly convey information in the TF2 developer commentaries - from the direction of the sun, the red side of the map is cast in warm light, while the blue base sits in cool shadows, subconsciously orienting players. Lighting is a remarkable high-bandwidth channel and human beings are given to pick up on patterns whether you point them out or not. Finally, lighting can serve as an emotional queue. Putting players on edge, plunging them into panic, bathing them in relief. Lighting also colors the larger emotional pallet of an environment. The "mood" of a place is in its lighting.
These three are the important functions of light and they are intertwined (e.g., when you impair senses, you elicit emotion). Dynamic lighting is well and good by it self, but its full potential lies in the service of these contexts. Spinning fans and fluttering shutters have their place, but I encourage you to design your lighting around its function rather than its effect. Most of all, leverage these fully dynamic lights to bring your environments to life! Let the lights change with the game. Make them mean something.
And you've thought of all of that before. Great. Can't wait for the next vidcast.<!--QuoteEnd--></div><!--QuoteEEnd--> I agree with this post in every motion, and it hasn't been quoted on this page yet, so here. It MUST be read.
My 2 cents, the lighting beams need to fall off more at long ranges, although, it's really something that should be controlled by the mapper (some lights should look like spotlights, many lights should not though). Bumpmapping and other surface treatments were mentioned... but I really didn't see any good examples of that in the demo, that being said, it can be deceptive looking at a game with both new tech and old tech art assets, because they don't always gel perfectly. My only other concern would be FPS and system compatibility... but meh, I guess with that we'll just have to see. Oh, and of course I feel gameplay can't possibly be taken seriously enough, but we haven't really had any news regarding gameplay designs, so I have nothing to comment on there.
I don't know if this has already been posted but... There's already an engine called the Evolution Engine. Digital Extremes( <a href="http://www.digitalextremes.com/" target="_blank">http://www.digitalextremes.com/</a> ) is using it to make Dark Sector ( <a href="http://www.darksector.com/home.html" target="_blank">http://www.darksector.com/home.html</a> )
From the Digital Extremes site:
Dark Sectorâ„¢ Utilizing the Evolution Engineâ„¢, proprietary next-gen technology developed by Digital Extremes, Dark Sector is designed exclusively for next-generation gaming.
<!--quoteo(post=1690075:date=Oct 12 2008, 11:49 AM:name=SgtHydra)--><div class='quotetop'>QUOTE(SgtHydra @ Oct 12 2008, 11:49 AM) <a href="index.php?act=findpost&pid=1690075"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I found it interesting that even the NS1 artwork looks FREAKING AMAZING with the new lighting system.<!--QuoteEnd--></div><!--QuoteEEnd--> The NS1 artwork was some of the best modeling and texture work on the HL engine. Really, you don't even need to do alot to fully update them to current gen tech, just generate spec/lighting maps and bump maps for them, and they'd look as good as any HL2 model.
[edit] That being said, those models have not been fully updated yet, they look so good simply because they are well modeled and textured.
schkorpioI can mspaintJoin Date: 2003-05-23Member: 16635Members
so now that you guys have the core tech and art standards in place are you going to hire heaps of people and ramp up production to get this thing out the door?
That was UBERCOOL! It kinda looks like almost ready game, which gets me nervous and excited! omgomg cant wait. <img src="style_emoticons/<#EMO_DIR#>/nerd-fix.gif" style="vertical-align:middle" emoid="::nerdy::" border="0" alt="nerd-fix.gif" />
Comments
I absolutely look forward to this game. It just jumped to #2 of my waiting-for-list. (number 1 being diablo3 <img src="style_emoticons/<#EMO_DIR#>/biggrin-fix.gif" style="vertical-align:middle" emoid=":D" border="0" alt="biggrin-fix.gif" /> )
I agree with the earlier post that radiosity is the crown jewel of Source's lighting, both as it applies to light maps and model lighting. Their "radiosity normal-mapping" technique is really quite clever and it give the impression that models "fit in" to their surroundings. It is a subtle but compelling effect. As mentioned, real-time radiosity calculations are not feasible and dynamically generated "bounce lights" have problems of their own (some of which Flayra mentioned). I'm not enough of a graphics guy to offer another solution, but I urge you to be on the lookout for a good way to approximate radiosity. Another thing you might want to cheat on is ambient occlusion; it really helps models "fit in."
The God Rays are another effect whose key is subtlety. The whirling columns of light slipping into the Hive chamber from a fan 20 meters away are a little distracting. The rays should fall off based on distance and view angle. They are there to establish atmosphere, not remind everyone how cool your engine is.
Which brings me to the larger purpose of graphics. As I have said, this tech demo is very impressive. I am confident that you guys will produce a terrific looking game. But lest we loose ourselves in the spectacle of window shutters, it servers everyone to remember the core strengths of Natural Selection; the reasons I return to NS time and again: mood and gameplay. These technical achievements are perfectly poised to enhance both, and I am sure you've kept these goals foremost in your development process, but allow me to explore them from my point of view (feel free to have thought of all of this years ago).
The Word for NS2 has been "dynamic." Infestation, lighting, environment. All aspire to dynamism. Much has been made of the game's variability, and rightly so. NS2 promises to be one of the most "dynamic" multi-paradigm multiplayer games on the block. But if this promise is to be realized, these individually expressive elements must be successfully combined into a living, breathing, dynamic whole. It must be more than window shutters and light switches (it's a tech demo, I know, and a really cool one). It is not enough to have procedural infestation over here and spinning fans over there. The endgame is a synthesis of all of these dynamic systems into an environment that feels real. That has character. That evokes emotion.
The two hallmark "dynamic lighting" schemes to date are time-of-day, in which the motions of heavenly objects are simulated "in real time!", and the ever-popular "dynamic lights!" which are like regular lights, only you can turn them off / shoot them out. The former offers no enhancement to gameplay (other than visualizing how much time one has spent in isolation from the real sun), and the latter involves deliberate (and tedious) intervention by the player to either find the light switch or take careful aim. Both are little more than curiosities. We are still waiting for the "dynamic lighting" system that means something.
And here I think you are blessed with your setting. Future industrial technology is replete with glowies, blinkies, auras, emissions, and indirect light of all sorts. There's "regular" mode, "power fluctuation" mode, "power failure" mode, "emergency lighting" mode, "proximity alert" mode, "red alert" mode, "lethal radiation alert" mode, "violent decompression alert" mode, and "power restored" mode, just to name a few. The procession of infestation through the map is a ripe excuse to fiddle with the interior lighting design. To excerpt a post from <a href="http://www.unknownworlds.com/ns2/forums/index.php?showtopic=103379&view=findpost&p=1663285" target="_blank">another thread</a>,
This touches on one of the three roles of lighting: sensory encumberance. The other two are information conveyance and emotional queue. Sensory encumbering light shifts give one team a tactical advantage by impairing the other team's vision. By forcing marines to use flashlights or night vision (or just suffer dim or sporadic illumination), the Kharaa stand a better chance in toe-to-toe encounters. Information conveying light shifts alert players to some change in game state. The approach of enemies, the completion of hives/structures, the progress of time-dependent variables, the occurrence of major events, &c.. Lighting can also convey information as it changes through space as well as through time: areas of the map near infestation may have different lighting characteristics, communicating the direction of the alien territory. Also, lighting effects can be used to provide specific tactical information. If marines are using flashlights, the light beam will signal their approach and position to aliens in ambush. Hive sight might also include any number of additional lighting effects to convey information on friendly and hostile activities. Valve discusses the use of lighting to subtly convey information in the TF2 developer commentaries - from the direction of the sun, the red side of the map is cast in warm light, while the blue base sits in cool shadows, subconsciously orienting players. Lighting is a remarkable high-bandwidth channel and human beings are given to pick up on patterns whether you point them out or not. Finally, lighting can serve as an emotional queue. Putting players on edge, plunging them into panic, bathing them in relief. Lighting also colors the larger emotional pallet of an environment. The "mood" of a place is in its lighting.
These three are the important functions of light and they are intertwined (e.g., when you impair senses, you elicit emotion). Dynamic lighting is well and good by it self, but its full potential lies in the service of these contexts. Spinning fans and fluttering shutters have their place, but I encourage you to design your lighting around its function rather than its effect. Most of all, leverage these fully dynamic lights to bring your environments to life! Let the lights change with the game. Make them mean something.
And you've thought of all of that before. Great. Can't wait for the next vidcast.<!--QuoteEnd--></div><!--QuoteEEnd-->
Quoted For Truth (QFT).
This post is dead on for how I feel about light as well - like say, if there was infestation near those fans, would they stop? I know, its just a lil tech demo, just trying to emphasize my agreement with that well written post by lunchtimemama.
Well, thank you for sharing and may your work towards deadlines be blessed Unknown Worlds <img src="style_emoticons/<#EMO_DIR#>/smile-fix.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile-fix.gif" />
<!--coloro:lime--><span style="color:lime"><!--/coloro-->As to performance<!--colorc--></span><!--/colorc-->, what FPS drop did you guys see (total) going from static to dynamic lighting, and what kind of system was that example running on?
Man, what an amazing post. As CanadianWolverine said, QFT!
I would just like to reiterate that I am very impressed with the work so far on engine. Please try to see that when we give these constructive comments, it is because to us new doors of opportunity have opened. We are not trying to criticise your current implementation of dynamic lighting, we are just getting excited over what this might evolve into.
Natural Selection 2, Diablo 3, Left 4 Dead /drool
Also, curious about the Hive, it's glowing by itself but is it really emitting any light? It shows such a sharp line of shadow instead of softly illuminating the walls. Some light entity that is located in a weird spot or?
One thing I'd like to see is a limitation of the engine that you've found. Somewhere that just looks wrong. Showing all the good sides of something is good, leaving limitations to the viewer's imaginations is generally a bad thing. I'd rather know exactly where this engine fails rather than leaving it as a vague niggle in the back of my mind.
I don't know about other people, but I always try to find the breaking point of any game I play. Find a situation that the engine or the level design just can't handle. While the rest of the game might be stellar, knowing where it breaks is somehow comforting.
--Scythe--
Side topic: We see in the hive room a Sensory Chamber plonked down on the floor. Have you guys got any plans to have the dynamic growth we've already seen 'sprout' into the terrain from the point the Chamber starts? It'd be great to see it look more like it'd grown there, rather than being a distinct, unintegrated entity. Oh, and I'm going to make a random (and probably near impossible at a code level) request. It would be awesomely cool if Cloaking (should it still exist) cause some sort of interaction/disturbance with light effects. Aside from sheer 'sweet' factor, could also have some value in things like areas you can light, and flashlight use, etc etc.
Otherwise, great to see work is still progressing nicely, keep up the good work!
Finaly real moving pix :')
<!--quoteo(post=1690000:date=Oct 11 2008, 01:06 PM:name=Ride)--><div class='quotetop'>QUOTE(Ride @ Oct 11 2008, 01:06 PM) <a href="index.php?act=findpost&pid=1690000"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Very good job there, now that's something we can look at while we wait. Don't let us wait that long again. Even if you're only showing us what you had to eat today, I'd still watch it!<!--QuoteEnd--></div><!--QuoteEEnd-->
I agree <img src="style_emoticons/<#EMO_DIR#>/tounge.gif" style="vertical-align:middle" emoid=":p" border="0" alt="tounge.gif" />
I agree with the earlier post that radiosity is the crown jewel of Source's lighting, both as it applies to light maps and model lighting. Their "radiosity normal-mapping" technique is really quite clever and it give the impression that models "fit in" to their surroundings. It is a subtle but compelling effect. As mentioned, real-time radiosity calculations are not feasible and dynamically generated "bounce lights" have problems of their own (some of which Flayra mentioned). I'm not enough of a graphics guy to offer another solution, but I urge you to be on the lookout for a good way to approximate radiosity. Another thing you might want to cheat on is ambient occlusion; it really helps models "fit in."
The God Rays are another effect whose key is subtlety. The whirling columns of light slipping into the Hive chamber from a fan 20 meters away are a little distracting. The rays should fall off based on distance and view angle. They are there to establish atmosphere, not remind everyone how cool your engine is.
Which brings me to the larger purpose of graphics. As I have said, this tech demo is very impressive. I am confident that you guys will produce a terrific looking game. But lest we loose ourselves in the spectacle of window shutters, it servers everyone to remember the core strengths of Natural Selection; the reasons I return to NS time and again: mood and gameplay. These technical achievements are perfectly poised to enhance both, and I am sure you've kept these goals foremost in your development process, but allow me to explore them from my point of view (feel free to have thought of all of this years ago).
The Word for NS2 has been "dynamic." Infestation, lighting, environment. All aspire to dynamism. Much has been made of the game's variability, and rightly so. NS2 promises to be one of the most "dynamic" multi-paradigm multiplayer games on the block. But if this promise is to be realized, these individually expressive elements must be successfully combined into a living, breathing, dynamic whole. It must be more than window shutters and light switches (it's a tech demo, I know, and a really cool one). It is not enough to have procedural infestation over here and spinning fans over there. The endgame is a synthesis of all of these dynamic systems into an environment that feels real. That has character. That evokes emotion.
The two hallmark "dynamic lighting" schemes to date are time-of-day, in which the motions of heavenly objects are simulated "in real time!", and the ever-popular "dynamic lights!" which are like regular lights, only you can turn them off / shoot them out. The former offers no enhancement to gameplay (other than visualizing how much time one has spent in isolation from the real sun), and the latter involves deliberate (and tedious) intervention by the player to either find the light switch or take careful aim. Both are little more than curiosities. We are still waiting for the "dynamic lighting" system that means something.
And here I think you are blessed with your setting. Future industrial technology is replete with glowies, blinkies, auras, emissions, and indirect light of all sorts. There's "regular" mode, "power fluctuation" mode, "power failure" mode, "emergency lighting" mode, "proximity alert" mode, "red alert" mode, "lethal radiation alert" mode, "violent decompression alert" mode, and "power restored" mode, just to name a few. The procession of infestation through the map is a ripe excuse to fiddle with the interior lighting design. To excerpt a post from <a href="http://www.unknownworlds.com/ns2/forums/index.php?showtopic=103379&view=findpost&p=1663285" target="_blank">another thread</a>,
This touches on one of the three roles of lighting: sensory encumberance. The other two are information conveyance and emotional queue. Sensory encumbering light shifts give one team a tactical advantage by impairing the other team's vision. By forcing marines to use flashlights or night vision (or just suffer dim or sporadic illumination), the Kharaa stand a better chance in toe-to-toe encounters. Information conveying light shifts alert players to some change in game state. The approach of enemies, the completion of hives/structures, the progress of time-dependent variables, the occurrence of major events, &c.. Lighting can also convey information as it changes through space as well as through time: areas of the map near infestation may have different lighting characteristics, communicating the direction of the alien territory. Also, lighting effects can be used to provide specific tactical information. If marines are using flashlights, the light beam will signal their approach and position to aliens in ambush. Hive sight might also include any number of additional lighting effects to convey information on friendly and hostile activities. Valve discusses the use of lighting to subtly convey information in the TF2 developer commentaries - from the direction of the sun, the red side of the map is cast in warm light, while the blue base sits in cool shadows, subconsciously orienting players. Lighting is a remarkable high-bandwidth channel and human beings are given to pick up on patterns whether you point them out or not. Finally, lighting can serve as an emotional queue. Putting players on edge, plunging them into panic, bathing them in relief. Lighting also colors the larger emotional pallet of an environment. The "mood" of a place is in its lighting.
These three are the important functions of light and they are intertwined (e.g., when you impair senses, you elicit emotion). Dynamic lighting is well and good by it self, but its full potential lies in the service of these contexts. Spinning fans and fluttering shutters have their place, but I encourage you to design your lighting around its function rather than its effect. Most of all, leverage these fully dynamic lights to bring your environments to life! Let the lights change with the game. Make them mean something.
And you've thought of all of that before. Great. Can't wait for the next vidcast.<!--QuoteEnd--></div><!--QuoteEEnd-->
I agree with this post in every motion, and it hasn't been quoted on this page yet, so here. It MUST be read.
My 2 cents, the lighting beams need to fall off more at long ranges, although, it's really something that should be controlled by the mapper (some lights should look like spotlights, many lights should not though). Bumpmapping and other surface treatments were mentioned... but I really didn't see any good examples of that in the demo, that being said, it can be deceptive looking at a game with both new tech and old tech art assets, because they don't always gel perfectly. My only other concern would be FPS and system compatibility... but meh, I guess with that we'll just have to see. Oh, and of course I feel gameplay can't possibly be taken seriously enough, but we haven't really had any news regarding gameplay designs, so I have nothing to comment on there.
From the Digital Extremes site:
Dark Sectorâ„¢
Utilizing the Evolution Engineâ„¢, proprietary next-gen technology developed by Digital Extremes, Dark Sector is designed exclusively for next-generation gaming.
The NS1 artwork was some of the best modeling and texture work on the HL engine. Really, you don't even need to do alot to fully update them to current gen tech, just generate spec/lighting maps and bump maps for them, and they'd look as good as any HL2 model.
[edit] That being said, those models have not been fully updated yet, they look so good simply because they are well modeled and textured.