Thank you so much for the update guys. And and extra thanks to you Max for giving the technical follow-up, and the community here is much more technical than average.
I do have few concerns, however. It seems as if your development hardware is a Vista OS based, Intel CPU run, software running on ATi hardware for rendering. Many of us, including myself prefer to use an XP (SP3 currently) based, AMD CPU run, software running on nVidia hardware for rendering. In the past, I've seen many, <b><i>many</i></b> developers develop software on just one piece of hardware and then they are plagued by incompatibility issues.
Examples?<ul><li>Halo 2, as Vista-only</li><li>Steam, when first released (non-beta), wouldn't work correctly on any OS other than WindowsXP (Win2000 was buggy and WinMe just flat out hung the computer if you moved the mouse after opening a GUI menu)</li><li>Ghost Recon, not working properly on nVidia hardware; then GRAW working inefficiently on ATi hardware</li><li>The Aurora Engine (NWN2), having been developed solely on Intel hardware, having inefficiencies in parsing graphics data to the graphics adapter in AMD systems</li></ul> I could keep going and I'm sure others here can come up with much better examples of applications (not even necessarily game programs) which have cross-platform hardware compatibility issues. I will simply ask these two important questions:
1. What re-assurances can you give your community and investors of the priority of application compatibility? 2. How would compatibility testing manifest in developement? (integrated into the creation of the source, alpha test fixes, beta test fixes, or post-production patches; I assume a combination of, but which? why?)
Anybody should feel free to comment on any of this, but only the developer's declarations and thoughts truly matter in this case. As always, thank you for your time in addressing this concern.
<!--quoteo(post=1690548:date=Oct 16 2008, 10:03 AM:name=modeling_man)--><div class='quotetop'>QUOTE(modeling_man @ Oct 16 2008, 10:03 AM) <a href="index.php?act=findpost&pid=1690548"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Man i want to see more of that marine! I bet whomever modeled it is totally awesome. Seriously, pimping stuff Max you make me so jealous. I can't wait to get my hands on the engine stuff! TQ2000 will live again! Tell Cory congrats for me, and tell everyone else hi. <img src="style_emoticons/<#EMO_DIR#>/smile-fix.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile-fix.gif" /><img src="style_emoticons/<#EMO_DIR#>/smile-fix.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile-fix.gif" /><img src="style_emoticons/<#EMO_DIR#>/smile-fix.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile-fix.gif" /> B<!--QuoteEnd--></div><!--QuoteEEnd-->
In case it's not completely obvious, <a href="http://subshape.phpwebhosting.com/?page_id=3" target="_blank">modeling_man</a> created the female marine...
Thanks for the update Max, it's always great to get some more info on how things are coming along.
I'm sure no matter what you do with the lighting it will look good. I don't think you guys would settle for second-best (well sometimes you have to compromise, but I know it will be far better than most people will expect).
I'm really impressed by how it's shaping up, though it would be nice to see some renders of the marine, and weapons if you have any completed yet, just to check out <img src="style_emoticons/<#EMO_DIR#>/tounge.gif" style="vertical-align:middle" emoid=":p" border="0" alt="tounge.gif" />
The editor looks really nice too, I like the look of the interface already.
I asked this earlier but I'll ask again, what 3d packages will be compatible with your engine for asset imports? I'd really like to know, as I'm really getting into Maya right now, and already know 3ds fairly well, and before NS2 is out I'd like to learn one of the packages that will be compatible, I'm assuming one of those two will be compatible.
Either way, completely impressed, can't wait to see more footage!
<!--quoteo(post=1690543:date=Oct 16 2008, 11:27 AM:name=Max)--><div class='quotetop'>QUOTE(Max @ Oct 16 2008, 11:27 AM) <a href="index.php?act=findpost&pid=1690543"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->If we had Star Trek-like environments with lots of white walls and bright lights then the bounce light is important, but in our current environments it just doesn't seem to be much of a factor.<!--QuoteEnd--></div><!--QuoteEEnd--> In the current NS there is a fairly large number of lighter concrete and tile textured environments, not to mention, while NS often uses dark colored textures, more often then not those textures are raw metals, meaning they would have much higher reflectivity than a wood, plastic, or painted wall the same shade. Ya, the game is often dark, but even then, that's not really an NS standard, and certainly not something to impose on mappers. The transition between light and dark environments is part of what makes NS maps so powerful. Bright lighting, and flood lighting, might not be an NS style thing, but bounce lighting is increadibly nice used on strong spot lights in abstract environments as well. Things like sun rays casting down from a porthole and reflecting in an oddly shaped room, or flood lights shining down a chasm landing strip, relfecting at odd angles off the walls. Some of that can be cheated on and done manually by the mapper, but making the environment too abstract, or too dynamic, makes that impossible.
moultanoCreator of ns_shiva.Join Date: 2002-12-14Member: 10806Members, NS1 Playtester, Contributor, Constellation, NS2 Playtester, Squad Five Blue, Reinforced - Shadow, WC 2013 - Gold, NS2 Community Developer, Pistachionauts
<!--quoteo(post=1690592:date=Oct 16 2008, 06:40 PM:name=Swiftspear)--><div class='quotetop'>QUOTE(Swiftspear @ Oct 16 2008, 06:40 PM) <a href="index.php?act=findpost&pid=1690592"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->In the current NS there is a fairly large number of lighter concrete and tile textured environments, not to mention, while NS often uses dark colored textures, more often then not those textures are raw metals, meaning they would have much higher reflectivity than a wood, plastic, or painted wall the same shade. Ya, the game is often dark, but even then, that's not really an NS standard, and certainly not something to impose on mappers. The transition between light and dark environments is part of what makes NS maps so powerful. Bright lighting, and flood lighting, might not be an NS style thing, but bounce lighting is increadibly nice used on strong spot lights in abstract environments as well. Things like sun rays casting down from a porthole and reflecting in an oddly shaped room, or flood lights shining down a chasm landing strip, relfecting at odd angles off the walls. Some of that can be cheated on and done manually by the mapper, but making the environment too abstract, or too dynamic, makes that impossible.<!--QuoteEnd--></div><!--QuoteEEnd--> Yeah, I was just thinking about how ns_hera made enormous use of strong lighting with clean surfaces, and looked amazing as a result.
Could you post an image with a big white light source that gives the room a minimum light and an additional colored light source with and without the bouncing effect. It's just a guess, but I think the bouncing effect is more visible with colored light.
The competitive staples include (among others) veil, tanith, eclipse, metal, origin, orbital (custom), lucid (custom).
Except for maybe origin, there's precious little issues with darkness in these maps.
We play(ed) these maps on max gamma and brightness as well to reduce the frustration of trying to aim in any of the few dark areas there might be.
Atmosphere is good but after the initial shock value of graphics as amazing as those found in the last podcast, extreme lighting contrast loses its appeal in many cases. I'm sure your mappers are intelligent enough to know that trying to aim in the dark is just frustrating, but from the way you're talking, it just seems like it could be said again, as constructive criticism.
In other words - awesome work, but please don't go crazy with your new toys. <img src="style_emoticons/<#EMO_DIR#>/lerk.gif" style="vertical-align:middle" emoid="::lerk::" border="0" alt="lerk.gif" />
hookeyedJoin Date: 2007-08-26Member: 62011Members, Squad Five Blue
Whatever sort of lightning you opt to go with, I'm confident mappers could tweak it to work. I'm still waiting on previews of the NS2TR, as the textures decide the flavour of the atmosphere in the end. I hope you release something to work with in the near future, I've been holstering map ideas for ages. Also, good choice on the modeller, his better work for Titan Quest looks perfect for those gritty kharaa forms.
Well, like people have been saying, I'm sure you'll pick whatever works best. In the end it's more about map composition and game balance than lighting, since players can alter their gamma/brightness and get around any intended lighting, it seems.
Looking forward to the next footage! <img src="style_emoticons/<#EMO_DIR#>/tounge.gif" style="vertical-align:middle" emoid=":p" border="0" alt="tounge.gif" />
Some interesting footage, but I must agree with the norm, if its possable can we get a video with DI being spawned on the walls and windows and hope it creeps over and begin takeing out the light and or if on the bight sun light glass pulse and eminate with a errie glow is Di capable of clogging up fans yet or completely blocking them and turning into a new prefab a kharaa style wind bag sphincta. - hum was going to post this in the other thread
on a note on the topics pressed in this thread. MAx have you consitered possable re-writeing the light code? or a plugin to re-write the light code. were needed a new hack
i am thinking have the first spot light in addition to projecting its light shoot out a invisable dynamic ball when the ball hits a surface it instantly sends a message back to the spot light, where it hits the surface it creates a new spot light with the new deflection faceing from were the ball hit and deflected the new light then fires another invisable dynamic box off in the direction the of the deflection of the last spot light and repeat. Information the invisible ball sends back off to the previous light are ambient drop off values as in how far the box traveled equals the amount the light dims and or radius changes as well as how soft the edges become. The new spot light also takes these values and caculates its new variables. rince and repeat for 2-3 bounces.
the cheats method i guess is have this fire once, or it can just bounce around and drop a point light at each location with deminishing values, to me this seams easy on the process, and some relatively easy code.
other things that could be done with the point light idea is keep the bouncey invis ball bouncying between the spot light and the point lights it created and if the ball path is broken tell it to turn off all other point lights from then on.
in turn you have spoti hiting surface dynamicaly minus gravity, deflecting hiting new surface making faded point light, deflecting makeing faded point light, as onos comes down coridor blocks last deflection path last point light turns off, if unblocked last point light turns back on, if the path between the spotie and or the first deflection gets broken all other point lights down the stream turn off. till the invis ball can bounce around the deflection path again, this seams a little more process exstensive, but would replicate efficent illuminosity, but may have a weird effect but not cast real time shadows, you could stop the point light casting shadows from its origin. other than having it invis ball broken.
the only issue i see with the spot light idea originaly is if one spot light is broken how to effect the other dimmed spot lights that follow, you could if beam is broken then spawn a point lights for all bounces after wards and dim them a heap, it just i think wiht out some other if else weird things could happen if the bounce invis ball hits a shine full speed onos carapace and spawns a couple hundred new spot lights in the span of a second disco onos anyone?, basicly it would fail if a onos broke one spot light beam and the rest werent dimmed or turned off. you could cache the bouncy spot lights before map begin and if a light path is broken, by said onos or movable object, spawn a temparay point light on said surface or hit box that broke beam, i think it would be as exstensive as a bunch of invisable partical effect fireing off consistantly, there are a lot of scripts there that would run constantly cheaking for a break in the beam.
I guess the only option with this is how real do you want to go, is it possable to write, how long will it take, and is it worth it for its over all effect it will have on the game.
honoustly guys this is all nice well and good but please just keep the game as arcady as possable its your bigest thing and every other company seams to be pulling away from it and loseing there fan bases, bar say team fortress garys mod, the odd zombie shooter a few sports or skater games and hte odd car game a couple of top down rts, and a few mmo's, are keeping arcadyness over realisum. that and the classic single player action adventure games, along with niche genre like guitar hero.
alot of *realisum games or serious games seam to be losing there fan base.. look at UT3, there less aussie servers spread across like 6 game modes than there is NS servers. and most of those are empty or useing a few bots. and UT is prolly the most arcady or run and gun gib game of all the serious/ realisum games out there, ghost recon died, swat died, cod4 all though cool i fear will die rapidly, some of these games do have the bonuses of being playable on all plate forums including consoles, but it will only exstend there life until the next gen console comes out.
basicly cod4, i fear wont hold its player base, and most will go back to cod2
its a fine art juggaling between realisum that effect game play, immersion and fun if anyone can do it its you guys.
oh and to all those pissing about gamma hacks i believe its been stated before that they plan to establish a unique balance between exstream contrasting of light and dark areas to the point were its almost like a strobe light so that there nothing a gamma hacker can do short of binding a unique network of gamam adjustment binds which only adds more keys to his play style and will surely effect and take away form his play, in turn not being worth it.. although i believe this information was stated in a much more efficent way. like 7 pod casts back or in a replie from max in one of its follow up threads.
I'd like to see performance when stuff is happening on screen.
I mean, looking at a room in Crysis gives me 150fps but drop in a player weapon, player models and gunfire then it drops to 40.
I still think developing their own engine is a huge risk with a lot of question marks (one example: if tried and tested developers make an engine that gets hacked with aimbots etc would a new engine from an untried team* make the risk of hacking greater or lower?)
The video UWE made looks good- the dynamic lighting is really impressive, but as a competitive player** all I care about is that the engine 'feels right' and the netcode is great.
* untried relative to making an FPS engine. ** i know public players won't have an issue with this, but I'm sure Flayra has said he wants NS2 to be the best competitive experience since Starcraft.
Max, I didn't recognize you in the video because you weren't wearing your hat like in your avatar <img src="style_emoticons/<#EMO_DIR#>/smile-fix.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile-fix.gif" />
One thing I noticed and I'm surprised that no one has commented on it yet. Dynamic player shadows.
I watched very carefully waiting for a point where the 'player' would walk through a light source and cast a shadow and watching through the video leads me to believe that you <i>VERY</i> carefully dodged that issue. It would seem either you haven't coded this aspect yet (light hitting a player and casting a shadow) or that you were dodging the subject for whatever other reason (doesn't look good yet, don't want to implement it, etc).
Is this feature implemented? Specifically, your beautiful sunlight streaming in through a window and hitting the player to create a shadow on the ground in the shape of a player. I did notice the shadows created by the sensory chamber and the fan blades and I'm hoping player shadows will complete the picture.
Are you planning on implementing it? If you haven't done so yet I can certainly understand. It seems like this video was released to finally give the community a feel for the engine (and I must say <u>WOW!</u>) and the lighting so far implemented rather than a full technical review. As you said yourself, you are still debating on the amount of time available for Bounce Lighting. I'm hoping this is the case as real shadowing is going to be a great game play aspect and feel issue; something that has been a long time coming since that trailer with the skulks lurking and leaping so long ago.
About Bounce Lighting: only you can decide if the 'feel' it generates sufficiently merits the time involved to implement. Some great notes have been said about the need for aliens to hide in darkness so I ask you to tread carefully as I think it's an important issue as well. That having been said, your demo of the tech looks extremely nice and I hope that this makes it into the game and engine. In some respects NS2 is going to be a Serious Sam style debut with the game detailing the technical aspects of the engine.
About the video in general: WOW! What you've built is just absolutely gorgeous! I love the way the sunlight filters in on the command chair in the first area. The spinning fan blades show the dynamic aspect of the light works marvelously. The hive glow across the sensory chamber is awesome effect. I've been looking at Dead Space trailers lately and I have to confess that I've been extremely unimpressed with their graphics and lighting while your demo just absolutely blows me away! I really do hope that NS2 becomes a force to be reckoned with because other game developers need to take a page from your book. You've obviously developed the workings of an engine that has a gorgeous look and feel and at the same time your system specs sheet will make the game accessible to the average gamer rather than die hard system builders.
Final notes: It will be amazing to see the brilliant game dynamics of Natural Selection take the world by storm. I've loved the RTS/FPS feel of Natural Selection for so long (since the day I started playing 1.4 in 2003) and I've been hoping it would make the jump to Source. Instead, you've exceeded my expectations with a whole new engine! If the final product looks half as good as your demo, I'll be in heaven when it releases. My fellow TSA Marines: say YES to group tactics, say YES to squads, say YES SIR, MAY I HAVE ANOTHER to your commander!
their own engine is a huge risk with a lot of question marks (one example: if tried and tested developers make an engine that gets hacked with aimbots etc would a new engine from an untried team* make the risk of hacking greater or lower. <img src="style_emoticons/<#EMO_DIR#>/lerk.gif" style="vertical-align:middle" emoid="::lerk::" border="0" alt="lerk.gif" />
locallyunsceneFeeder of TrollsJoin Date: 2002-12-25Member: 11528Members, Constellation
edited October 2008
<!--quoteo(post=1690794:date=Oct 19 2008, 09:15 AM:name=sherpa)--><div class='quotetop'>QUOTE(sherpa @ Oct 19 2008, 09:15 AM) <a href="index.php?act=findpost&pid=1690794"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I still think developing their own engine is a huge risk with a lot of question marks (one example: if tried and tested developers make an engine that gets hacked with aimbots etc would a new engine from an untried team* make the risk of hacking greater or lower?)
* untried relative to making an FPS engine.<!--QuoteEnd--></div><!--QuoteEEnd--> The company may be new, but Max and Charlie have been doing this for a while so they know how to avoid obvious exploits. And Flayra may not have written gldsrc, but he sure changed it around enough that I'd say he has experience with FPS engines. I'm not too worried about them cutting corners and leaving exploitable bugs in either.
I think the high contrast between dark and light areas will suit the NS world, shouldnt be to hard for mappers to fake some bounces were needed anyway. Roll on ns2 <img src="style_emoticons/<#EMO_DIR#>/smile-fix.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile-fix.gif" />
<!--quoteo(post=1690794:date=Oct 19 2008, 02:15 PM:name=sherpa)--><div class='quotetop'>QUOTE(sherpa @ Oct 19 2008, 02:15 PM) <a href="index.php?act=findpost&pid=1690794"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I still think developing their own engine is a huge risk with a lot of question marks (one example: if tried and tested developers make an engine that gets hacked with aimbots etc would a new engine from an untried team* make the risk of hacking greater or lower?)<!--QuoteEnd--></div><!--QuoteEEnd-->Hacking is all about fame and glory, so this would depend on how difficult it is to crack and how popular the engine/game is. If NS2 is still niche then it won't be as much of a target for hackers, but it is something to consider. If UWE goes with Steam they can still count on Steam-wide bans thanks to VAC. Otherwise some other form of anti-cheat protection would be needed.
I think generally punishing users who use hacks is a better form of anti-cheat than trying to make your product cheat-proof (a virtual impossibility).
<!--quoteo(post=1690856:date=Oct 20 2008, 08:15 AM:name=Thuro)--><div class='quotetop'>QUOTE(Thuro @ Oct 20 2008, 08:15 AM) <a href="index.php?act=findpost&pid=1690856"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->One thing I noticed and I'm surprised that no one has commented on it yet. Dynamic player shadows.
I watched very carefully waiting for a point where the 'player' would walk through a light source and cast a shadow and watching through the video leads me to believe that you <i>VERY</i> carefully dodged that issue. It would seem either you haven't coded this aspect yet (light hitting a player and casting a shadow) or that you were dodging the subject for whatever other reason (doesn't look good yet, don't want to implement it, etc).
Is this feature implemented? Specifically, your beautiful sunlight streaming in through a window and hitting the player to create a shadow on the ground in the shape of a player. I did notice the shadows created by the sensory chamber and the fan blades and I'm hoping player shadows will complete the picture.<!--QuoteEnd--></div><!--QuoteEEnd--> The players can cast shadows. Right now in first person mode you don't see a shadow from yourself because you don't have a body like in NS (and a lot of other first person shooters), although you do see shadows from other players. This is potentially a problem since you might be casting a shadow that could be visible by other players and not to yourself, so we've been considering giving the player a body like in Crysis or Dark Messiah of Might and Magic. But that's a bit of a departure from NS so we're not sure about it yet.
<!--quoteo(post=1690874:date=Oct 20 2008, 10:17 AM:name=Max)--><div class='quotetop'>QUOTE(Max @ Oct 20 2008, 10:17 AM) <a href="index.php?act=findpost&pid=1690874"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->The players can cast shadows. Right now in first person mode you don't see a shadow from yourself because you don't have a body like in NS (and a lot of other first person shooters), although you do see shadows from other players. This is potentially a problem since you might be casting a shadow that could be visible by other players and not to yourself, so we've been considering giving the player a body like in Crysis or Dark Messiah of Might and Magic. But that's a bit of a departure from NS so we're not sure about it yet.<!--QuoteEnd--></div><!--QuoteEEnd--> I personaly think it would be nice to see your own body and shadows though I'm not sure what the downsides would be.
<!--quoteo(post=1690886:date=Oct 20 2008, 07:09 PM:name=KungFuDiscoMonkey)--><div class='quotetop'>QUOTE(KungFuDiscoMonkey @ Oct 20 2008, 07:09 PM) <a href="index.php?act=findpost&pid=1690886"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I personaly think it would be nice to see your own body and shadows though I'm not sure what the downsides would be.<!--QuoteEnd--></div><!--QuoteEEnd-->
The obvious downsides being development time (to code and develop a fairly minor feature) and processing time (the HD, RAM, CPU, and GPU requirements would increase).
That aside, I think player shadows are important enough to me (after all I was the one who brought it up to begin with) that I think the time to make them should be taken, on the other hand I'd also like to see bounce lighting, real-time physics, destructible environment, and dynamic infestation take major roles in the game so maybe I'm a bit greedy! I love the idea that was mentioned about Dynamic infestation taking over lights and altering gameplay the way a welder does for marine (impassible areas when infested, closing doors, shutting down lights so only aliens can see, etc).
When using shadowmaps, making the player cast a shadow is basically "free" complexity wise. I also think it should be there, it looks really odd if you can walk around as if you were some kind of ghost.
<!--quoteo(post=1690888:date=Oct 20 2008, 12:30 PM:name=Thuro)--><div class='quotetop'>QUOTE(Thuro @ Oct 20 2008, 12:30 PM) <a href="index.php?act=findpost&pid=1690888"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->The obvious downsides being development time (to code and develop a fairly minor feature) and processing time (the HD, RAM, CPU, and GPU requirements would increase).<!--QuoteEnd--></div><!--QuoteEEnd--> Well you're already rendering every other part of the world so you could probably just render the same 3rd person model that is being rendered for all the other players. It might not be this simple. *shrugs*
You can skip the body all you want, but at least cast the shadow. If a skulk or such is trying to hide, it will be mandatory to tell where their shadow is and if their shadow will give them away, like you said. It may look a little odd, but is a good compromise.
Immersion's always good in my books, being able to look down and see yourself and shadows would help plant the player in the world, as if youre really there. Having one extra model being rendered wont kill your computer, so i dont see much of a downside
Comments
I do have few concerns, however. It seems as if your development hardware is a Vista OS based, Intel CPU run, software running on ATi hardware for rendering. Many of us, including myself prefer to use an XP (SP3 currently) based, AMD CPU run, software running on nVidia hardware for rendering. In the past, I've seen many, <b><i>many</i></b> developers develop software on just one piece of hardware and then they are plagued by incompatibility issues.
Examples?<ul><li>Halo 2, as Vista-only</li><li>Steam, when first released (non-beta), wouldn't work correctly on any OS other than WindowsXP (Win2000 was buggy and WinMe just flat out hung the computer if you moved the mouse after opening a GUI menu)</li><li>Ghost Recon, not working properly on nVidia hardware; then GRAW working inefficiently on ATi hardware</li><li>The Aurora Engine (NWN2), having been developed solely on Intel hardware, having inefficiencies in parsing graphics data to the graphics adapter in AMD systems</li></ul>
I could keep going and I'm sure others here can come up with much better examples of applications (not even necessarily game programs) which have cross-platform hardware compatibility issues. I will simply ask these two important questions:
1. What re-assurances can you give your community and investors of the priority of application compatibility?
2. How would compatibility testing manifest in developement? (integrated into the creation of the source, alpha test fixes, beta test fixes, or post-production patches; I assume a combination of, but which? why?)
Anybody should feel free to comment on any of this, but only the developer's declarations and thoughts truly matter in this case. As always, thank you for your time in addressing this concern.
<b><!--coloro:#DD0000--><span style="color:#DD0000"><!--/coloro-->x5<!--colorc--></span><!--/colorc--></b>
Seriously, pimping stuff Max you make me so jealous.
I can't wait to get my hands on the engine stuff!
TQ2000 will live again!
Tell Cory congrats for me, and tell everyone else hi.
<img src="style_emoticons/<#EMO_DIR#>/smile-fix.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile-fix.gif" /><img src="style_emoticons/<#EMO_DIR#>/smile-fix.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile-fix.gif" /><img src="style_emoticons/<#EMO_DIR#>/smile-fix.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile-fix.gif" />
B<!--QuoteEnd--></div><!--QuoteEEnd-->
In case it's not completely obvious, <a href="http://subshape.phpwebhosting.com/?page_id=3" target="_blank">modeling_man</a> created the female marine...
I'm sure no matter what you do with the lighting it will look good. I don't think you guys would settle for second-best (well sometimes you have to compromise, but I know it will be far better than most people will expect).
I'm really impressed by how it's shaping up, though it would be nice to see some renders of the marine, and weapons if you have any completed yet, just to check out <img src="style_emoticons/<#EMO_DIR#>/tounge.gif" style="vertical-align:middle" emoid=":p" border="0" alt="tounge.gif" />
The editor looks really nice too, I like the look of the interface already.
I asked this earlier but I'll ask again, what 3d packages will be compatible with your engine for asset imports? I'd really like to know, as I'm really getting into Maya right now, and already know 3ds fairly well, and before NS2 is out I'd like to learn one of the packages that will be compatible, I'm assuming one of those two will be compatible.
Either way, completely impressed, can't wait to see more footage!
In the current NS there is a fairly large number of lighter concrete and tile textured environments, not to mention, while NS often uses dark colored textures, more often then not those textures are raw metals, meaning they would have much higher reflectivity than a wood, plastic, or painted wall the same shade. Ya, the game is often dark, but even then, that's not really an NS standard, and certainly not something to impose on mappers. The transition between light and dark environments is part of what makes NS maps so powerful. Bright lighting, and flood lighting, might not be an NS style thing, but bounce lighting is increadibly nice used on strong spot lights in abstract environments as well. Things like sun rays casting down from a porthole and reflecting in an oddly shaped room, or flood lights shining down a chasm landing strip, relfecting at odd angles off the walls. Some of that can be cheated on and done manually by the mapper, but making the environment too abstract, or too dynamic, makes that impossible.
Yeah, I was just thinking about how ns_hera made enormous use of strong lighting with clean surfaces, and looked amazing as a result.
Except for maybe origin, there's precious little issues with darkness in these maps.
We play(ed) these maps on max gamma and brightness as well to reduce the frustration of trying to aim in any of the few dark areas there might be.
Atmosphere is good but after the initial shock value of graphics as amazing as those found in the last podcast, extreme lighting contrast loses its appeal in many cases. I'm sure your mappers are intelligent enough to know that trying to aim in the dark is just frustrating, but from the way you're talking, it just seems like it could be said again, as constructive criticism.
In other words - awesome work, but please don't go crazy with your new toys. <img src="style_emoticons/<#EMO_DIR#>/lerk.gif" style="vertical-align:middle" emoid="::lerk::" border="0" alt="lerk.gif" />
I hope you release something to work with in the near future, I've been holstering map ideas for ages.
Also, good choice on the modeller, his better work for Titan Quest looks perfect for those gritty kharaa forms.
^^ <3 haymo
Well, like people have been saying, I'm sure you'll pick whatever works best. In the end it's more about map composition and game balance than lighting, since players can alter their gamma/brightness and get around any intended lighting, it seems.
Looking forward to the next footage! <img src="style_emoticons/<#EMO_DIR#>/tounge.gif" style="vertical-align:middle" emoid=":p" border="0" alt="tounge.gif" />
on a note on the topics pressed in this thread. MAx have you consitered possable re-writeing the light code? or a plugin to re-write the light code. were needed a new hack
i am thinking have the first spot light in addition to projecting its light shoot out a invisable dynamic ball when the ball hits a surface it instantly sends a message back to the spot light, where it hits the surface it creates a new spot light with the new deflection faceing from were the ball hit and deflected the new light then fires another invisable dynamic box off in the direction the of the deflection of the last spot light and repeat. Information the invisible ball sends back off to the previous light are ambient drop off values as in how far the box traveled equals the amount the light dims and or radius changes as well as how soft the edges become.
The new spot light also takes these values and caculates its new variables.
rince and repeat for 2-3 bounces.
the cheats method i guess is have this fire once, or it can just bounce around and drop a point light at each location with deminishing values, to me this seams easy on the process, and some relatively easy code.
other things that could be done with the point light idea is keep the bouncey invis ball bouncying between the spot light and the point lights it created and if the ball path is broken tell it to turn off all other point lights from then on.
in turn you have spoti hiting surface dynamicaly minus gravity, deflecting hiting new surface making faded point light, deflecting makeing faded point light, as onos comes down coridor blocks last deflection path last point light turns off, if unblocked last point light turns back on, if the path between the spotie and or the first deflection gets broken all other point lights down the stream turn off. till the invis ball can bounce around the deflection path again, this seams a little more process exstensive, but would replicate efficent illuminosity, but may have a weird effect but not cast real time shadows, you could stop the point light casting shadows from its origin. other than having it invis ball broken.
the only issue i see with the spot light idea originaly is if one spot light is broken how to effect the other dimmed spot lights that follow, you could if beam is broken then spawn a point lights for all bounces after wards and dim them a heap, it just i think wiht out some other if else weird things could happen if the bounce invis ball hits a shine full speed onos carapace and spawns a couple hundred new spot lights in the span of a second disco onos anyone?, basicly it would fail if a onos broke one spot light beam and the rest werent dimmed or turned off.
you could cache the bouncy spot lights before map begin and if a light path is broken, by said onos or movable object, spawn a temparay point light on said surface or hit box that broke beam, i think it would be as exstensive as a bunch of invisable partical effect fireing off consistantly, there are a lot of scripts there that would run constantly cheaking for a break in the beam.
I guess the only option with this is how real do you want to go, is it possable to write, how long will it take, and is it worth it for its over all effect it will have on the game.
honoustly guys this is all nice well and good but please just keep the game as arcady as possable its your bigest thing and every other company seams to be pulling away from it and loseing there fan bases, bar say team fortress garys mod, the odd zombie shooter a few sports or skater games and hte odd car game a couple of top down rts, and a few mmo's, are keeping arcadyness over realisum. that and the classic single player action adventure games, along with niche genre like guitar hero.
alot of *realisum games or serious games seam to be losing there fan base.. look at UT3, there less aussie servers spread across like 6 game modes than there is NS servers. and most of those are empty or useing a few bots. and UT is prolly the most arcady or run and gun gib game of all the serious/ realisum games out there, ghost recon died, swat died, cod4 all though cool i fear will die rapidly, some of these games do have the bonuses of being playable on all plate forums including consoles, but it will only exstend there life until the next gen console comes out.
basicly cod4, i fear wont hold its player base, and most will go back to cod2
its a fine art juggaling between realisum that effect game play, immersion and fun
if anyone can do it its you guys.
I mean, looking at a room in Crysis gives me 150fps but drop in a player weapon, player models and gunfire then it drops to 40.
I still think developing their own engine is a huge risk with a lot of question marks (one example: if tried and tested developers make an engine that gets hacked with aimbots etc would a new engine from an untried team* make the risk of hacking greater or lower?)
The video UWE made looks good- the dynamic lighting is really impressive, but as a competitive player** all I care about is that the engine 'feels right' and the netcode is great.
* untried relative to making an FPS engine.
** i know public players won't have an issue with this, but I'm sure Flayra has said he wants NS2 to be the best competitive experience since Starcraft.
I watched very carefully waiting for a point where the 'player' would walk through a light source and cast a shadow and watching through the video leads me to believe that you <i>VERY</i> carefully dodged that issue. It would seem either you haven't coded this aspect yet (light hitting a player and casting a shadow) or that you were dodging the subject for whatever other reason (doesn't look good yet, don't want to implement it, etc).
Is this feature implemented? Specifically, your beautiful sunlight streaming in through a window and hitting the player to create a shadow on the ground in the shape of a player. I did notice the shadows created by the sensory chamber and the fan blades and I'm hoping player shadows will complete the picture.
Are you planning on implementing it? If you haven't done so yet I can certainly understand. It seems like this video was released to finally give the community a feel for the engine (and I must say <u>WOW!</u>) and the lighting so far implemented rather than a full technical review. As you said yourself, you are still debating on the amount of time available for Bounce Lighting. I'm hoping this is the case as real shadowing is going to be a great game play aspect and feel issue; something that has been a long time coming since that trailer with the skulks lurking and leaping so long ago.
About Bounce Lighting: only you can decide if the 'feel' it generates sufficiently merits the time involved to implement. Some great notes have been said about the need for aliens to hide in darkness so I ask you to tread carefully as I think it's an important issue as well. That having been said, your demo of the tech looks extremely nice and I hope that this makes it into the game and engine. In some respects NS2 is going to be a Serious Sam style debut with the game detailing the technical aspects of the engine.
About the video in general: WOW! What you've built is just absolutely gorgeous! I love the way the sunlight filters in on the command chair in the first area. The spinning fan blades show the dynamic aspect of the light works marvelously. The hive glow across the sensory chamber is awesome effect. I've been looking at Dead Space trailers lately and I have to confess that I've been extremely unimpressed with their graphics and lighting while your demo just absolutely blows me away! I really do hope that NS2 becomes a force to be reckoned with because other game developers need to take a page from your book. You've obviously developed the workings of an engine that has a gorgeous look and feel and at the same time your system specs sheet will make the game accessible to the average gamer rather than die hard system builders.
Final notes: It will be amazing to see the brilliant game dynamics of Natural Selection take the world by storm. I've loved the RTS/FPS feel of Natural Selection for so long (since the day I started playing 1.4 in 2003) and I've been hoping it would make the jump to Source. Instead, you've exceeded my expectations with a whole new engine! If the final product looks half as good as your demo, I'll be in heaven when it releases. My fellow TSA Marines: say YES to group tactics, say YES to squads, say YES SIR, MAY I HAVE ANOTHER to your commander!
* untried relative to making an FPS engine.<!--QuoteEnd--></div><!--QuoteEEnd-->
The company may be new, but Max and Charlie have been doing this for a while so they know how to avoid obvious exploits. And Flayra may not have written gldsrc, but he sure changed it around enough that I'd say he has experience with FPS engines. I'm not too worried about them cutting corners and leaving exploitable bugs in either.
Ninja edit: Thanks <b>Cxwf</b>.
I think generally punishing users who use hacks is a better form of anti-cheat than trying to make your product cheat-proof (a virtual impossibility).
I watched very carefully waiting for a point where the 'player' would walk through a light source and cast a shadow and watching through the video leads me to believe that you <i>VERY</i> carefully dodged that issue. It would seem either you haven't coded this aspect yet (light hitting a player and casting a shadow) or that you were dodging the subject for whatever other reason (doesn't look good yet, don't want to implement it, etc).
Is this feature implemented? Specifically, your beautiful sunlight streaming in through a window and hitting the player to create a shadow on the ground in the shape of a player. I did notice the shadows created by the sensory chamber and the fan blades and I'm hoping player shadows will complete the picture.<!--QuoteEnd--></div><!--QuoteEEnd-->
The players can cast shadows. Right now in first person mode you don't see a shadow from yourself because you don't have a body like in NS (and a lot of other first person shooters), although you do see shadows from other players. This is potentially a problem since you might be casting a shadow that could be visible by other players and not to yourself, so we've been considering giving the player a body like in Crysis or Dark Messiah of Might and Magic. But that's a bit of a departure from NS so we're not sure about it yet.
I personaly think it would be nice to see your own body and shadows though I'm not sure what the downsides would be.
The obvious downsides being development time (to code and develop a fairly minor feature) and processing time (the HD, RAM, CPU, and GPU requirements would increase).
That aside, I think player shadows are important enough to me (after all I was the one who brought it up to begin with) that I think the time to make them should be taken, on the other hand I'd also like to see bounce lighting, real-time physics, destructible environment, and dynamic infestation take major roles in the game so maybe I'm a bit greedy! I love the idea that was mentioned about Dynamic infestation taking over lights and altering gameplay the way a welder does for marine (impassible areas when infested, closing doors, shutting down lights so only aliens can see, etc).
<img src="style_emoticons/<#EMO_DIR#>/asrifle.gif" style="vertical-align:middle" emoid="::asrifle::" border="0" alt="asrifle.gif" /> <!--coloro:#696969--><span style="color:#696969"><!--/coloro-->(run skulkie run!)<!--colorc--></span><!--/colorc--> <img src="style_emoticons/<#EMO_DIR#>/skulk.gif" style="vertical-align:middle" emoid="::skulk::" border="0" alt="skulk.gif" />
I also think it should be there, it looks really odd if you can walk around as if you were some kind of ghost.
Well you're already rendering every other part of the world so you could probably just render the same 3rd person model that is being rendered for all the other players. It might not be this simple. *shrugs*