Flashlight fixed/improved in HL2?

[Pseudonym_]

Don't be rediculous, gameplay isn't an element of an engine, its something the game developers create through arrangment of situations within the engine.

There is nothing revolutionary abour Source, except maybe the dynamic facial animation.

 

[Pseudonym_]

yes I predicted about 5 games in total, after the more advanced engines are out and people are playing the games, game devs wont want to license inferior graphics engines. It's not good business.

 

[Letters]

There have been two confirmed lisences, and the engine isn't even finished yet. Lighting isn't everything - what Source is really focusing on is actual gameplay, which is quite revolutionary for an engine.

W0rd to that...

Pseudonym_, some of things you come up with about Half-Life 2 and Source are ridiculous.

 

[Pseudonym_]

We are talking graphics here. Everytime a discussion is started on graphics hl2 fans quickly turn it around on gameplay. "yea but gameplay..........blah blah". The reason being you know what I say is true.

Anyway, as I said gameplay has nothing to do with the engine.

 

[Letters]

Oh, we're talking about graphics!

What's the problem? Looks fine...

 

[Shuzer]

Oh, we're talking about graphics!

What's the problem? Looks fine...

The problem? Well, if one person prefers HL2's overall graphical style over Doom 3's, they're clearly clueless.
Atleast, that's what it seems like every Doom 3 fanboy I've ever talked to seems to think.

 

[Pseudonym_]

Source looks fine yes. Don't get me wrong it's a great engine. I love it. But, graphically it can't compete with technology like Doom³ and U3. Plain and simple.

 

[Shuzer]

Source looks fine yes. Don't get me wrong it's a great engine. I love it. But, graphically it can't compete with technology like Doom³ and U3. Plain and simple.

And Doom 3 can't compete with U3 and their respective engines.. why? U3 is years down the road, and will have way more to offer.

 

[Pseudonym_]

You people are turning this into a doom³ fanboy rant against hl2, it isn't. I think by all indications I will like hl2 more infact. I am simply talking graphical capabilities in the different engines.

Source won't last in a market where there are a number of graphically superior engines. It doesn't matter if some people prefer the overall look of Source, what matters is pure graphical capability. This is a market where people shell out hundreds and hundreds a year in order to get the best graphics rendered on their computer, this is a graphics hungry market, and a graphically inferior engine will not last very long.

 

[Pseudonym_]

Doom³ is exactly what U3 would look lik eif they decided to cut back so that the game could be playable on the target hardware Doom³ is shooting for. What you see is U3 developers skipping ahead where Doom³ is taking steps. Doom³ technology will evolve to spurpass that of U3 in 2 years time. I think you all will be pleasently surprised with how much the technology will have advanced for quake4.

 

[Apos]

OFCOURSE every pixel on screen being lit by dynamic lighting where every light casts a shadow and every object casts a shadow, including on itself and other objects is more advanced than years old lightmap technology coupled with some models casting a simple projected shadow map on the ground, but not on other objects or on itself. Don't kid yourself.

You're ignoring things like Radiosity, HDR glare, and other things that are key to "real" lighting which Doom3 does not have. No game within the next year or two will even begin to offer a true "full" lighting simulation.

Look at the new games coming out, Doom³, Thief3, and now the granddaddy on the horizon Unreal3,

Unreal3 far surpasses any of these games, because it brings together all the disparate elements. Of course, that's not as hard as it might be given that it's targeting hardware that can handle it and shader effects thats basically give it that functionaltiy.

coupled with games already out like DE:IW and farcry,

DE:IW had Doom3's lighting system, and it just wasn't all that overwhelmingly impressive to look at, frankly. Doom3's lighting looks a bit better than DeusEx2's, but it's basically the same effect.

Source and Far Cry do basically the same sorts of things in regards to lighting: they can do dynamic lights (like swinging lights moving shadows around) if they want to spend the resources on it.

And Source is no where near that level. Sorry, it just isn't.

Shrug. It does different things. None of them does everything except Unreal 3.0 I doubt any of the devs of any of these engines would insist that there is the one right priority for light rendering once you realize you can't do it all with acceptable framerates. As I see it, Doom3 puts all its eggs in one basket. Source is more versatile, giving us most of the important shadowing and shading elements that dynamic lights get you, but also not slacking off on other lighting elements either.

Sourc's rendering engine will have to be completely re-written to even try to compare. in a market where advanced lighting models are expected, no one will liscense Source.

It's not really that hard to do what Doom3 does: the major problems are the performance cost, not the basic technology idea, which is actually a quite simple, brute force method.

I doubt their lighting engine would have to be totally rewritten, considering that it already does basically the exact same process that Doom3 uses, it just doesn't do it for every lightsource all the time.

I predict maybe 5 games in total this year will decide to go with Source, but by that time is up Doom³ will be out and people will expect games to look as good or better from that point on. Developers will not liscense Source after that, it will be too dated within a year of hl2 being released.

Shrug. We'll see. The engine can and will be updated as time goes on, and I wouldn't be surprised if they already have versions available with choices of other feature priorities than they are targeting for HL2.

 

[Pseudonym_]

Yes, the Source rendering engine would have to be re-written to do what Doom³ does. Just because it now has a few projected shadow maps on the ground doesn't mean it's only a matter of adding more projected shadow maps and bingo, Doom³ in source. The way Source does shadows is years old technology, while the way Doom³ does it is revolutionary in games. Source would have to completely change the way it handles lighting and shadows, not just add more of what it has already, because what it has already IS NOT what Doom³ is.

 

[PvtRyan]

yes I predicted about 5 games in total, after the more advanced engines are out and people are playing the games, game devs wont want to license inferior graphics engines. It's not good business.

There were more advanced engines than Quake 3 in terms of graphics, yet many developers chose it over those engines. There's more than just graphics or the lighting system in an engine.
Until engines can do GI/radiosity fully realtime -which won't be until 2007- lightmaps still have a future. Ultimately, lightmaps will be dumped and a fully dynamic lighting solution will be the way to go, but that won't be for a while.

 

[BuddMan]

If what I mean by "pigs can fly" is that pigs wallow in the mud, I guess pigs can fly.

I guess you don't understand the concept of advancement. Games started out using predetermined light maps. Then games moved to a mix of predetermined light maps and dynamic lighting. Doom 3 has moved to a fully dynamic lighting system....it ADVANCED to this. This is why I say Doom 3's lighting system is more advanced than HL2's since HL2 still uses an older method of lighting in that it uses perdetermined lighting maps with dynamic lighting. Many methods of how it uses dynamic lighting and predetermined lighting maps could be somewhat different, but the basic element is the same.

No DX7 game's lighting looks like HL2's lighting.

But basically it is the same thing. I'm talking about the basics.

This distinction doesn't exist. Both are effects, both are part of a complete lighting system.

Yes it does. I'm talking about a lighting system versus lighting effects that can be added to the lighting system. When OpenGL has an ext that enables the use of HDR, you can make a game based on Doom 3's lighting system and add HDR. In fact you can disable HDR in HL2 and still have lighting effects - you didn't disable the actual lighting "system." If you disable dynamic lighting in Doom 3, you disable the entire lighting system that Doom 3 uses (bleh I'm getting tired of using the phrase lighting system). You're trying to say that the effects are part of the lighting system. Yes that is true, but again I'm really just talking about the basics of it.

No. OpenGl doesn't support it natively yet, and indeed doesn't really go over Dx8 level effects in general.

Now this I didn't know. I assumed there was an ext in OpenGL that allowed for HDR effects. Oh well, just something I haven't read up on.

 

[Asus]

The source engine is about realism with dynamic and real-time environments, material interactions, facial expressions, lip-synching, light, shadows, and HDR. It's strength doesn't rely on one or two (lights/shadows) of these alone. I havn't touched gameplay yet but I am talking more than just lights and shadows.

D3's advantage is unified lighting system so everything is made sure to be lite and have a shadow from each source calculated in Real-Time. They don't have to worry if someone forgets give an object a shadow from an unexpected light since the engine does it all. Everything is lite from the same light. Everything from walls/floors to missles/models.

Both have physics and bump mapped models so...

If we are talking lighting/shadowing alone then we have pros and cons for both direct lighting and HDR.
My take on this.
1 - Developer shouldn't forget to shadow objects from lights. Unified lighting is easy for the developers since the engine does it all but it isn't that big of a plus for me the gamer. IMO
2 - Both HL2+D3 are in first person. If I was in 3rd person, direct lighting would be a must since I would see behind the object the flashlight is pointing at and see other possible shadows that I don't see in first person. The benefits of direct lighting are minimal. IMO
3 - IMO, HDR has a bigger impact for the gamer. HDR reduces or increases the intesity of the brightest light or darkest lit room and spreads that indirect lighting around the room.

Course I'd like to see the final results of HL2/D3 and see if they have made any changes.

BTW The Flashlight is much improved and I would think look very similar to Farcry's. hehe

 

[Apos]

Yes, the Source rendering engine would have to be re-written to do what Doom³ does.

Do you even know what Doom3 does? The method it is using? Why are you pretending that it's some incredible trade secret? It's actually a very, very simple technology: it basically just involves a brute force calculation of lightsources across all surfaces.

Just because it now has a few projected shadow maps on the ground doesn't mean it's only a matter of adding more projected shadow maps and bingo, Doom³ in source.

No, it just means swtiching over fulltime something it already does parttime. And it means replacing a lot of complicated but speedy methods with a single, simple, but both effective and very costly method.

The way Source does shadows is years old technology, while the way Doom³ does it is revolutionary in games.

It's one step towards a unified lighting system, but it's not a complete solution. HL2 takes steps towards a unified lighting system that Doom3 also does not take. Neither is complete.

Source would have to completely change the way it handles lighting and shadows, not just add more of what it has already, because what it has already IS NOT what Doom³ is.

All if would have to do is remove the baked in lighting information in the bsp and then render all the lights dynamically. This is not that hard at all.

 

[Apos]

If Valve did want to re-write their renderer to be fully dynamic, they almost certainly wouldn't use simple perpixel lighting with stencil shadows: which is pretty much the simplest and crudest dynamic simulation of lighting, one which places a lot of technical limits on other elements. They would almost certianly go for something like photon mapping (which Halo 2 will be using) or some simulation of Spherical Harmonic Mapping (yeah right!) which allow for better global illumination and color blends (both of which lightmaps are better at simulating the look of, just not dynamically)

 

[Varsity]

Don't be rediculous, gameplay isn't an element of an engine, its something the game developers create through arrangment of situations within the engine.

And Source allows those game developers to do things they have never done before and very probably won't be able to do for a long while afterwards. It's subtle, but it's incredibly important.

 

[Styloid]

Correct me if I'm wrong but didn't I hear in the UE3 vid that shaders are used for the lighting; particularly for the colourful lamp that rotates and lights up a monster? They also said that modders can write their own shaders specific to their mods. Does that mean that you can write a shader to do that kind of lighting (which looked pretty good)?
I just remember Valve saying that many visual effects in the near future will be handled by shaders and that modders for HL2 can also write their own shaders. It just seems like real-time lighting isn't as big a thing as it was a while ago.

 

[Apos]

Sort of: basically most of the neat surface properties we are going to be seeing will indeed be shaders. Real-time lighting is more of a feature than just writing a shader though.

 

[Murray_H]

I wonder what happens when you shine the light into some water...or when you see someone under water turn their torch on/vice versa

 

[oopsmaster]

Eeep! I post something late at night, go to work, and come back later in the evening to find 5 pages of replies!

Thanks for all the interesting discussion folks, I'm sure I read an answer to my original question in there someplace. :E

 

[Crusader]

I wish people would stop with this comparing of games nobody has played...
We won't see which game has won the battle for a year, maybe more... I wonder if some people will come back here and eat their hats, or just slink away to some other forum

Why are you people even arguing about HL2 graphics versus anything else? VALVe already stated they can simply pump up the image quality when they think OUR hardware is up to the task.

That means this whole argument is pretty redundant, as the source engine is capable of much more than any of us have yet seen. It's the hardware currently available we should be arguing about.

This is why all of HL2's art assets were originally created far more detailed and at higher resolution, and scaled down to meet todays hardware capabilities. Also, VALVe seem to plan on upgrading and developing HL2 for years after its release. I wonder which game we'll be playing in 5 years time...

Besides, are we really supposed to feel emotionally charged when interacting with characters that look like they've been varnished? /jk

 

[Falcon(Nate)]

Yea, Doom 3 looks amaaaaazing, but characters DO have a plastic feel to them. EQ2 though, THERE is some plasticide for ya.