night lighting - any plans that can be revealed?

[Chock 17,769]

It may have taken LM until a MUCH later iteration of P3D to get some decent lighting in there, but that doesn't mean it will take that same length of time for some decent lighting to go into FSW. The only way to know for sure about any of that and what is and is not developed or in development, would be to actually be part of their development team in a position to see what their long term plan is, and good luck with that considering LM are the guys who make super secret aeroplanes and such. One can nevertheless assume that LM's priority with P3D is not to please casual flight simmers, but to have a working usable platform for its true application as a mission simulator for its serious clients, such as those training to operate the Joint Strike fighter off new carrier ships.

No combat mission planner who uses it it for that (i.e. its intended purpose and the reason LM make the thing) is ever going to say, 'well we can't use this for planning our assault on those hijackers, the lighting isn't convincing and we need more attractive volumetric cumulous clouds'. Take a look at the clouds in a Level D simulator, they look like a joke compared to what we can have in FS; the nice visuals may be there of course, but it's simply not a top priority for a mission simulator to have pretty visuals, it's response time and fidelity to the real craft's movements which are the priorities for such sims. These are what LM will be concerned about, and that is a good thing of course if you really want a realistic simulation of a craft. I'm not saying they won't want it looking realistic, of course they will, and have clearly lready done some work to make that so, but pretty lights are not what will sell P3D to a company which wants a tactical mission planning platform. As you probably know, P3D can simulate night vision with a green overlay, but anyone who has used night vision will know that bright lights at night can cause flares and light trails on the NV equipment, so I'd imagine simulating that would be more of a priority for them than to have beautiful crepuscular sunbeams in the thing so flight simmers can post lovely screenshots, whereas with FSW on Steam, the ability to post lovely screenshots on their product pages is going to be a bit more of a priority if they want to sell stuff to gamers.

It already is apparent that the lighting in FSW is not stock FSX when you see how the lights work (or in some cases don't work) on ported over FSX aircraft. DTG have quite a lot of experience in how to make things look good, you only have to look at a screenshot of their train sims to know that. Now granted, their train sim is using the Unreal engine, so it isn't directly comparable in terms of how one might implement lighting in FSW, but the fact that DTG chose to use that engine for the train sim is at least indicative that they are aware how much good looks on screenshots sell stuff, and they will be similarly aware of that for FSW.

So FSW and P3D will probably become more and more widely divergent from their FSX origin as time passes, and given the differing focus of the two platforms, i.e. one is a serious tactical mission planner and the other is a flying game sold on Steam for 20 quid, I suspect what we'll see, is FSW getting prettier and P3D gaining more fidelity in terms of simulation, with them both going for the other stuff as a secondary priority.

[SierraHotel 976]

12 hours ago, rtodepart said:

i was also hoping for some lighting changes on P3Dv4 but as of today for what i been told v4 did not get any changes in the lighting department 

 

Did you not watch any of the videos? One of the main changes is the lighting.

 

[Paraffin 2,197]

1 hour ago, Jiri Kocman said:

P3Dv4 will have limited light source - 256 - if i remember correctly. Nothing huge and will be used on some specific parts of scenery or AI simobjects. 

Just 256 light sources? That isn't enough for a large airport, let alone a city with all its street lights and car headlights. Once you go with individual light sources, it starts to look very obvious where you're not seeing them. Like the complaints about the limited LOD radius of night lighting in XP11. 

Maybe LM's priorities are different and that's the reason. I hope it's not a limitation of the underlying code, because FSW needs more light sources than that, if they're able to build in dynamic lighting at some point.

[TechguyMaxC 516]

1 minute ago, Paraffin said:

Just 256 light sources? That isn't enough for a large airport, let alone a city with all its street lights and car headlights. Once you go with individual light sources, it starts to look very obvious where you're not seeing them. Like the complaints about the limited LOD radius of night lighting in XP11. 

Maybe LM's priorities are different and that's the reason. I hope it's not a limitation of the underlying code, because FSW needs more light sources than that, if they're able to build in dynamic lighting at some point.

Modern 3d engines aren't really designed to handle individual light sources on the order of thousands (or even millions) of lights.  The hardware just isn't fast enough.  You can do point source lights and not perform the calculations for ray tracing, but then the light sources look unnatural.  Sure, there's a point which appears to be lighter than the surrounding area, but the area of illumination is generally limited to a small sphere in a pre-calculated radius.  

So what we have instead is a physical limit on the number of dynamic lights (in order not to overwhelm the hardware), which do cast rays that are calculated and displayed properly.  This doesn't mean LM and 3rd parties can't still use point lighting techniques for distant lights off toward the horizon, and dynamic lights for those closer to the camera.  In fact, that's precisely what I would expect to see.  

The major sticking point for me in all of this comes down to whether or not they have a true global illumination model for major light sources (i.e. sun and moon).  If they've done so, the overall quality of lighting in the sim will be drastically better than the static lighting model which we've seen thus far.  This is the reason you hear 3rd party developers constantly talk about "baked" lighting and shadows in their add-ons.  FSX/ESP in all prior variants has made it so that developers have to perform their own lighting calculations during development and create different textures to use depending on the time of day.  

 

[n4gix 4,947]

5 hours ago, Jiri Kocman said:

P3Dv4 will have limited light source - 256 - if i remember correctly. Nothing huge and will be used on some specific parts of scenery or AI simobjects. 

The dynamic lighting will also be used on aircraft, which from the pilot's POV is very important. Having "lights" that actually illuminate all visible objects is very important IMHO. Also, aircraft navigation/strobe lights will now illuminate the ground as well. This is something that I and other developers have had to struggle with using multiple effect files, along with some fancy scripting of XML to accomplish ever since FS9.

For the most part, airport terminal ramp lights, and moving vehicles having "lights" that illuminate other objects is also a major plus from the standpoint of night operations.

[fsxkitty 62]

Something that really breaks the immersion is when you're flying at night and turn on the cockpit lights. The entire cockpit textures get replaced with the "lit" textures, and as a result there are no textures for a second or two.

It would be really great if there was some dynamic lighting, at least for cockpits.

[Gregg_Seipp 565]

Someone asked somewhere if P3Dv4 lights will illuminate clouds and I didn't see an answer.  Anybody know?

[sightseer 895]

you guys do know this is the FSW forum right?

[Gregg_Seipp 565]

19 minutes ago, sightseer said:

you guys do know this is the FSW forum right?

Sorry.