Friday, 1 April 2011

Night lights, Ranges and Vehicle Damage

AD posted some screens of the city at night over at SimHQ in his Afghan diary forum topic. Using non-shadowed point lights with EntityViewRange culling, Leadwerks 2.4x handles 350+ streetlights without a hitch.



Damaged / rusting vehicles

Completed the full set. Rusting hulk versions of all major vehicles have been completed for use as firing range targets and set dressing.



Three stages of vehicle, normal, broken/wreck, destroyed.



Out on the firing range they currently look like this...





Firing range observation post.


The Cougar stands in for the now dated Humvee. I'm toying with the idea of adding a selectable ground view (traditionally known as "Tower") which locks onto the nearest observation point, such as these towers.


*edit*

An image from a real passive infra-red system for a comment below...

8 comments:

  1. Brilliant idea three stages -not just two as everywhere.Vehicles broken / wreck no longer represents a threat?
    I understand that there will be any random damage, just a single piece of the damaged vehicle?(random scattering wheels for example)
    Is there any chance of falling off a tank turrets?;)
    In Gunship! microprose it's great solved.
    And the final question-whether the damage to one vehicle will have an impact on neighboring damage?In EECH this is a well solved, one hellfire can destroy 4-5 vehicles if are quite close.


    Btw you are the last hope for a good OPTIMISED simulator .All recently
    issued simulators are simply unplayable, even the most expensive computers :(
    Few days ago, released il Sturmovik, which practically does not work.
    Before a DCS a-10, black shark, all these simulators may extreme hardware requirements.
    I hope that CH kick all ass.

    ReplyDelete
  2. Not sure what you mean by Optmised. It's GPU heavy and strongly dependent on fill-rate. That's an engine requirement.

    There are many things we can do to let you tweak performance, the heaviest are post-processing effects and cockpit lighting, these are all part of user options. I'd go with an nVidia 460 for happy performance. Costs for those have come down. But You might also get away with a 9800 GTX or equiv and that's a two (three?) year old video card?

    It needs a solid GPU as everything from lighting, terrain, tree sway, lighting etc is all done in shaders. your CPU is left with the tasts of filling it up and processing avionics/game world.

    I guess we'll have to have some benchmark program. But Steam hardware surveys suggest over 80% of their customers have PCs that can run it.

    ReplyDelete
  3. Simple answer-optimized = Apache Air Assault.I know
    little map, not dynamic campaign,few missions,arcade game, but can do effective exquisite engine.

    ReplyDelete
  4. Cool! I like textures. Very very nicely done. Im also happy with the three-phase damaging. Question: its obvious that these wrecks no longer emit heat signature. How pilots train to use their FLIR for engagement? Or do I sound completely nonsense? :-)

    ReplyDelete
  5. I shall have to put that question about FLIR training around when we come to implementing heat maps. Quite a good question.

    We're still undecided on which technique to do synthetic FLIR imaging for Combat-Helo.

    Heat signature I would guess...isn't much of a factor since non of the weapons you deploy are heat seeking and your sensor suite doesn't rely of heat either. Modern FLIR looks pretty different to what it used to look like. I was advised not to waste too much time degrading images to blobs, which was sound advice. Take a look at the image I just added at the bottom of this blog entry.

    Looking at the image, first things come to mind, a lot of resolution, no smearing or bloom. It's also remarkable in that there's little heat signature from the engines, testimony to heat baffle efficiency.

    Synthetic heat maps would be my preference, they are very low resolution and can be combined with existing materials in a pixel shader.

    ReplyDelete
  6. It looks to me as if the Apache in that FLIR image is being illuminated by something (the source would be off to the lower left of the image if it is). If it's just passive radiation (rather than reflection of a source) I don't understand the distribution at all. Any ideas anyone?

    ReplyDelete
  7. I would suggest the ground is a large heat source as evidenced by underside of rotor blades (at the right pitch angle to reflect) and stabiliator. I wouldn't rule out a composite of several wavelengths either.

    ReplyDelete
  8. Picture is great. I really dont know, how to create this kind of imaging into sim engine. Ive been always a little confused on what source or principle the FLIR systems are working. Once they are talking about thermal imaging once they are talking about infra-red imaging......anyway...Im looking forward to see, how its gonna look, when You implement it to the sim. Once my father told me about the infrared device and it was about the infrared emitter and the reciever. The final image and its green intensity was dependent on the speed of the return the emitted infrared waves. I dont know. Maybe its wrong. :-o

    ReplyDelete