Jump to content
The Dark Mod Forums

nbohr1more

Development Role
  • Posts

    12100
  • Joined

  • Last visited

  • Days Won

    195

Everything posted by nbohr1more

  1. Yeah, I can't resist making with the "peanut gallery" \ "color commentary" et al. One of my superiors at work has an email signature that I really have trouble abiding:
  2. I imagine that this is kind of a call to the community to make versions of these for the existing assets ??? If so, do you have a list of the assets that have gone through this process? Maybe we could create a wiki and any mapper could write "Got it" next to entries that are listed as incomplete? (or just put the table on the "discussion" tab for the existing tutorial wiki?) (I will even try this if I can squeeze enough PC time from my restrictive wife...)
  3. Yep, sikkpin's definitions show many vfp's in one material definition. It appears that he is using conditional "If" shaderparms for each vfp invoke. Just following by example you could probably combine the two vfp's with a meaningless conditional (one that always satisfies the "If")? postProcess/softShadows { sort postProcess // soft shadow sample { if ( Parm3 == 0.0 ) Program ss_sample.vfp fragmentMap 0 _ssRender fragmentMap 1 _currentRender } // shadow mask blur { if ( Parm0 == 1.0 && Parm3 == 1.0 ) Program ss_blur_box.vfp vertexParm 0 Parm1 // offset scale vertexParm 1 Parm2 // epsilon fragmentMap 0 _ssMask fragmentMap 1 _ssDepth fragmentMap 2 _currentRender } { if ( Parm0 == 2.0 && Parm3 == 1.0 ) Program ss_blur_poisson.vfp vertexParm 0 Parm1 // offset scale vertexParm 1 Parm2 // epsilon fragmentMap 0 _ssMask fragmentMap 1 _ssDepth fragmentMap 2 _currentRender } { if ( Parm0 == 3.0 && Parm3 == 1.0 ) Program ss_blur_x.vfp vertexParm 0 Parm1 // offset scale vertexParm 1 Parm2 // epsilon fragmentMap 0 _ssMask fragmentMap 1 _ssDepth } { if ( Parm0 == 4.0 && Parm3 == 1.0 ) Program ss_blur_y.vfp vertexParm 0 Parm1 // offset scale vertexParm 1 Parm2 // epsilon fragmentMap 0 _ssMask fragmentMap 1 _ssDepth fragmentMap 2 _currentRender } // final blend for unblurred shadow mask { if ( Parm3 == 2.0 ) Program ss_blend.vfp fragmentMap 0 _currentRender fragmentMap 1 _ssMask } } _ssRender { { map _ssRender } } _ssMask { { map _ssMask } } _ssDepth { { map _ssDepth } } ssRender { sort postProcess { map _ssRender scale 1, -1 translate 0, -1 } } ssMask { sort postProcess { map _ssMask scale 1, -1 translate 0, -1 } } ssDepth { sort postProcess { map _ssDepth scale 1, -1 translate 0, -1 } }
  4. Good luck ungoliant! I wonder if you could just call each vfp in a separate stage in the same material definition? I know that even sikkpin uses the material definitions to pull-off tricks that cannot be done entirely in one vfp. Another sikkpin trick is to place an invisible patch in front of the viewer so that post processing effects can be rendered to that surface and look like they are being applied to the whole scene. (you should check out the Test 5 pk4's at D3W...) I will do so again as well to see if I can fathom anything relevant...
  5. Those pretty new Melan textures are all (compressed) DDS formatted, correct?
  6. I have been pouring over the vfp's with my scant free-time and see that there are some digestible operations. I think with that, unlike other engines, the material definition and blend design process is a good stepping-stone to "shader understanding" as it is a close parallel to the way that hardware actually processes these operations. While I doubt that many members would ever come close to mastering the existing material blend system enough where they would need to "move-on" to the GPU level, (I don't think anyone has actually fully mastered it anyway...) it might be helpful to understand (to some degree) how a vfp affects the results of a material blend (for example). It would be cool if we had a simple explanatory example for ARB vfp syntax example in a similar manor that this modwiki article shows explicit blend syntax: http://www.modwiki.net/wiki/OpenGL_Blendmode_%28Materials%29 It obviously won't help for shaders that are like: "See this complex 20 variable equation? I'm approximating it by multiplying two registers and subtracting the following decimal value." ...but perhaps some degree of what is "under the hood" can be de-mystified and that may (at least) lead to better material designs. I personally would love to know how Doom transmits cubemap samples to a material stage and whether a vfp can inherit custom cubemap data or be coerced to perform a cubemap sample operation from the six tga textures it is made of? Any takers for a layman's ARB shader wiki? (please )
  7. Given my limited DR time, this is probably silly but I only have one issue so far... I have been trying to line-up a detailed light projection-image with a far wall between two walls that have architectural decor. The image is designed to account for the decor but was designed from an Orthoview where it was easier to align. If there was a way to make the camera view render "Ortho-Style" (no 3D distance perspective) or if there was a way to show the projection image as a transparent overlay for the Orthoview grid, then this alignment would be much easier. Feel free to laugh if there's some two-key solution that I've overlooked by not finishing all my DR training
  8. I don't know what you are up to Sparhawk but that comment amused me... But, yes, now America is riddled with the malady of risk aversion and "know-it-all". Look at the story of the VCR, we scared the inventor overseas to our former enemy, Japan (when we were still ambivalent about the Japanese even, rather than now where we enjoy and mass-consume their cultural contributions...).
  9. Demagogue has introduced a more interesting angle on this discussion. We have "smart-phones" that self-destruct when an end-user tries to open and modify them. That has to be so much more stifling than <whatever-impediments-may-have-been-present-in-the-past> for innovations. I dread the days when the very raw materials themselves are embedded with these kinds of self-destruction methods. You won't be able to make a popsicle-stick bird-house without corporate approval...
  10. I see that my examples are also a little "sensationalized" as well. I just like the narrative of TDM members coming up with miracles. It's a better story in my mind. For example, I like to think that Baddcog could make the worlds best looking low-poly objects. He has one of the best eye's for "economy of style". So it's legendary that he would help with LOD which needs that skill. (I would love to see Baddcog's really high-poly stuff too). See, it all makes for a good story when it's exaggerated a little.
  11. Yeah, Edison wasn't an angel by any means, but I idealize Nikola so much that I like to portray him as "engineering inside his mind". Most of the European engineering establishment had written-off Nikola's ideas as well (or at least that is what is written in our sensationalized versions of the story ). Foolish corporate activity is definitely another wet-blanket. Maybe the idea that half-baked ideas can lead to cool new stuff is just comforting to me as I am always running around with partial understanding under my belt
  12. I have always been fascinated by the rise of Thomas Edison and his nemesis(?) Nikola Tesla. While I believe that Nikola was clearly the superior of the two, Thomas Edison seemed to embody this naive sense of "Can do" that we associate with the golden era of American innovation. Edison was cranking out gadget after gadget that the engineering minds of Europe had deemed "impossible", he was too dumb to know that these things "shouldn't work". Now that America has mostly "grown-up", it seems that all the "impossible" innovations are coming from Pacific rim nations that still seem to have a sense of mysticism about what is within the realm of possible rather than a stern engineering mind-set. I think this concept of "just try it, even if it makes no sense" can be applied even here to this mod. For example, many of the better veteran mappers had concluded that Melan's "Return to the City" was simply a design that could never be made to visportal well, yet Bikerdude simply put some elbow-grease in and "trial 'n error"-ed his way to 60FPS. Until the Ground Zero mod, LOD was deemed a wash for any performance benefit by most of the smartest folks at Doom3world. Conversely, Rich_is_Bored pulled-off a more Nikola Tesla style innovation by carefully studying Vertex blending properties for his "Real Translucency" method. Are there any other examples that you know of where not knowing what's "impossible" has lead to a break-through? Would it be better all around if we encouraged more random trials for things that engineers and scientists have "written-off"?
  13. My best friend was getting pretty good at the ole skateboard in his mid-twenties. He was always visiting skate parks and centers. One day, in a parking-lot near his apartment he went to grind a small curb and fell backwards to fracture his skull. He has been pretty shy about 'boarding ever since. Be careful. And Get well soon.
  14. Clue found: The Normalization Cubemap is a mathematical structure used to help bypass hardware limits for normal vector angle calculations. (Therefore has no direct bearing on precomputed cubemap based "Image-Space Lighting" solution approaches... ...) More studying...
  15. nbohr1more

    HI GUYS!

    That's: www.thedarkmod.com Biker... "www.darkmod.com" is a nasty cyber-squatter site... (thank you FF noscript add-on). (although... the Domain Host may deserve the fallout from visiting the wrong site anyway... )
  16. Thanks Biker! Title is still pending eh?
  17. Now that I've gotten this silly hack idea out of my system, I have tried to fathom some of Rebb's work in the ambient fix. I see that the current implementation essentially looks for an RGB 01 value from any (?) projection image's alpha to essentally say "Don't paint ambient light here" (?) I see that an experimental build is sampling two "Light Normalization" cubemaps. Which means that each face of the cubemap is the average inbound light to the player position? Could this system be made to dynamically accept more detailed (pre-rendered) "scene" cubemaps that would be weighted against the normalization cubemap depending on the player position to the scene? All the mapper would then have to do is collect an envshot (with lights added to simulate bounces) and plug the cubemaps and positions into "the system". (Maybe you guys are already cooking-up something akin to this...) (Probably missing some clue here anyway )
  18. HappyCheeze's "Internal Halloween FM" is slated for release to the public "Sometime in the future" and thus has been added to the Upcoming FM wiki.
  19. Thanks for the info HC! I will place this on the "Upcoming Fan Mission" wiki with an "Unknown" arrival date !
  20. I have a feeling that you are going to bring about some interesting changes grayman. Can't wait.
  21. 1) Springheel is waiting on vocals. 2) Horse is being fixed? 3) "Damagable" system SDK framework is being upgraded...? 4) The devs need to put a tiny unfathomable flaw into the code and let grayman try to fix it (for hazing purposes).
  22. Yep.

    Just collecting references, etc.. :)

    More:

    http://www.moddb.com/games/overgrowth/news/simd-optimization

    1. nbohr1more

      nbohr1more

      C:\Windows\System32\cmd.exe /C start "" /affinity 1 "C:\whatever.exe"

  23. Heck, I'm just happy about the foliage. (Plus I see a Horse is being worked on... )
  24. According to the wiki: http://wiki.thedarkmod.com/index.php?title=LOD ... the LOD slated for v1.03 wont be applicable for AI. I suppose you could make an AI script that inherits distance checks from the Interleaved Thinking mechanism then re-skins the AI (If I've read my D3W threads correctly, a Skin can define a whole other model...), but the new "skin" would have to be rig-able with AI animations so I'm not sure how much you could pare-down... Or spawn an AI in place of a statue? (just silly talk... I hope no Holy-War comes of this... )
  25. You have a great mind for details. I can see how the other (early) Doom 3 modders that tried and failed to get LOD to work would have been stumped by these issues. Kudos and Great Job!
×
×
  • Create New...