@ "The eyes glint based on a RADIOSITY calculation and local illumination. They"
This qute is 100% accurate. It does not states that its using real time radiosity. Make an effort and read the HL2 shading model paper. And learn what the term "radiosity" means, because with all due respect, I do not think you understand the term. Then you will see why Gabe statement is true, and does not imply real time radiosity. I do not try to discredit anyone. Rad that phrase very carefully, and tell me where is "real time radisoity" mentioned in that phrase. Please pay attention to details. They make all the difference. In this case the detail is the very problem debated :P @It would make sense to base the realtime off pre-processed data Yes it makes a lot of sense. I did not debated support for off-line radiosity processors. I did not debated that. It was about real-time radiosity solution, and the claim to support this, Do you see the difference ? "preprocess .... " "real-time" , 2 different terms . ----- Original Message ----- From: "HoundDawg" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Monday, November 29, 2004 6:24 PM Subject: RE: [hlcoders] Realtime radiosity & Source ? > > @ " G-Man's eyes utilizing "realtime radiosity lighting"." > > > > Hmmm , I dont think so. Those WEB pages and sites > > reproduces a flawed technical statement. Check the paper > > above, it will > > enlight you how dynamic geometery is shaded in Source Engine. While > > As much as you'd like to discredit what doesn't agree with you, you can look > here: > http://www.hl2source.com/?content=news&month=05&year=03 > > ...and notice that it states Gabe talking about it with this quote: > > "An enormous number of details go into creating a character like the G-Man. > The eyes glint based on a radiosity calculation and local illumination. They > self-shadow and follow you as you move." > > They showed it off at E3, which wasn't a pre-built movie and was an > interactive G-Man. > > > it looks extremly cool, it uses a solution which SAMPLES at run time > > data from stored by the off-line radisoity preprocessor. And depending > > by the desity off the data stored by the radiosity solution, > > this can or not > > result in accurate dynamic model lighting. Im sure it looks > > great most of > > the time, > > and results in great lighting effect. > > It would make sense to base the realtime off pre-processed data. After all, > the maps control the lighting, which is processed by vrad. This would then > explain that other elements, like model actors or whatever the coders > create, can use the same principle to create the same realtime effect. > > Of course, this isn't as realtime as other engines, but other engines don't > exactly provide MOD teams with the level of tools that Valve does either. > Most of them only release a scripting kit rather than a C++ base SDK. Makes > me wonder what they're hiding? Parts of the HL2 code (or other games) > maybe? > > > > @ Until we see the SDK, we really won't know to what extent > > the realtime > > radiosity exists. > > > > A SDK containing game code wont enlight us on this subject. > > Anyway, a person > > which > > played HL2 , and with minimum 3D knowledge, can easily verify > > this claim. > > We have seen a great deal more in 3rd party HL1 MODs than we ever saw in > HL1, or even in TFC. It is very possible that after the SDK is released, > other coders would utilize the tools and underlying engine to create the > desired realtime effects some of us would like to see in HL2, but wasn't > available (for whatever reason). > > IMO, HL2 SP only gave us a small sampling of what can be done with this new > engine. A very good base to work off of. I'm eager to see what others on > this list can do with the SDK. I've seen a lot from the last one, and I > know that we're going to see a great deal more this time around. > > > - HoundDawg > > > _______________________________________________ > To unsubscribe, edit your list preferences, or view the list archives, please visit: > http://list.valvesoftware.com/mailman/listinfo/hlcoders > _______________________________________________ To unsubscribe, edit your list preferences, or view the list archives, please visit: http://list.valvesoftware.com/mailman/listinfo/hlcoders
