Guest JonS Posted November 16, 1999 Share Posted November 16, 1999 Hi All, I've just been flicking through Fionns' review of CM at TGN, and came across the bit about weather – there will be fog and snow and rain, and a host of other environmental conditions available for scenarios. ‘Cool,’ I thought, ‘variable weather.’ Then something occurred to me. If we can have TWO levels of fog, and all the other weather treatments, why not a better handling of smoke? After all, for most - if not all - practical purposes fog and smoke are the same. In one of his replies to an earlier smoke thread Steve raised the issue that to accurately model smoke dispersion (incl. wind eddies, drift, differences in surface temperature, humidity, accurate wind currents around buildings and trees, et al) would require – to paraphrase broadly - the services of a professional weather man and an SGI workstation. Well, what can I say? I think this answer is disingenuous at best. Is this stuff modelled for the different weather states? What about factoring it into the ballistic path of projectiles? I have no way of knowing, but I would suspect not. Granted, the effect on DF weapons, especially at the scale being modelled, would be slight. Weather is a different story though – rain never falls consistently for 30-60 minutes, at least not where I live. At twilight the visibility steadily, and quickly, deteriorates as the sun sinks. Vice versa at dawn. Fog can quickly lift, and just as rapidly settle. It can also be very localised – visibility 10-15 m in one spot, and unlimited just 30m away. This is especially the case with changes in altitude. I can’t give references for what I have just stated, but I can cite many years of empirical evidence to back this up. As I said, I have no way of knowing whether this stuff is modelled in the final game. If it is – fantastic! If it isn’t, then using the difficulty of modelling real weather as an excuse to give smoke limited handling seems a little lame to me. To get this post back on to a constructive footing … could someone give me a quick brief on how fog (and other weather) is modelled in CM – including the practical effects on gameplay. Also a short bit on why the same model - as it currently stands, warts and all - can or cannot be used as a proxy for smoke? I understand the VRAM thing – well actually my technical grip on it is a bit shaky, but I accept it – and why that limits options with regard to smoke, but I can’t see why that same argument wouldn’t effectively preclude fog. Thanks Jon ------------------ Quo Fas et Vino du Femme Link to comment Share on other sites More sharing options...
Fionn Posted November 16, 1999 Share Posted November 16, 1999 Ok, I read this and figured I'd take a few shots from a scenario I was tweaking. This is a scenario set at night during fog. Night= reduced visibility. Fog= reduced visibility. Fog at night = can't see more than 20 metres or so usually. it's really bad. Of course the COMMANDER can see more than 20 metres otherwise plotting moves would be impossible and the movie would suck bigtime. PLEASE bear in mind that jpeg compression was NOT kind to my shots and that my monitor brightness and contrast settings will have a BIG impact on how these shots look on ur monitor. If you have a REALLY bright monitor then the fog at night won't look all that impressive as CM's graphics from which I took this pic are adapted to my monitor. Basically I am trying to say your settings could screw up the darkness and contrast since I probably use different setting to you all. With that said here's a few shots. Go to http://combathq.thegamers.net/pix/fog1.JPG NB. it must be JPG and not jpg I think. then go to fog2.JPG sitrep.JPG and zoom.JPG Fog 1 = high level view from about 300 metres or so from a town. 2 buildings are already on fire. fog 2= As the Panther advances some of my arty lands near it and the town becomes more visible sitrep.jpg = a platoon commander's-eye view of one of the roads leading into the village. Can't see much can you? Welcome to warfare at night zoom= using zoom x 4 from the position in sitrep.jpg I've taken a nice shot of the Panther at the LIMIT of the player's viewrange at night ( the troops could only see about 40 metres from the location they were at while the tank is about 400 away but I can see it since I'm the god-like commander and I fiddled with the controls etc to get the shot ). Enjoy. As for the question.... the way fog is modelled ic CM is VERY different to the way the smoke is modelled. I know you can't realise this until you've actually seen fog in the game ( the screenshots don't give you a good enough impression to draw conclusions BTW) but the way fog is modelled is inherently different and is so obviously different just from looking that it seems obvious to me the code etc etc for modelling fog can't be transferred to model smoke. I hope that makes sense.. See with computers there can be two ways to draw two similar things which look grey and similar. However fog is basically something which blocks view progressively over a long distance while smoke is just grey and stops view. The way you code those two pretty similar-looking things is probably very different and thus whilst superficially similar you couldn't use smoke code instead of fog code or vice versa. That's just my take on it but I think I'm pretty close to what it is... Maybe when more VRAM comes available though. Remember, there are actually people running this on Pentium 100s with NO 3d card which is pretty amazing when you think of it. ------------------ ___________ Fionn Kelly Manager of Historical Research, The Gamers Net - Gaming for Gamers Link to comment Share on other sites More sharing options...
kingtiger Posted November 16, 1999 Share Posted November 16, 1999 Fionn, Nice picture. Hhhmm, Uhmm, Err... is that a little King Tiger lurking in the fog. Richard Kalajian Link to comment Share on other sites More sharing options...
Guest Big Time Software Posted November 16, 1999 Share Posted November 16, 1999 JonS, smoke and fog are totally different elements. Fog is persistant and constant in CM. The scenario designer determins if it is going to be present or not. There is no variability. Smoke, on the other hand, is something that is far too variable to deal with. Previous threads have expanded upon the reasons why. Short of it is "fog is doable, smoke is not". Steve Link to comment Share on other sites More sharing options...
Guest JonS Posted November 16, 1999 Share Posted November 16, 1999 thanks for the shots Fionn. Clears things up for the fog (eh?) But Fionn, your comment "However fog is basically something which blocks view progressively over a long distance while smoke is just grey and stops view", in my experience, just isn't true. Smoke requires depth and body for it to block LOS. Its a little hard to explain, but if you've ever seen smoke grenades in use you'll know what I mean. 1 or 2 of those things make little impact on LOS - it takes lots of them to create a screen. It's the same for artillery delivered smoke - it can take 3-4 minutes for a decent smoke screen to build up, and even then objects moving through the smoke don't just suddenly appear in full clarity on the near edge of the screen. They appear to emerge as if out of, well, out of fog. Of course the distances may be compressed when compared to fog, but not that greatly. The example you give is 20m (at night), which is comparable to a good smoke screen. The real difference I suppose is that smoke occurs in discrete blobs across a battlefield, while fog can be approximated to being the same throughout the combat area. Later: Oops, started writing after Fionns, but before BTSs. Ok, thanks guys. Regards Jon ------------------ Quo Fas et Vino du Femme [This message has been edited by JonS (edited 11-16-99).] Link to comment Share on other sites More sharing options...
Guest Bobb Posted November 16, 1999 Share Posted November 16, 1999 I very much look forward to playing a really good representation of night battle as Fionn's screen shots depicted. It is good as is. I also look forward to future developments as this new to computers "science" is in its birth pangs. The cutting edge of advance is but the infant of things to come. Depicting night effects no doubt will gain greatly as hardware becomes more capable. Vision at night is much more monochromatic than depicted so far. Those color receptors just don't function at low levels of light. On my computer the effect of fire seen through fog could have been better. The brightness of fire seen through fog is still bright enough to be spectacularly colorful. Here the effect I saw was too monochromatic. It never looks greyish in my experences. Seems that I can't be satisfied, complaining of too much color on one hand and too little on the other. It is my observation that CM's basic design concept contains a difficulty of which I am reminded by Fionn's indicating that the Player-Commander's view is different from his troops's view. Theirs is of the abstracted LOS kept much closer to the reality of night, whereas the commander's is greater so as to make it possible for him to give orders to his troops. This stems necessarily from the several hats that the screen view must wear. That is to say, it is at the same time all of the the following: a map; a commanders view, the commander's map as updated by staff; subordinant commander,s views and maps; trooper's views. These many functions have been very cleverly combined into a workable game system that is obviously successful and fun to play. Yet, I think it must be recognised that this entails some compromises that may occasionally be noticable. It will be interesting to watch this in future developments. I can envision separation of map and views. A commander seeing both from his position and the map: individual units seeing from their positions positions: the player having use of both sets of views combined and giving orders via map. And, then the a movie given of the action phase as is presently implimented. One previous thread anticipates somewhat this state of affairs by voluntary confining views to the commander's position forgoing moving the camera to others. I am sure the folks at BTS would have some strong opinons on this picture of posssibilites. For now they are wisely confining their efforts to the job at hand; but, since my hardware is and will be waiting for upgrading for a while, what else have I to do than visualize and speculate? Link to comment Share on other sites More sharing options...
Lee Posted November 16, 1999 Share Posted November 16, 1999 Oh man, the notion of poking around in that town is frightening. Looks exceedingly dangerous... Link to comment Share on other sites More sharing options...
Fionn Posted November 16, 1999 Share Posted November 16, 1999 JonS, Thank god Steve stepped in. What he said is what I was trying to get across. I'm not good with techie stuff especially since I'm just describing what I see. Bobb, yup, game design is all about compromises. make the right ones and it works out. Make the wrong ones and... Lee, yup, the Panther is actually within the American position and has US troops all around it without even knowing it. Night combat in CM is as vicious, quick and deadly as you've heard. ------------------ ___________ Fionn Kelly Manager of Historical Research, The Gamers Net - Gaming for Gamers Link to comment Share on other sites More sharing options...
Los Posted November 16, 1999 Share Posted November 16, 1999 Hey guys since we're talking about nigt how about a quick discussion of illumination? First off, nice pictures Fionn. Is there a way, within the engine, to illuminate a certain radius around any given point once a starshell of PL-fired starcluster is fired? (It would be easy for me to get the specs of what that radius should be.) While it might be a large portion of the map the shells would not last very long and would probably be in limited supply thus requiring their judicious use. If the answer is yes, previously hidden units or sound contacts could then show up with some reater level of intel. If no, then you could still have the units show up without the graphical representation. Another concept, flashes or some graphical representation of firing which takes place out of your night visual range. You don't see the actual units of course depending where you are observing on the map) but you see tracers and better yet flashes. (Like your opaque explosions but more brighter and yellowish?) Regarding battlefield smoke it would be nice as a scenario goes on, in say dry conditions, that the fog model begins to show up over certain areas to simulate the inevitable fires and smoke that always erupts from fighting. Just treat it as fog, though I understand the technical limitations of the engine might not support it. It's just that not modelling it at all is less realistic then modelling it only half right. Los Link to comment Share on other sites More sharing options...
Recommended Posts