Autogen Exclusions

[Guest gorchi]

Wow Arno!I can say that I am a little bit surprised. I thought that Area() block will make exclusion but using two refpoints and the second will not exclude autogen.. interesting.So I persume, that if FSSC/Airport/Groundmaker would use only one area for groundpolys/roads/rivers and refpoints only to start each of the object, we can see also trees on the roads&polys :)? I must make a test also with roads, I think this test should be quite interesting!Oh Arno, can You please make snapshots by the day so the trees could be more obvious?Best regards,Goran BrumenFS Slovenija 2002 teamhttp://slovenia.avsim.net

[arno 1]

Hi Goran,They are by day :). Only I forgot to remove the texture folder in my new landclass scenery map, so therefore the texture didn't load. With the forest texture displayed on it they are even worse visible :). So I think this is the best it can get (it was clear enough for my).I know that Airport indeed uses a single Area and RefPoint for each polygon. FSSC and GroundMaker both have an option where you can fit more then one polygon in an Area. For GroundMaker for example if you don't layer the polygons (then the order in the source controls which one displays first) then they all go in one Area with each an own RefPoint. I haven't test it yet, but in principle I think expect the first polygon all others should be covered by the autogen. I'll continue with more tests to see if this is true....Arno

---

Member Netherlands 2000 Scenery Team[a href=http://home.wanadoo.nl/arno.gerretsen]http://home.wanadoo.nl/arno.gerretsen/banner.jpg[/a]

[rhumbaflappy 113]

Hi Arno.I was surprised to note that autogen appears to be tied to a distance measurement ( so many meters ) rather than degrees of latitude and longitude. Now I am surprised, again, that it uses a simple X,Y rectangular shape for the exclusion.In your example, if a "dummy" refpoint doesn't work, then surely a tiny poly ( perhaps using the same point for all three verticies ) should work. This behavior may also explain why the dummy RotatedCall-TransformCall usage we all discussed earlier may work. And it might be applied somehow to this example.Very interesting, as sometimes we'd want polys to still have autogen.I'm not that familiar with traditional polygons.. but is an invisible poly possible? That could exclude autogen without excluding anything else?Very good topic to study.Dick

[rhumbaflappy 113]

Hi J.R.Does the necessary Area16 flatten suppress autogen by itself? I'm thinking it doesn't. And, assuming it doesn't, then the autogen suppression is from the elevated, invisible runway.That runway would probably conform roughly to the length-width ratios I found on the ground. But, something else I found, was that a runway of less than 10 feet width had little effect on the autogen around it. In fact, a 1 foot wide runway has no detectable effect on the autogen.What happens if the helipad's invisible runway is 1 foot wide, rather than a square? Can we still land? Does the 'copter "fall through", like earlier "runway-less" tests? Or is a 1 foot wide invisible runway still good for helipads, while retaining the surrounding autogen? ( At least I hope it is! )Dick

[arno 1]

Hi Dick,>In your example, if a "dummy" refpoint doesn't work, then >surely a tiny poly ( perhaps using the same point for all >three verticies ) should work. Yes, that might be a good thing to try. But are reversing the topic, we were trying to find out why the autogen dissapeared and not why it doesn't :).>This behavior may also explain why the dummy >RotatedCall-TransformCall usage we all discussed earlier may >work. And it might be applied somehow to this example. Yes, might be something similar to the second RefPoint. Maybe these commands also confuse the method that checks for the autogen size and therefore result in no exclusion.>I'm not that familiar with traditional polygons.. but is an >invisible poly possible? That could exclude autogen without >excluding anything else? Possible. I could just force it to draw the other side of the polygon and that was is of course not visible from the air :).I'll continue with more tests like this, this weekend. But as I have no internet on my room in Brussels, don't expect any posts about it until monday.Arno

---

Member Netherlands 2000 Scenery Team[a href=http://home.wanadoo.nl/arno.gerretsen]http://home.wanadoo.nl/arno.gerretsen/banner.jpg[/a]

[rhumbaflappy 113]

Hi J.R.I'm going to answer some of my own helopad questions. :)I made a building with helipad in the Chicago area.I tested the Area16n flatten... no bad effects on the autogen.I tested the building I was placing... It does effect autogen, and then I used a dummycall to eliminate the effect. It turns out the dummycall was unneeded, as the invisible runway excludes a larger area than the building.I then played with the pad's runway. But the invisible runway needs to be the same size as the pad, or you'll get "fall through".I even tried the invisible runway alone, and it is the cause of the exclusion that is larger than the building.The rough ratios of 2:1 length and 6:1 width are a little oversized for runway exclusion, as the picture will show.http://www.flatface.net/~rhumba/Pics/ChicagoHelopad.JPGI'm attaching the SCASM code as a text file for reference ( some odd coding you may enjoy there! ... I also borrowed some of your code I found lying around ).Dick

[Guest JR Morgan]

Hi again Dick...I think we're proving tht the invisible rwy is definitely an AGN hog and unlike grouped 3D objects, ground polys, etc, the dummy calls don't seem to help that condition. However, for FS2002, the inv rwy is the only game in town (so far) compared with the old Area16 elevated hard surfaces we used in prior sims. You may be happily sitting on an elevated Area16 pad for many minutes and then FS seems to 'hiccup' and there you go -- falling right through the pad.The Area16 poly bgls used for flatten areas seem to be stable in open terrain and useful for local variation of mesh elevations and don't seem to destroy AGN. Beneath elevated rwy helipads, they behave very well until you depart/re-enter your scenery area whereupon they don't seem to always 'refresh' when re-entering the (add-on) scenery area.David Morgan of England (no relation) discovered that placing another dummy (here we go with the dummys again :-) ) helipad at ground level immediately below the elevated one seems to further nail things down when re-entering. That's what I use and it's working pretty well.Yep, I also played with the narrow inv rwys which allow more AGN but as you've discovered, it's pretty hard to balance on 1 (heli) skid on a 1 ft wide rwy :-).About the best combo I've experienced is using a square 86 x 86 ft elevated inv rwy. That size is chosen to be compatible with the Apt26 texture bitmap. Making quick measurements, the AGN seems to be killed equally in all directions around it using the attached macro. You might find some bugs in it or be able to fine-tune it to be better?I'm not sure I'm getting the same AGN destruct areas you're getting, with your code though. As mentioned, as near as I can tell, I think the AGN exclusions are fairly symmetrical around the elevated inv rwys. The top-view pic I posted yesterday is a little misleading because there was actually no native AGN within approx 90 m north of the pad shown in the top-down view (top is north).Unless someone finds a better way, I'm pretty much resigned to use of the elevated inv rwys as elevated helipads and live with the presently known AGN destruct zones. As mentioned, buildings seem to divert the viewer's attention from the loss of AGN, especially when using the FS2000 advanced buildings and dummy calls if necessary; i.e. the below scrnshot shows that one can keep the trees pretty close to the action.The basic macro I used for AGN testing here is atchd.The thing that really has me interested is what's the nature of the FS Engine that lets these 'band-aid' dummy calls work to save AGN :-).CUL -- 'been a long day.. My daughter's coming down from SFO tomorrow to pick up my car (her's is broke and not worth fixing).'Regards;J.R.

[rhumbaflappy 113]

Hi all.SurfaceType interacts with autogen... dependant upon the v2 setting!

; RussianSurfaceType.sca  N64* 10.923'   E49* 10.277'Header( 1 N65:00:00.0000 N64:00:00.0000 E050:00:00.0000 E049:0:00.0000 )LatRange( N64:00:00.0000 N65:00:00.0000 )Area( 5 N64:10:55.3800 E049:10:13.6200 )	RefPoint( rel :Surface_Fail  1.00  N64:10:55.3800  E049:10:13.6200  V1= 6000 V2= 1000 );	the v2 can control the area of autogen excluded!;	a v2 of 999 will exclude autogen in a rectangle of 500 x 999 (?);	a v2 of 500 will exclude autogen in a square of 500;	a v2 of 1000 will not exclude autogen at all;	so the v2 should be equal to or greater than the surface rectangle;	this also suggests v2 is not a round radius, but square x,y with dimensions dependant on the refpoint scaling.	RotatedCall( :Surface_Area 0 0 0 )	Jump( : ):Surface_Area	SurfaceType( 0 500 1000 0 ); 0 = smooth   1= rough   2 = water	Return:Surface_Fail	ReturnEndA

This code affects AGN buildings in the same way. This is the first time I've actually seen the v2 affect autogen. And now, v2 appears not to be a radius at all, but a square, in this case of 1000x1000 meters.Dick

[rhumbaflappy 113]

Hi J.R. I redid a 'floating' helipad in the Russian forest...It does seem that generic autogen trees may only be excluded right under the pad. As per your suggestion, I used a flatten, a groundlevel invisible runway, an elevated invisible runway, and a smoothed elevated bitmapped poly.I'll attach the code... and here's a screenshot with the excluded trees marked:

[rhumbaflappy 113]

Hi Arno."So this makes me conclude that, for polygons, autogen is excluded around everything that is in the first RefPoint of an Area block. For all further RefPoints no autogen is excluded." - ArnoYes, I agree. The first instance of Refpoint usage in a defined "Area" seems to be the only instance that effects autogen, for these polys... and that may have some bearing on "chained" transform or other types of calls.And if you follow your first Poly(), with a Dummy call, none of the autogen is affected in the Area.

Header( 1 41 39 -39 -41 )LatRange( 39 41 )Area( 5 40 -40 35 )LayerCall( :L 8 )Jump( : ):LRefPoint( 7 :R 1 40 -40 v1= 5000 v2= 300 )Points( 0-200 0 -100; 0200 0 -100; 10 0 300; 2)RGBSColor( EF 255 0 0 )Poly( m 0 32767 0 0.0 0 1 2 )RotatedCall(  :R  0  0  0   ) ; dummy call here saves all the autogen in the defined "Area"RefPoint( 7 :R 1 d 300 600 v1= 5000 v2= 300 )Points( 0-200 0 -100; 0200 0 -100; 10 0 300; 2)RGBSColor( EF 255 0 0 )Poly( m 0 32767 0 0.0 0 1 2 ):RReturnEndA

Dick

[Guest gorchi]

Hi Rhumba!Regarding VTP polys and roads. It looks like that autogen can grow even from VTP roads but it is quite rare occassion while on lakes there are no autogen. Autogen trees come really close to lakes but stay away for let say VTP 30 units (the more I work on VTP lines, the more it seems that 1 unit is 1/2 of meter or maybe even 1 foot?).Take a look on enclosed pictures.Best regards,Goran BrumenFS Slovenija 2002 teamhttp://slovenia.avsim.net

[rhumbaflappy 113]

Hi Goran.Great pictures. I think the actual center of the autogen must be within a poly for it to be excluded (?) As these are lines, perhaps the center of the autogen is never enclosed. VTP1 lines ( strips ) like those of cellgrid, are actually polys... I don't recall seeing autogen pop through them. Also, VTPs have a blending at their edges, so the actual edge of a line or poly is a little hard to define.If that is a line you made, try to widen it, and see if the tree remains.Dick

[Guest gorchi]

Wow Rhumba!You were very quick...Oaky, I am attaching two pics, same position, the only difference between them is width of road and one exit of FS :) First has road (roadssu.bmp) with width 26 units, second 226 units. Now let's see by which width it will dissappear... In couple of minutes will know that also!Best regardsGoran BrumenFS Slovenija 2002 teamhttp://slovenia.avsim.net

[Guest gorchi]

Okay, the tree is still on its place with road of width 32, if the road is wider then the tree is gone (and used for some furniture :))Oh Rhumba; CellGrid polys are on which layer? 4? Maybe this is also the reason why autogen is still visible. On shorelines autogen is not visible. And If I remember, layer 4 is still accepted as ground polygon adhering more or less to mesh...Best regardsGoran BrumenFS Slovenija 2002 teamhttp://slovenia.avsim.net

[rhumbaflappy 113]

Hi Goran.So, you have discovered VTP lines do exclude autogen. Good job!For Cellgrid I used VTP Method1 lines... but they were only a width of 3, so they shouldn't easily show autogen exclusion... but all VTPs will exclude autogen, as VTP1s ( fans and strips ) are both polygons. I didn't know if the VTP2 lines would exclude... but you have shown they do.So, VTPs exclude underlying autogen. Actually, they replace autogen.I think if you take that wide line you made, and use a dense city landclass texture type, it might show houses on the "road", as VTPs use their own landclass autogen. :-lol Dick