Sign in to follow this  
Guest product

Terrain Mesh Question

Recommended Posts

Hi All,I've got a question on the terrain mesh setting. What impact does keeping terrain mesh at a low level- say 50-60% - if you fly predominantly over very flat areas? In particular, I now do most of my flying over South Florida (using the excellent land class files available in the library). I'm trying to squeeze out all the FPS I can, and know from flying in more mountainous regions that the mesh slider has a big impact.I guess there are two questions here. First, over very flat terrain, what impact on visuals does setting mesh low cause? I've done some tests but can't be sure if changes I'm seeing are due to the low slider setting or my imagination.And conversely, does setting the mesh slider low over flat landscape provide any benefit in the way of framerates in the first place? Again I've done some tests but am unsure of the mesh impact in this regard.Any thoughts / theories appreciated, thanks!Best,Joel

Share this post


Link to post
Share on other sites
Help AVSIM continue to serve you!
Please donate today!

HI Joel,>I guess there are two questions here. First, over very flat>terrain, what impact on visuals does setting mesh low cause?>I've done some tests but can't be sure if changes I'm seeing>are due to the low slider setting or my imagination.>Very little, if any, impact over flat areas. The slider sets the radius at which the prevailing LOD loads around your aircraft. Higher settings, larger radius, more detail farther out, more processor overhead. So, setting the slider way back will offer hardly any visual degradation in flat areas, and boost performance.>And conversely, does setting the mesh slider low over flat>landscape provide any benefit in the way of framerates in the>first place? Again I've done some tests but am unsure of the>mesh impact in this regard.>>Any thoughts / theories appreciated, thanks!>This will depend on the speed of your processor and your quantity of RAM. Higher-end systems will show a smaller benefit. Low-end systems a larger benefit. If you have a 10 Gh processor and 5 Gb of RAM, it won't matter at all. If you have a 500 Mh processor and 128 Mb of RAM, it could make a big difference.Hope this helps . . .-------Justinhttp://www.fsgenesis.netHigh Quality Scenery for FS200x

Share this post


Link to post
Share on other sites

Thanks for the reply, much appreciated!Best,Joel

Share this post


Link to post
Share on other sites

Hi Justin,>The slider sets the radius at which the prevailing LOD loads around your aircraft. >Higher settings, larger radius, more detail farther out, more processor overhead.That doesn't seem to be the case. Remember this situation? This is a series of images of Diamond Head, Hawaii. The test uses 10m data at LOD12, which means that the radius in which the High Res mesh under the plane is displayed is quite limited. The red line in the first image shows that boundary quite clearly. In the foreground, we see the edge of the high res version of the crater, with the default mesh/crater at the top of the image. The Terrain Mesh Complexity slider is set to 100%http://forums.avsim.com/user_files/22594.jpgIn the second image, I've lowered the setting to 0%, and the beginnings of the high res crater remain, at the same distance, but seriously degraded.http://forums.avsim.com/user_files/22595.jpgIn the third image, I've disabled the LOD12 mesh, leaving just the default active. As you can see, the foreground edge of the crater is gone, confirming that the ridge is, in fact, included in the LOD12 mesh.http://forums.avsim.com/user_files/22596.jpgFinally, I move closer so the entire crater falls within the radius within which LOD12 mesh is rendered. Well worth waiting for!http://forums.avsim.com/user_files/22597.jpgIf I had to guess, I'd say the slider may be lowering the threshold level at which the CLOD engine collapses the polygons, reducing detail. As far as I know, only LOD affects the radius at which mesh is displayed.Steve"mesh matters" at www.fs-traveler.com.

Share this post


Link to post
Share on other sites

I'm sorry, I'd love to understand this but am not totally familiar with the terminology you've used- is there a way to say all that from a novice standpoint in terms of its practical effects?Thanks,Joel

Share this post


Link to post
Share on other sites

HI Steve,My answer pertained to only "flat" situations. When the terrain is predominantly flat, LOD=5 will looks pretty much the same as LOD=13.But . . .From a reliable source:TERRAIN_ERROR_FACTOR=96.000000 -- "This is a value from 0 to 100 that controls the DEM radius (it is linear in area)"-------Justin"Come check out the NEW FSGenesis Web Site"http://www.fsgenesis.netHigh Quality Scenery for FS200x

Share this post


Link to post
Share on other sites

Sure. Greater terrain complexity requires more computing power. To reserve some of that power for other aspects of the sim, you can reduce the terrain complexity.One way to do this is to fly over flatter terrain. Another is to lower the setting of the terrain mesh complexity slider - doing so does degrade the appearance of the terrain. Either one or both together can improve FPS, but how much depends on your system and how complex the terrain is in the first place.Practically speaking, you probably won't see much difference in either the terrain or FPS in your case. But reducing the slider setting can have a significant impact on both in more complex terrain. I believe Justin and I agree on this much. Our so far unreconciled difference of opinion concerns the specific nature of the degradation. Steve

Share this post


Link to post
Share on other sites

Hi Justin,Perhaps we can move ahead a bit on this issue.You say that "The slider sets the radius at which the prevailing LOD loads around your aircraft." This statement seems to be based on information "From a reliable source". I take the "Trust, but verify" approach where possible. My images simply do not support that claim. Now you have revealed the actual basis for your interpretation: "This is a value from 0 to 100 that controls the DEM radius (it is linear in area)". Hmmm. I wasn't an English major(:)), but isn't "linear in area" a contradiction in terms? Sounds more like a phrase from a tech writer or PR person than from a programmer who actually understands the details. SOME DEM data is used from the foreground to the horizon, or nearly so (the terrain may actually be rendered as flat there but I suspect not). So this sentence makes no sense to me at all.Nor to you either, I suspect, so you rephrase it, assuming it refers to the radius of the "prevailing LOD", which at least makes sense. But I don't believe your interpretation is correct. I wonder if it is really telling us that the slider controls the radius of the degradation of the terrain in some way? That might make sense. If you look carefully at my images, you can see that the 0% setting has degraded the appearance of the default crater, which lies outside the radius of my prevailing LOD (12) mesh. I have not examined this possibility in any detail yet, but it may be worthwhile to do so.Cheers, Steve'... The Promethean spirit that animates scientific enquiry, that terrifying curiosity that inhabits the human soul, always proves stronger than the fear of knowledge that opposes it.' - Dylan Evans

Share this post


Link to post
Share on other sites

HI Steve>Hi Justin,>>Perhaps we can move ahead a bit on this issue.>>You say that "The slider sets the radius at which the>prevailing LOD loads around your aircraft." This statement>seems to be based on information "From a reliable source". I>take the "Trust, but verify" approach where possible. My>images simply do not support that claim. >FWIW, the reliable source is a lead scenery developer for FSCOF. I'd rather not elaborate since I may be stepping on some NDA boundary here, I don't know.>Now you have revealed the actual basis for your>interpretation: "This is a value from 0 to 100 that controls>the DEM radius (it is linear in area)". Hmmm. I wasn't an>English major(:)), but isn't "linear in area" a contradiction>in terms? Perhaps, but I interpret that to mean the area enclosed in a "radius" from the aircraft location.>Sounds more like a phrase from a tech writer or PR>person than from a programmer who actually understands the>details. I would think a lead developer for FSCOF would understand the details. :-)>SOME DEM data is used from the foreground to the>horizon, or nearly so (the terrain may actually be rendered as>flat there but I suspect not). So this sentence makes no sense>to me at all.>Keep in mind the context of my answer. I was speaking of predominently flat areas to begin with, not mountainous areas.>Nor to you either, I suspect, so you rephrase it, assuming it>refers to the radius of the "prevailing LOD", which at least>makes sense. But I don't believe your interpretation is>correct. >By prevailing LOD, I mean the "highest" LOD in the database, which always displays automatically.>I wonder if it is really telling us that the slider controls>the radius of the degradation of the terrain in some>way? That might make sense. This is also the case in an inverse way. If it controls the display of the LOD >inside< the radius, it also, by definition, controls the display >outside< the radius.>If you look carefully at my>images, you can see that the 0% setting has degraded the>appearance of the default crater, which lies outside the>radius of my prevailing LOD (12) mesh. I have not examined>this possibility in any detail yet, but it may be worthwhile>to do so.>I think a couple of other cfg settings are also factored in here, for instance, TERRAIN_DEFAULT_RADIUS and TERRAIN_EXTENDED_RADIUS. I think these settings all work together in the formula to solve what is finally displayed.-------Justinhttp://www.fsgenesis.netHigh Quality Scenery for FS200x

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this