Dense forest (Full Version)

All Forums >> [Current Games From Matrix.] >> [World War II] >> Panzer Command: Ostfront



Message

leridano -> Dense forest (6/4/2011 12:57:37 PM)

Love the graphics of this game, particularly now because of the many changes that has been introduced regarding this. PCO has achieved far more inmersion than the other two titles. But one thing I miss is about dense forest terrain. In Russia there are many dense forest areas but in the scenarios we can only see light forest areas. Thoughts?








HintJ -> RE: Dense forest (6/4/2011 1:10:03 PM)

Sometimes I would place trucks behind trees, hoping to block line of sight, only to have enemy tanks occasionally see through them and destroy them. Perhaps I could open the map in Mapmaker, change the light woods to medium, and medium to heavy?

Maybe just make more maps w/heavy woods?



junk2drive -> RE: Dense forest (6/4/2011 1:37:08 PM)

Everything on top of the map costs computer power. The more trees, the slower things get. MR has a few battles with lots of trees. He actually had to cut back on the number of trees to get reasonable loading times and FPS.

Are there any particular historical battles involving heavy woods that you would like to see in a scenario?



junk2drive -> RE: Dense forest (6/4/2011 1:39:03 PM)

HintJ the underlying terrain only effects movement and cover. The LOS is effected by the sighting rays hitting or missing tree models.



sztartur2 -> RE: Dense forest (6/4/2011 5:31:37 PM)

I believe a generic LOS calculation regarding forested areas could be more effective then calculating trees one by one. It would allow bigger maps as well.





Stridor -> RE: Dense forest (6/4/2011 11:31:39 PM)

Using the grass system it is very possible to make very dense forests without much impact on performance. In fact whilst it came late in development using the special grass system I put in place for dense forests not only will large maps play faster they will load much faster as well. Because it came so late in development a lot of the older maps were not retrofitted with this improvement, but you can easily do it yourself for your own maps.

In terms of making maps larger than 2x2, well it is not just a question of machine performance.

PCO unlike CM uses an underlying 1mx1m grid for its physics/mechanics/LOS, etc computations. This means certain game datastructures are getting very close to the 32 bit memory limit of managed direct X on 2x2km maps. I think from my own vague memory I could have made 2.4 x 2.4km maps before hitting that hard limit, but no more without doing a massive datastructure rewrite.

I hope that all helps

Regards

S.



sztartur2 -> RE: Dense forest (6/5/2011 12:56:04 AM)

I would suggest making 2X3 km if possible. And yes, every extra size helps. I find sometimes very frustrating that setup and reinforcement areas are under direct fire possiblity of the opponent in many scenarios. It sould not be like that! Most likelíy this is also a reason because of the lack of space. You should have a dense forest or some elevation protecting the startup area IMO!



sztartur2 -> RE: Dense forest (6/5/2011 1:22:22 AM)

quote:

ORIGINAL: Stridor
...

In terms of making maps larger than 2x2, well it is not just a question of machine performance.

PCO unlike CM uses an underlying 1mx1m grid for its physics/mechanics/LOS, etc computations. This means certain game datastructures are getting very close to the 32 bit memory limit of managed direct X on 2x2km maps. I think from my own vague memory I could have made 2.4 x 2.4km maps before hitting that hard limit, but no more without doing a massive datastructure rewrite.

...




Is it possible to use 2X2m grid? Then we have could have the desired size of maps.

I believe CM used 20X20m grids if I recall right.





junk2drive -> RE: Dense forest (6/5/2011 1:34:39 AM)

No but in your spare time you could learn fragmotion and reduce all the 3d models by half. Then edit all the range and distance data by half. Then you would have 2k = 4k. [;)]



sztartur2 -> RE: Dense forest (6/5/2011 8:37:26 AM)

Just trying to be constructive ... [:)]



sztartur2 -> RE: Dense forest (6/5/2011 8:38:53 AM)

Actually if range data is in XMLs I could do it. Fragmotion is not my business though.



Jacko -> RE: Dense forest (6/5/2011 9:10:19 AM)

quote:

ORIGINAL: Stridor

Using the grass system it is very possible to make very dense forests without much impact on performance. In fact whilst it came late in development using the special grass system I put in place for dense forests not only will large maps play faster they will load much faster as well. Because it came so late in development a lot of the older maps were not retrofitted with this improvement, but you can easily do it yourself for your own maps.

In terms of making maps larger than 2x2, well it is not just a question of machine performance.

PCO unlike CM uses an underlying 1mx1m grid for its physics/mechanics/LOS, etc computations. This means certain game datastructures are getting very close to the 32 bit memory limit of managed direct X on 2x2km maps. I think from my own vague memory I could have made 2.4 x 2.4km maps before hitting that hard limit, but no more without doing a massive datastructure rewrite.

I hope that all helps

Regards

S.


So larger maps are quite out of the question. That is disappointing.



sztartur2 -> RE: Dense forest (6/5/2011 11:25:57 AM)

Without bigger maps this game will have a serious disadvantage over other games on even mid term.



Jacko -> RE: Dense forest (6/5/2011 1:17:58 PM)

Yes, I agree, but I'm a bit tired of the whole big map discussion at the moment. [:)] We will see what the future brings. Or doesn't bring. 



Erik Rutins -> RE: Dense forest (6/5/2011 2:20:43 PM)


quote:

ORIGINAL: Jacko
So larger maps are quite out of the question. That is disappointing.


Stridor basically said the same thing I said earlier. There are some structural memory limitations with the current design. The discussion on what we can do to get around those and how long it will take is the development discussion I've been talking about. Nothing is out of the question until we've had that discussion. There'a a lot that we can still optimize and restructure, but we need to have a development discussion to figure out what makes sense.

Regarde,

- Erik



leridano -> RE: Dense forest (6/5/2011 3:57:16 PM)


quote:

ORIGINAL: Stridor

Using the grass system it is very possible to make very dense forests without much impact on performance.


Dense forest would be fine not only because map scenarios would gain even more inmersion but also because it would create more covered areas from enemy fire. I´m ok with the map scales at the moment but I think that dense forest terrain would be a good addition to the game.







rickier65 -> RE: Dense forest (6/6/2011 3:20:12 AM)


quote:

ORIGINAL: leridano


quote:

ORIGINAL: Stridor

Using the grass system it is very possible to make very dense forests without much impact on performance.


Dense forest would be fine not only because map scenarios would gain even more inmersion but also because it would create more covered areas from enemy fire. I´m ok with the map scales at the moment but I think that dense forest terrain would be a good addition to the game.







I've done some testing of creating dense forrests and should be able to have some maps with denser looking woods in a few weeks. I'm not able to work on them right now.

Thanks
Rick



leridano -> RE: Dense forest (8/14/2011 11:19:38 AM)

Is there something regarding this question in the near future?









rickier65 -> RE: Dense forest (8/14/2011 6:24:19 PM)


quote:

ORIGINAL: leridano

Is there something regarding this question in the near future?



I am going to work on a map over the next week and will try a few things to see how they look. In the meantime, several of the scenarios have farily ense woods right now. take a look at some of the scenarios by MR, as well as the Kobrin scenario. There are several others as well.

Thanks
rick



Ratzki -> RE: Dense forest (8/31/2011 8:27:02 PM)

Woods is such a varied terrain to try to pin down a way to represent LOS ect. properly. You can have trees that are 8 inches dia. spaced 24 inches apart with small dead braches that interlock between trees that make for a wall of darkness only a few yards from your location, as well as huge blowdown lines 10 feet+ high ringed with brush and interspersed with larger spaced trees that would block LOS just as effectively. Where I live, I have hunted in 2+ feet deep moss with larger trees (36+ inches dia.) with low growth underbrush that is just about as restricted as far as LOS would be concerned as the other two examples. I think that LOS could be based on the depth of the terrain as opposed to the tree structures themselves. Each pixiel/tile of terrain could modify chances of seeing through to the next one until after enough pixels/tiles are passed through that the chances effectively become zero. This could still have some variation built in and yet be somewhat more predictable, as well, it might make it easier on the system resources as a bonus.



rickier65 -> RE: Dense forest (8/31/2011 9:23:03 PM)


quote:

ORIGINAL: Ratzki

Woods is such a varied terrain to try to pin down a way to represent LOS ect. properly. You can have trees that are 8 inches dia. spaced 24 inches apart with small dead braches that interlock between trees that make for a wall of darkness only a few yards from your location, as well as huge blowdown lines 10 feet+ high ringed with brush and interspersed with larger spaced trees that would block LOS just as effectively. Where I live, I have hunted in 2+ feet deep moss with larger trees (36+ inches dia.) with low growth underbrush that is just about as restricted as far as LOS would be concerned as the other two examples. I think that LOS could be based on the depth of the terrain as opposed to the tree structures themselves. Each pixiel/tile of terrain could modify chances of seeing through to the next one until after enough pixels/tiles are passed through that the chances effectively become zero. This could still have some variation built in and yet be somewhat more predictable, as well, it might make it easier on the system resources as a bonus.


I agree with your comments concerning the variability of woods terrain. Particularly the way in which brush and other undergrowth impacts LOS in and through wooded areas. We can create maps for PCO that utilize some of the concepts you describe above, using the depth of the terrain as a factor in determining LOS.

Jacko1 is working on a scenario using a map that makes use of a combination of that technique together with the standard way that actual tree models block terrain. I'm anxious to see how it turns out and how the performance is.

Thanks
rick



Ratzki -> RE: Dense forest (9/1/2011 12:37:15 AM)

I am a big map fan so freeing up system resources is behind my train of thought. My other thought would be to create tree tiles that fit together like puzzle pieces or a hex grid. This way they could still be placed like any other structure but varied orientation and an assortment of tiles could create treed areas that would not look patterned. This could also reduce the number of structures on the maps and free up resources.



Mobius -> RE: Dense forest (9/1/2011 3:59:17 PM)

We don't have any good winter forest underbrush.  I don't know if it all dries out and dies but the woods in the winter maps look bare.



sztartur2 -> RE: Dense forest (9/1/2011 4:15:56 PM)

In the past years I frequently played in the woods (Airsoft is someone knows about that) In winter there are the bush like undergrowth fully naked without leaves, snowed but still there. In thick forest there should be a lot of barren bushes from 20cm till even 200cm high depending on the thickness the map designer wants. LOS is not that restricted like in summer but almost. Especially when snowed LOS is much restricted, a bit lesser if not.



Page: [1]

Valid CSS!




Forum Software © ASPPlayground.NET Advanced Edition 2.4.5 ANSI
0.71875