Demon
Register | User Profile | Member List | F.A.Q | Privacy Policy | New Blog | Search Forums | Forums Home
Doomworld Forums : Powered by vBulletin version 2.2.5 Doomworld Forums > Classic Doom > Doom Editing > Visplane Error
 
Author
All times are GMT. The time now is 22:32. Post New Thread    Post A Reply
999cop
Constrained Owing to Lot


Posts: 3173
Registered: 12-00
I've received an error such as this
"R_DrawPlanes: visplane overflow (130)" in the middle of the game when I was running my wad under vanilla Doom2 port. Does anyone have a clue of what it is all about?
Old Post 10-13-01 05:57 #
999cop is offline Profile || Blog || PM || Email || Homepage || Search || Add Buddy IP || Edit/Delete || Quote
Espi
Möh.


Posts: 663
Registered: 06-00
Your map is too complex. Cut down some detail.
Old Post 10-13-01 10:05 #
Espi is offline Profile || Blog || PM || Email || Homepage || Search || Add Buddy IP || Edit/Delete || Quote
Enjay
ASK ME ABOUT FOOTBALL / STEAM / DEAD CELEBRITIES / THE BLAIR WITCH PROJECT


Posts: 6093
Registered: 12-00
Or run it on a source port.

Pretty much all source ports have eliminated this problem.
Old Post 10-13-01 12:36 #
Enjay is offline Profile || Blog || PM || Email || Homepage || Search || Add Buddy IP || Edit/Delete || Quote
ReX
Senior Member


Posts: 1732
Registered: 05-00
Does anyone have a clue of what it is all about?
This means that there are too many visible planes (or sidedefs) in view. Basic DooM2's engine cannot handle the display of so many sidedefs at one time, so it crashes. Do as Nigel suggests, and use a source port. If it's your own map, and you want it to run on vanilla DooM, do as Espi suggests and reduce the level of detail.
Old Post 10-13-01 14:18 #
ReX is offline Profile || Blog || PM || Email || Homepage || Search || Add Buddy IP || Edit/Delete || Quote
Jayextee
Darth Jrrr


Posts: 321
Registered: 10-00
Apparently, judging from what I have read in the textfile that comes with BSP, it's not about linedefs, but SEGS/SSECTORS (One of the two). I gather it's something about how areas (Sectors) have to be broken into triangles to be drawn, or something, and it's many of those in view at the same time that causes VPO. Linedefs just cause HOM.

I have no idea, so don't quote me 100%, but BSP can reduce the chance of it happening. Read the docs.
Old Post 10-13-01 15:46 #
Jayextee is offline Profile || Blog || PM || Email || Homepage || Search || Add Buddy IP || Edit/Delete || Quote
ReX
Senior Member


Posts: 1732
Registered: 05-00
it's not about linedefs, but SEGS/SSECTORS (One of the two)
The thing is, if you create complex floor or ceiling sectors without assigning them a height parameter (i.e., they are flat on the floor or ceiling), then you won't get a VPO error. These are the linedefs that you refer to. However, if you assign the sectors height parameters, so that sidedefs are created, then the VPO error is likely to occur. It's sidedefs, not linedefs, that cause the problem. Likewise, it's sidedefs without textures that cause HOM.

Thanks for the info on BSP. I didn't realize that the node-builder breaks up sectors into triangular segments.
Old Post 10-14-01 15:26 #
ReX is offline Profile || Blog || PM || Email || Homepage || Search || Add Buddy IP || Edit/Delete || Quote
deep
as in Deepsea


Posts: 1066
Registered: 09-01
A Linedef always -has- to have at least 1 Sidedef which -has- to point to a "sector" no matter what the "height" is. IOW, a "sector" does NOT define the geometry of a level. It's something a lot of "tutorials" get confused on. For example, a "sector" can be all over the map, that's what "packing sectors" is all about.

Nodes are nodes are nodes and BSP is not any different. What was thought to work actually works only in certain small levels. Basically not a useful option:)

Nodes are not triangles - they define a bounding area which in turn define convex areas.

SSECTORS/SEGS are related. SSECTORS tell how many SEGS there are in the area they define. SEGS define the actual "line" geometry: point-to-point, angle and linedef relation. Nodes point to SSECTORS.

If you take an area with enough "pedestals" of the same height to cause a vis overflow (a simple way to verify this stuff), all you have to do is combine all the "sectors" defining the pedestals and it no longer will have the problems. The number of linedefs and sidedefs stay exactly the same. The sidedef sector references have been consolidated by "merging/packing".

DOOM code does a SSECTOR->SEGS lookup in NODES to define the area.

The end
Old Post 10-14-01 19:29 #
deep is offline Profile || Blog || PM || Email || Homepage || Search || Add Buddy IP || Edit/Delete || Quote
Jayextee
Darth Jrrr


Posts: 321
Registered: 10-00
Yes, I have found that BSP is only *sometimes* useful at reducing the chances of VPO. But, it worked for my E1 contest map.
Old Post 10-16-01 18:29 #
Jayextee is offline Profile || Blog || PM || Email || Homepage || Search || Add Buddy IP || Edit/Delete || Quote
Quasar
Moderator


Posts: 5150
Registered: 08-00
Visplanes have absolutely nothing to do with sidedefs. 1 visplane is generated for every unique flat/lightlevel/height combination of floor areas that are currently in view.

In other words, if you make a checkerboard with 500 squares but all have the same flat, lightlevel, and height, then only 1 visplane is generated for the entire area. But change every other tile into a different flat or raise the sectors up 1 unit, and suddenly the engine needs to generate a ton of visplanes.

If you don't believe me, then load up Lee Killough's visplane overflow demo wad that comes with BSP. The problem with HOM is from the 2s sidedefs. When the game exits, however, the problem is with visplanes, and those visplanes are generated when you hit the switch in the middle that lowers the light level of every other square in the checkerboard.

Remember -- 2S HOM = sidedef problem
visplane overflow = sector problem
Old Post 10-19-01 19:52 #
Quasar is offline Profile || Blog || PM || Email || Homepage || Search || Add Buddy IP || Edit/Delete || Quote
deep
as in Deepsea


Posts: 1066
Registered: 09-01
Visplanes have absolutely nothing to do with sidedefs.
Sort of correct - indirectly they do since sidedefs define sectors:) Did you read what I wrote above? That's actually a simpler way to demonstrate the problem without hom artifacts involved.

Carefully read what Lee wrote. He was "guessing" how the engine works, I wrote exactly how it works since the code came out since then.

Now, the interesting thing is that if you rebuild the Conduits.wad (the one you referenced) with DeePbsp (using default settings from DeePsea) pressing the switch won't cause a vis overflow. Yet, technically, using the "standards" given it shouldn't work:) So you see, it's a tricky little devil.

You can examine both types of nodes using the DeePsea node viewer - on the Edit menu / Node Viewer.
Old Post 10-19-01 22:55 #
deep is offline Profile || Blog || PM || Email || Homepage || Search || Add Buddy IP || Edit/Delete || Quote
ReX
Senior Member


Posts: 1732
Registered: 05-00
In other words, if you make a checkerboard with 500 squares but all have the same flat, lightlevel, and height, then only 1 visplane is generated for the entire area. But change every other tile into a different flat or raise the sectors up 1 unit, and suddenly the engine needs to generate a ton of visplanes.
My mistake about the nomenclature (linedef vs. sidedef) -- as Jack correctly pointed out every linedef has 2 sidedefs, even if they are not textured. I guess I was using the terminology (textured sidedef vs. sidedef) interchangeably, but my meaning was as stated above, and I described the reason for the problem in a non-technical manner. What I said was exactly what's in the quote above -- "The thing is, if you create complex floor or ceiling sectors without assigning them a height parameter (i.e., they are flat on the floor or ceiling), then you won't get a VPO error".
Old Post 10-20-01 17:06 #
ReX is offline Profile || Blog || PM || Email || Homepage || Search || Add Buddy IP || Edit/Delete || Quote
deep
as in Deepsea


Posts: 1066
Registered: 09-01
"The thing is, if you create complex floor or ceiling sectors without assigning them a height parameter (i.e., they are flat on the floor or ceiling), then you won't get a VPO error".
Not really (see Quasar's statement).

If you give -flat- sectors different light levels it will cause a VPO. Similarly, the flat graphics have to be the same. So it's not a "height" issue, but a "difference" issue. That's why packing sectors will get rid of the problem.

The nodes/vis type discussions seem to come and go. There is a lot of incorrect data written about this and other Doom editing facets because people never check out the facts:)
Old Post 10-20-01 19:08 #
deep is offline Profile || Blog || PM || Email || Homepage || Search || Add Buddy IP || Edit/Delete || Quote
All times are GMT. The time now is 22:32. Post New Thread    Post A Reply
 
Doomworld Forums : Powered by vBulletin version 2.2.5 Doomworld Forums > Classic Doom > Doom Editing > Visplane Error

Show Printable Version | Email this Page | Subscribe to this Thread

 

Forum Rules:
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts 		HTML code is OFF
vB code is ON
Smilies are OFF
[IMG] code is ON
 

< Contact Us - Doomworld >

Powered by: vBulletin Version 2.2.5
Copyright ©2000, 2001, Jelsoft Enterprises Limited.

Message Board Statistics