[Paul B]

Good day eh,

I was in the process of mapping in my new map when I came across a few problems with SE 31.

I have updated to the latest EDUKE r3723 and I still experience the same problems. I'm not sure what release build this started occuring in.

1) Problem with SE 31 not aligning itself with the height of the Sector Effector. (It is always off by a bit).

2) Duke can't crouch walk through the Sector Effector 31 sector. It's like it becomes a blockable sector. Yet if Duke stands up, he can walk through it no problem.

Please see the attached map I made to demonstrate the problems.

Any advice or help is appreciated.
Thanks,
Paul

Attached File(s)

•  SE31 Bug.zip (502bytes)
  Number of downloads: 210

[Helixhorned]

This is the same behavior as in the original Duke3D. Please check bug reports against it the next time. It's only a matter of starting Duke under DOSBox.

[Paul B]

Helixhorned, on 04 May 2013 - 04:27 AM, said:

This is the same behavior as in the original Duke3D. Please check bug reports against it the next time. It's only a matter of starting Duke under DOSBox.

I'm not trying to waste anyone's time. I have two instances in my map where I thought SE 31 works properly and another spot where it doesn't. (One area just wasn't as obvious). I've used this SE 31 before in other places in previous maps and I didn't recall having this problem which was why I posted a reply. I think my memory is failing me. If this is an original bug in the game is this something not worth fixing?

I'm using TROR exclusively in my user maps so unfortunately I have stopped checking things in the original DOS version.

Anyway, I am definitely am not trying to nit pick, it just looks weird when I build my trap with a raising floor and it doesn't lift the sprite to the proper desired height.

Thanks for your time and next time before I post a bug I will make sure to test it in build first, which totally slipped my mind.

[Helixhorned]

Paul B, on 04 May 2013 - 09:05 AM, said:

If this is an original bug in the game is this something not worth fixing?

We aim at keeping the original behavior for all sector effects because changing them might break existing maps that rely on it. There might be instances where it's possible to change things in a backward-compatible way, but the height difference is not one of them. The collision detection glitch is somewhat annoying, but I consider clipping code to be frozen as well.

[LAW]

I get something about GetRawInputData info is missing in my User32.dll. Just don't tell me there is no work-around other than upgrading my Win 2000

I am really pissed off, as I wanted to play DNF mod.


Woops. My problem is with the Eduke, not Mapster, so you can delete this post.

[DavoX]

When you copy a sector into another with ALT GR , it ask for creating an inner loop, but if you keep copying more than 2 times it won't ask you and it will make the walls of the first copies white.

[Micky C]

I'm doing a section of a map which involves subways, however the bulk of the subway is actually built up of nested sectors. When the player gets on the subway, he slides of 95% of the time because of this and it's very frustrating. Is this supposed to happen? The floor relativity bits are set everywhere.

[Paul B]

Micky C, on 17 May 2013 - 02:11 AM, said:

I'm doing a section of a map which involves subways, however the bulk of the subway is actually built up of nested sectors. When the player gets on the subway, he slides of 95% of the time because of this and it's very frustrating. Is this supposed to happen? The floor relativity bits are set everywhere.

Hi Micky, according to my previous experiences with Subways you can't have more than one sector for a subway track. The track itself has to be all one continuous sector for each subway system. The reason why Duke can't stay on the subway car is probably because of a sector clipping issue with an adjacent sector when the subway car passes through it. Just make sure the subway track is all one big open sector and you should be fine.

Let me know if this helps you?

*** Edit *** As for the individual subway cars, the sectors for each subway car must contain a unique sector hi tag per car . Make sure each one has the appropriate hi-tag for the subway cars child sectors as well. Seems to me that the sectors have to be created within the perimeter of the subway car and not attached to the outside walls of the subway car. Don't try and split/connect the outside walls of the subway car with another sector. In any case, when it happens in game check your 2D overhead view. You might see what offending wall is being stretched and forcing Duke off the subway car. I'd also just make sure that your angle for the S.E is set properly to indicate the front of the subway car. Also I think all child sectors within the subway car have to be balanced or equal parts. Similar to my Brooklyn Bridge Aircraft Subway vehicle. Unfortunately, you can't really get too fancy with that feature.

Good Luck, i want to play your map!

[DavoX]

He probably meant that the subway CAR is made out of various sectors. I'm sure gambini is the man to ask for this question.

[Kyanos]

I've got sprites not showing up in the map when I add them, as in this pic, I added three oceansprites, the third wasn't visible in 3d mode. I immediately switched to 2D mode, saw that it wasn't on the grid and had an odd flag, 54. I first caught this using rev. 3780, it persists in 3786. I back checked my next newest version 3737 where the behaviour was proper.

[Paul B]

Drek, on 17 May 2013 - 04:38 PM, said:

I've got sprites not showing up in the map when I add them, as in this pic, I added three oceansprites, the third wasn't visible in 3d mode. I immediately switched to 2D mode, saw that it wasn't on the grid and had an odd flag, 54. I first caught this using rev. 3780, it persists in 3786. I back checked my next newest version 3737 where the behaviour was proper.

Capture.PNG

Are you sure they aren't showing up as one sided sprites? I noticed after doing an update as well some of the sprites I had placed became one sided in the opposite direction I wanted. So I had to go back and re-adjust the side I wanted them to be one sided on.

Are you using TROR in the map?

[Kyanos]

No TROR. I didn't see any other changes like one sidedness. I came to say it just happened again in rev. 3737. I'm on my laptop with an ati video card. This time the sprite flag was 50.


Added, next morning with some sleep in me.

So yes the sprites are becoming one sided, wall aligned, and not landing where they should on a grid intersection. Here is a 3d mode pic; and the log;
Log:

Spoiler

Mapster32 2.0.0devel r3786
Compiled May 17 2013 03:46:00
Using C:/cbp9/ for game data
Remap 8 key to 4B
Remap 9 key to 4D
Remap A key to 50
Remap B key to 48
Remap 2B key to 9C
Remap C7 key to 46
Windows 7 (build 6.1.7600)
Automatic fullscreen size determination failed! 1366 768 -> 683 384.
Initializing DirectDraw...
Loading "m32help.hlp"
Loading NAMES.H
Loaded 740 names.
Initializing DirectInput...
- Enumerating attached game controllers
- No game controllers found
Uninitializing DirectInput...
Initialized 24.0M cache
Enabling desktop composition...
Setting video mode 1024x600 (8-bit windowed)
Loading sounds from "GAME.CON"
warning: duplicate sound #261, overwriting
warning: duplicate sound #323, overwriting
Loaded 374 sound definitions.
Loaded V8 map /CBP9.map successfully
Disabling desktop composition...
Setting video mode 1024x600 (32-bit windowed)
polymost_glreset()
Enabling ATI FBO color attachment workaround.
OpenGL Information:
Version: 3.2.9704 Compatibility Profile Context
Vendor: ATI Technologies Inc.
Renderer: AMD M880G with ATI Mobility Radeon HD 4250
polymost_glreset()
Opened "textures" as cache file
gltexinvalidateall()
gltexinvalidateall()
Enabling desktop composition...
Setting video mode 1024x600 (8-bit windowed)
polymost_glreset()
Disabling desktop composition...
Setting video mode 1024x600 (32-bit windowed)
polymost_glreset()
Enabling ATI FBO color attachment workaround.
polymost_glreset()
Opened "textures" as cache file
gltexinvalidateall()
Board autosaved to AUTOSAVE.MAP
Saved screenshot to capt0000.png
polymost_glreset()
polymost_glreset()
Uninitializing DirectDraw...

Another note: My mouse cursor is off a little compared to "current tile"

[Mark]

Off topic but... You know how sometimes you see a face or other pattern in a seemingly random texture. I spotted this right away when I saw your pic. A one toothed man face.

Attached thumbnail(s)

• 

[Helixhorned]

Drek, on 17 May 2013 - 04:56 PM, said:

My mouse cursor is off a little compared to "current tile"

This is a known issue with Polymost's mouse picking.

For the sprite insertion bug, a test case would be best since otherwise I have no starting point about how to debug it.

[Helixhorned]

DavoX, on 14 May 2013 - 09:48 AM, said:

When you copy a sector into another with ALT GR , it ask for creating an inner loop, but if you keep copying more than 2 times it won't ask you and it will make the walls of the first copies white.

It won't attempt to make an inner loop if any of its walls intersect with existing ones. Or are you saying that on the third time, Mapster32 changes geometry that doesn't belong to the highlighted sectors? Anyhow, a test case would be great.

[DavoX]

Lets say you make a platform out of a sector, then you copy that platform somewhere else it will make the inner loop properly. But if you keep making copies of the platform more than once it will leave the previous pasted platforms with their outer walls white and these won't be visible from the main sector.


[/URL]

[Helixhorned]

DavoX, on 19 May 2013 - 11:49 AM, said:

Lets say you make a platform out of a sector, then you copy that platform somewhere else it will make the inner loop properly. But if you keep making copies of the platform more than once it will leave the previous pasted platforms with their outer walls white and these won't be visible from the main sector.

Ah, I see. Automatic inner loop insertion only happens when you clear the sector highlight with [RAlt], not when you duplicate the highlighted portion with [Insert]. Note how the short "height indicators" on the white walls point inward. That square sector is independent from the large rectangular one. When you highlight it and unhighlight it right afterwards, Mapster32 will create the corresponding walls in the outer sector and set the nextwall links ("make red walls"). This even works with multiple loops: the "Attached new inner loop to sector" will be printed as many times as inner loops that were created.

[Kyanos]

Helixhorned, on 19 May 2013 - 11:12 AM, said:

This is a known issue with Polymost's mouse picking.

For the sprite insertion bug, a test case would be best since otherwise I have no starting point about how to debug it.

The map in question is a group project, there was a section that had some extreme SOS bridge which just got erased. No more problems with sprite insertion. Voila!

[DavoX]

Helixhorned, on 20 May 2013 - 02:21 AM, said:

Ah, I see. Automatic inner loop insertion only happens when you clear the sector highlight with [RAlt], not when you duplicate the highlighted portion with [Insert]. Note how the short "height indicators" on the white walls point inward. That square sector is independent from the large rectangular one. When you highlight it and unhighlight it right afterwards, Mapster32 will create the corresponding walls in the outer sector and set the nextwall links ("make red walls"). This even works with multiple loops: the "Attached new inner loop to sector" will be printed as many times as inner loops that were created.

I've had a problem... let's say that instead of one platform I copy and past 2 at the same time, when I paste them both for the first time it will ask me to create an inner loop and indeed it will create it, but when you look at it from 3d mode or the game, strange artifacts show up, and some times the platform dissapears in front of you, as if it didn't belong to the big sector.

[Helixhorned]

DavoX, on 20 May 2013 - 11:27 AM, said:

I've had a problem... let's say that instead of one platform I copy and past 2 at the same time, when I paste them both for the first time it will ask me to create an inner loop and indeed it will create it, but when you look at it from 3d mode or the game, strange artifacts show up, and some times the platform dissapears in front of you, as if it didn't belong to the big sector.

The square white ones in your screenshot do not belong to the outer rectangular sector, as I have said.

Otherwise, by "test case", here I mean a map along with clear instructions how to reproduce a particular issue. I mean... it's easier than looking for the right words to describe the problem[1], so unless it takes more than a couple of minutes, I suggest bug reporters of map effects / Mapster32 issues provide them up-front. Everyone saves time by doing that.

[1] Yes, you will still have to describe it, but then some sloppiness may be excused .

[DavoX]

Any particular reason in 3d mode you point to a wall and it shows it's stats in white but on another wall it shows them in yellow font? also they both walls have the same looking brightness but in numbers they have different shades.

[Helixhorned]

DavoX, on 21 May 2013 - 10:42 AM, said:

Any particular reason in 3d mode you point to a wall and it shows it's stats in white but on another wall it shows them in yellow font?

Yes, I added this as a visual aid when you're aiming at the bottom part of a bottom-swapped wall (cstat bit 2). In this case, the bottom part facing you uses some members of the wall on the other side:

r3308 said:

Mapster32: in 3D mode's tile info, highlight members of swapped bottom walls.

- If aiming at a swapped bottom wall, display "Wall <wallnum> -> <otherwallnum>"
- highlight pic, shade, pal, cstat in yellow then

[DavoX]

So I guess that's related to setting the shade on the outside of the wall and it getting the same shade on the other side :S how to make it look normal again? (I.E. Different shades on both sides)

[Helixhorned]

DavoX, on 22 May 2013 - 09:42 AM, said:

So I guess that's related to setting the shade on the outside of the wall and it getting the same shade on the other side :S

That's how bottom-wall swapping works: some members, including shade, are set on the other side but displayed on the facing one.

Quote

how to make it look normal again? (I.E. Different shades on both sides)

Disable the bottom swapping on the offending wall (key [2] in 3D mode) if your intended texturing allows that.

[Mark]

I think we should gather up a bunch of money to buy Helix a train pass so he can travel the country and visit us mappers for one-on-one lessons on all the many,many features of Mapster we don't know about.

[DavoX]

I knew about the swapping but didn't know it made a change on the other side of the wall, it's true that you learn something new every day

[Paul B]

I don't have a bug report, but a question. I thought I could be tricky by using a TROR extended sector to use a different first wall then the parent sector. Would it be possible to include a new feature to TROR that would allow the parent sector to have a different "first wall" then the extended TROR sector ceiling? This would allow mappers the ability to slope floors one direction while sloping ceilings in the extended TROR layer differently.

From what I've seen, the extended sector retains the same first wall angle as the parent sector, therefore TROR only slopes in the same direction as the original parent sector. I am pretty sure this was how it was intended to be, but I was hoping we could have more flexibility in this area.

[DavoX]

I've found a bug... as you can see in the attached picture the player is inside a really small sector. When I have the textures activated in 2d mode (backspace) and zoom all the way in, the editor crashes to desktop. Doesn't happen when there is no texture 2d mode.

Attached thumbnail(s)

• 

[Helixhorned]

DavoX, on 28 May 2013 - 12:37 PM, said:

I've found a bug... as you can see in the attached picture the player is inside a really small sector. When I have the textures activated in 2d mode (backspace) and zoom all the way in, the editor crashes to desktop. Doesn't happen when there is no texture 2d mode.

Could you post the test map? Also, an eduke32_or_mapster.crashlog can be useful to compare with similar open bugs (one by LeoD). Are you running the x86 or x86_64 build, i.e. using the ASM rendering routines or their C replacements?

[blizzart]

Does anybody know, what this brown colored sprite in 2D mode means?

See here:


Sprite no.: #9809