EDuke32 thread

[Puritan]

Thanks!

Everything works fine now.

Here is the log:

Scanning for GRP files...
Using group file 'duke3d.grp' as main group file.
Using group file 'autoload/duke3d.grp/duke3d_hrp.zip'.
Using group file 'autoload/duke3d.grp/hrp_update.zip'.
Using group file 'autoload/duke3d.grp/maphacks36a.zip'.
Using group file 'autoload/duke3d.grp/xxx_pack.zip'.

***

Are you saying that I should delete the "Maphacks36a.zip file...?

[hummertime]

Also if you go out of map in dnclip mode, the game crashes immediately.

[The Commander]

Originally Posted by hummertime

The latest release "The Dream" doesn't properly work. Some added sounds are screwed or missing.

Because you have hardely provided any details or information then all I can asume is that it is the problem with the map it's self and should be posted in the thread of the map and not here. And If you running Vista and you conplaining about some sound bug, We all know about that.

Originally Posted by hummertime

Also if you go out of map in dnclip mode, the game crashes immediately.

Dont you mean you die straight away. Using DNCLIP has never made Duke chrash before on my system. If your using DNCLIP also use GOD MODE as well.

[hummertime]

I gave the information I know. Some sounds didn't appear from the level with the new EDuke snapshot. I made a GRP file, but nothing has changed. It's not the maps fault, because I played it in the previous day, and all those extra sounds were there. But that snapshot was buggier with the sounds, I decided to put the new one in.

You are wrong with the DNCLIP. In the Atomic Edition you can clip through the vacuum without dying in most of the case. You possibly mistaken with the parallaxed bigorbit skies in the 2nd episode. They were originally built those maps (at least the AHB made ones) for space suit use.

By the way I used god mode, because it was just a search for secret places in The Dream map, not a normal play.

[Hunter_rus]

@ hummertime
You can attach logs("eduke32.log" file) so that we can have more information about your problem. If it's EDuke32 related, you can attach logs for both versions(new and old).

[hummertime]

Originally Posted by Hunter_rus

@ hummertime
You can attach logs("eduke32.log" file) so that we can have more information about your problem. If it's EDuke32 related, you can attach logs for both versions(new and old).

Could you post a link to the first DSound-fixed snapshot (which had some bugs)? The one in the "Sounds on Vista" topic is dead.

[Hunter_rus]

"Sounds on Vista" topic has the latest OpenAL snapshot and the link isn't dead BTW. But this snapshot is too outdated(based on September's source code). The OpenAL snapshots will be finally removed when the Music Pack is supported by latest EDuke32.
The first Direct Sound fixed snapshot is here(6th February). But I think it crashes as well.

It would be nice of you if you would provide some information(log, which map, which map location) which can be useful for fixing the problem for the latest EDuke32.

[hummertime]

I am talking about A.dream which was released by Piterplus recently. Here is it: http://files.duke4ever.eu/dukemaps/A.dream_install.zip

And sorry, the openAL snapshot was just temporarily down, now it works. I reinstalled it, and the map sounds work perfectly!!! Strange, but the sounds aren't working with older versions as well, including JonoF. Only with the OpenAL snapshot.

There isn't much in it actually. New sounds, a worthless user.con which doesn't screw anything and the map file. New sounds like gameovr5.voc, octapn.voc and such and these are just silent in the map.

Maybe it's a map specific thing, but strange that it works fine with one EDuke version, and screwed in the others.

Here is my logfile from a rushing gameplay. I don't think it can help but watch it.

From the OpenAL (working) snapshot

EDuke32 1.4.0 svn (Jan  4 2008 14:42:54)
Copyright (c) 1996, 2003 3D Realms Entertainment
Copyright (c) 2007 EDuke32 team
Using DEF file: A.dream.def.
addsearchpath(): Added C:/Users/nandi/Saved Games/eduke32_20060718/
Initialising Windows DirectX/GDI system interface
OS: Windows Vista (Version 6.0.6000)
Loading OPENGL32.DLL
Loading GLU32.DLL
Initialising DirectDraw...
  - Loading DDRAW.DLL
  - Enumerating display devices
    * Elsődleges képernyőillesztő
  - Creating DirectDraw object
  - Checking capabilities
Detecting video modes:
  - 1280x800 8-bit fullscreen
  - 1024x768 8-bit fullscreen
  - 848x480 8-bit fullscreen
  - 800x600 8-bit fullscreen
  - 720x576 8-bit fullscreen
  - 720x480 8-bit fullscreen
  - 640x480 8-bit fullscreen
  - 1280x800 32-bit fullscreen
  - 1280x800 16-bit fullscreen
  - 1024x768 32-bit fullscreen
  - 1024x768 16-bit fullscreen
  - 848x480 32-bit fullscreen
  - 848x480 16-bit fullscreen
  - 800x600 32-bit fullscreen
  - 800x600 16-bit fullscreen
  - 720x576 32-bit fullscreen
  - 720x576 16-bit fullscreen
  - 720x480 32-bit fullscreen
  - 720x480 16-bit fullscreen
  - 640x480 32-bit fullscreen
  - 640x480 16-bit fullscreen
  - 1024x768 8-bit windowed
  - 800x600 8-bit windowed
  - 640x480 8-bit windowed
  - 640x400 8-bit windowed
  - 512x384 8-bit windowed
  - 480x360 8-bit windowed
  - 400x300 8-bit windowed
  - 320x240 8-bit windowed
  - 320x200 8-bit windowed
  - 1024x768 32-bit windowed
  - 800x600 32-bit windowed
  - 640x480 32-bit windowed
  - 640x400 32-bit windowed
  - 512x384 32-bit windowed
  - 480x360 32-bit windowed
  - 400x300 32-bit windowed
  - 320x240 32-bit windowed
  - 320x200 32-bit windowed
Using config file 'duke3d.cfg'.
Scanning for GRP files...
Using group file 'duke3d.grp' as main group file.
Using group file 'A.dream.zip'.

Compiling: GAME.CON (151133 bytes)
Including: DEFS.CON (35992 bytes)
Including: USER.CON (43133 bytes)
Duke Nukem 3D v1.4+ style CON files detected.
GAME.CON: In state `blimphitstate':
GAME.CON:476: warning: expected a action, found a define.

Compiled code size: 16209 bytes
1903/11264 labels, 285/2048 variables
0 event definitions, 207 defined actors

Initialising DirectInput...
  - Loading DINPUT.DLL
  - Creating DirectInput object
  - Enumerating attached input devices
    * MOUSE: Egér
    * KEYBOARD: Billentyűzet
  - Creating keyboard device
  - Creating mouse device
Initialising mouse
CONTROL_Startup: Mouse Present
Loading art header...
initcache(): Initialised with 33554432 bytes
Loading palette/lookups...
mmulti: This machine's IP is 192.168.1.101
Warning: Failed including duke3d_hrp.def on line A.dream.def:1
Warning: Failed including sd_duke.def on line A.dream.def:1
Definitions file 'A.dream.def' loaded.
RTS Manager Started.
RTS file DUKE.RTS was not found
Initializing OSD...
Setting video mode 640x480 (8-bit fullscreen)
  - Creating primary surface
  - Getting back buffer
  - Allocating offscreen buffer
  - Creating palette
Checking music inits...
Checking sound inits...
Demo demo1.dmo is of an incompatible version (116).
Cache time: 1007ms
E1L1: THE DREAM
AMMO FOR PISTOL
LARGE MEDKIT: +30
AMMO FOR ROCKET LAUNCHER!
AMMO FOR PISTOL
LASER TRIPBOMB!
NIGHT VISION GOGGLES
AMMO FOR PISTOL
SMALL MEDKIT: +10
AMMO FOR SHRINKER!
AMMO FOR SHOTGUN
SHOTGUN
BLUE ACCESS CHIP REQUIRED
SMALL MEDKIT: +10
AMMO FOR CHAINGUN CANNON!
DEACTIVATED
ACCESS CHIP
SMALL MEDKIT: +10
CHAINGUN CANNON!
SMALL MEDKIT: +10
MOUSE AIMING OFF
GIVING EVERYTHING!CHEATER!
cheater mode: ON
WARNING: SE23 i<0!
AMMO FOR SHOTGUN
JETPACK ON
JETPACK OFF
AMMO FOR SHOTGUN
AMMO FOR CHAINGUN CANNON!
AMMO FOR PISTOL
Uninitialising DirectDraw...
  - Releasing palette
  - Releasing back-buffer surface
  - Releasing primary surface
  - Freeing offscreen buffer
  - Releasing DirectDraw object
  - Unloading DDRAW.DLL
Uninitialising DirectInput...
  - Releasing keyboard device
  - Releasing mouse device
  - Releasing DirectInput object
  - Unloading DINPUT.DLL

And here is the log file from the new stuff, where the sounds are not working.

EDuke32 1.4.0 svn (Feb  5 2008 21:25:35)
Copyright (c) 1996, 2003 3D Realms Entertainment
Copyright (c) 2007 EDuke32 team
Using DEF file: A.dream.def.
addsearchpath(): Added C:/Users/nandi/Saved Games/eduke32_20060718/
Initialising Windows DirectX/GDI system interface
OS: Windows Vista (Version 6.0.6000)
Loading OPENGL32.DLL
Loading GLU32.DLL
Initialising DirectDraw...
  - Loading DDRAW.DLL
  - Enumerating display devices
    * Elsődleges képernyőillesztő
  - Creating DirectDraw object
  - Checking capabilities
Detecting video modes:
  - 1280x800 8-bit fullscreen
  - 1024x768 8-bit fullscreen
  - 848x480 8-bit fullscreen
  - 800x600 8-bit fullscreen
  - 720x576 8-bit fullscreen
  - 720x480 8-bit fullscreen
  - 640x480 8-bit fullscreen
  - 1280x800 32-bit fullscreen
  - 1280x800 16-bit fullscreen
  - 1024x768 32-bit fullscreen
  - 1024x768 16-bit fullscreen
  - 848x480 32-bit fullscreen
  - 848x480 16-bit fullscreen
  - 800x600 32-bit fullscreen
  - 800x600 16-bit fullscreen
  - 720x576 32-bit fullscreen
  - 720x576 16-bit fullscreen
  - 720x480 32-bit fullscreen
  - 720x480 16-bit fullscreen
  - 640x480 32-bit fullscreen
  - 640x480 16-bit fullscreen
  - 1024x768 8-bit windowed
  - 800x600 8-bit windowed
  - 640x480 8-bit windowed
  - 640x400 8-bit windowed
  - 512x384 8-bit windowed
  - 480x360 8-bit windowed
  - 400x300 8-bit windowed
  - 320x240 8-bit windowed
  - 320x200 8-bit windowed
  - 1024x768 32-bit windowed
  - 800x600 32-bit windowed
  - 640x480 32-bit windowed
  - 640x400 32-bit windowed
  - 512x384 32-bit windowed
  - 480x360 32-bit windowed
  - 400x300 32-bit windowed
  - 320x240 32-bit windowed
  - 320x200 32-bit windowed
Using config file 'duke3d.cfg'.
Scanning for GRP files...
Using group file 'duke3d.grp' as main group file.
Using group file 'A.dream.zip'.

Compiling: GAME.CON (151133 bytes)
Including: DEFS.CON (35992 bytes)
Including: USER.CON (43133 bytes)
Duke Nukem 3D v1.4+ style CON files detected.
GAME.CON: In state `blimphitstate':
GAME.CON:476: warning: expected a action, found a define.

Compiled code size: 16209 bytes
1904/11264 labels, 285/2048 variables
0 event definitions, 207 defined actors

Initialising DirectInput...
  - Loading DINPUT.DLL
  - Creating DirectInput object
  - Enumerating attached input devices
    * MOUSE: Egér
    * KEYBOARD: Billentyűzet
  - Creating keyboard device
  - Creating mouse device
Initialising mouse
CONTROL_Startup: Mouse Present
Loading art header...
initcache(): Initialised with 33554432 bytes
Loading palette/lookups...
Warning: Failed including duke3d_hrp.def on line A.dream.def:1
Warning: Failed including sd_duke.def on line A.dream.def:1
Definitions file 'A.dream.def' loaded.
RTS Manager Started.
RTS file DUKE.RTS was not found
Initializing OSD...
Setting video mode 640x480 (8-bit fullscreen)
  - Creating primary surface
  - Getting back buffer
  - Allocating offscreen buffer
  - Creating palette
Checking music inits...
Checking sound inits...
Initializing MultiVoc...
  - Maximum voices: 8
  - Using 1280 byte mixing buffers
Initializing DirectSound...
  - Loading DSOUND.DLL
  - Creating DirectSound object
  - Creating primary buffer
  - Setting primary buffer format
      Channels:    2
      Sample rate: 22050Hz
      Sample size: 8 bits
  - Creating secondary buffer
Demo demo1.dmo is of an incompatible version (116).
Cache time: 122ms
E1L1: THE DREAM
AMMO FOR PISTOL
LARGE MEDKIT: +30
cheater mode: ON
MOUSE AIMING ON
GIVING EVERYTHING!CHEATER!
WARNING: SE23 i<0!
AMMO FOR PISTOL
JETPACK ON
JETPACK OFF
JETPACK ON
JETPACK OFF
Uninitializing MultiVoc...
DirectSound: Waiting for sound thread to exit
DirectSound: Sound thread has exited
Uninitializing DirectSound...
  - Releasing secondary buffer
  - Releasing primary buffer
  - Releasing DirectSound object
  - Unloading DSOUND.DLL
Uninitialising DirectDraw...
  - Releasing palette
  - Releasing back-buffer surface
  - Releasing primary surface
  - Freeing offscreen buffer
  - Releasing DirectDraw object
  - Unloading DDRAW.DLL
Uninitialising DirectInput...
  - Releasing keyboard device
  - Releasing mouse device
  - Releasing DirectInput object
  - Unloading DINPUT.DLL

[Hunter_rus]

I see no crash in the logs. But now I know all I need to try get your crash. I'll try.

EDIT
As for the crash. I can't get the crash. Where does it happen?
As for sound/music. Some sound and music work only for OpenAL snapshot. It's allowable for the music(because it's OGG) but all the VOC sounds could have been specified in the USER.CON file because any EDuke32 can play VOC.

[hummertime]

Fixed.

I didn't say crash. It just the new sounds that doesn't work well in the other versions. The nightmare part of the map starts after the red keycard door when you have to enter a vent. I already know if it works at that point. Under OpenAL it gives you a strange sound, under the other versions Duke says "game over".

You said it might be the special voc files to blame here? I don't really know what to do then.

[hummertime]

I think I got the problem generally with the openAL port: the shotgun! Every time I fire the shotgun, the sounds got screwed and usually replays the last sound (for example the choke of a liztroop) with slower tempo like it was underwater.

[Hunter_rus]

This snapshot was based on old EDuke32. It's too outdated and the development is discontinued. So there is no need to report about problems with it.
It has only one advantage over the latest EDuke32: support OGG for sound and music. It's expected this feature will be brought to the latest EDuke32 soon.

[hummertime]

Originally Posted by Hunter_rus

This snapshot was based on old EDuke32. It's too outdated and the development is discontinued. So there is no need to report about problems with it.
It has only one advantage over the latest EDuke32: support OGG for sound and music. It's expected this feature will be brought to the latest EDuke32 soon.

Thanks. Hopefully "the dream" problem will be solved too.

[DeeperThought]

That mess is a model sitting on the ground. Possibly an unexploded mortar, though I'm not sure -- I am sure that the mess stayed like that until I blew up something that was on the ground.

I get bizarre model distortions like this frequently, but usually it only lasts a moment because it is a short-lived sprite (e.g. a jib flying through the air). This happened only rarely on my old computer. I've heard reports from Commander and Spiker of the same thing, in two of my mods, and also in SD_Duke IIRC.

Could CON code trigger this rendering error somehow?

[TerminX]

Don't think so, unless it's some kind of freak problem that occurs with the pitch and roll controls that only shows up on certain models. Usually this kind of stuff is caused by a defective GPU, I think... I'm sure Plagman will chime in soon enough with his insights on the issue though.

[The Commander]

Originally Posted by DeeperThought

I've heard reports from Commander and Spiker of the same thing, in two of my mods, and also in SD_Duke IIRC.

Yes that is correct. Ill post picture for refrence.

Originally Posted by TerminX

Usually this kind of stuff is caused by a defective GPU,

I must disagree with that. Both DT and I both own brand new graphics cards and the chances of both of our GPU's being defective must be like 1 in a million. Must be something else to do with the renderer IMO... But I could allways be wrong

[Aherin]

Yeah I have that happen to me every now and then as well. It usually occurs when jibs spawn after explosions and sometimes projectiles. On my comp they mostly look like the texture of the particular jib or projectile that is supposed to be there, just so large that they fill up the entire screen. They seem to rotate a little sometimes as well.

They usally disappear within a second or two, or if I rotate the mouse left or right off screen and back again.

I can't say it's a massive problem, but it does happen at least once a session.

[Spiker]

Yeah like it was said before, exactly the same happens to me, And I also have a new and very good graphic card.

[gt1750]

I experienced this stuff when playing Piterplus jr.'s Moonbase episode with sd_duke. Blewing up the turrets and gibbing enemies sometimes spawned squares with the ground or wall texture.
My graphic card is not new at all ,it's a 64MB Radeon x600.

[The Commander]

Maybe it has something to do with ATI cards? Because mine is also an ATI. Not sure about DT's or Spiker's though.

[DeeperThought]

Originally Posted by The Commander

Maybe it has something to do with ATI cards? Because mine is also an ATI. Not sure about DT's or Spiker's though.

Nope. My card is a GeForce 8500 GT.

I'm not sure if gt1750 is talking about the same thing, since "spawning squares" sounds a little different.

[Spiker]

NVIDIA GeForce 8400 GS

[KillerBudgie]

Mine is an NVIDIA GeForce 7600 GS. And I sometimes get the same problem you lot have been getting.

[gt1750]

Originally Posted by DeeperThought

Nope. My card is a GeForce 8500 GT.

I'm not sure if gt1750 is talking about the same thing, since "spawning squares" sounds a little different.

Well not exactly squares. It wasn't as wild as on your screenshot, but it was there. I'll try to grab a screen when I play Duke again.

[DeeperThought]

Originally Posted by gt1750

Well not exactly squares. It wasn't as wild as on your screenshot, but it was there. I'll try to grab a screen when I play Duke again.

It think I know what you mean, actually. There is a model used in SD_Duke (or at least it used to be), "plane.md3", which is a simple plane used for smoke and blood effects. I've been using it too, and it definitely exhibits the problem sometimes.

[Hunter_rus]

I haven't noticed that glitch before.
I was trying hard to get that glitch with SD_Duke and WMD mods. After placing lots of monsters and shooting everywhere I noticed several times some huge debris models(I thing it was "scrap?.md3"). I happened only several times. Even though it was hard to get that, I could see that limits of sprites weren't reached, lots monsters and shooting just increase a chance of that glitch. It happens so rarely for me that I can't investigate it more. It looks like a random memory corruption but it would rather crash due dynamically allocated arrays.

Is it that glitch or it's something another?
How to increase a chance of glitch?

[DeeperThought]

Originally Posted by Hunter_rus

How to increase a chance of glitch?

I noticed it happening a lot in redrock.map using my latest (unreleased) version of DNWMD. That screenshot that I posted was from that map, and it happens a lot in that area near those boxes. I can't imagine why the map would have anything to do with it.

I am trying to make it happen less, even though I don't know what's causing it. I am replacing the plane.md3 model with a sprite. Also, when I change pitch and roll on models, instead of just adding to the number, I will check the number and if it is > 2047, I subtract 2048 first (in case overflow was causing a glitch).

[dood!]

I don't know what causes it, but I think I know how to make it stop:

set r_animsmoothing to 0.

[Plagman]

It looks like animation is smoothing from one still model frame to another unrelated one. Do you have smoothing disabled for all your frame definitions for this model?

[DeeperThought]

Originally Posted by Plagman

It looks like animation is smoothing from one still model frame to another unrelated one. Do you have smoothing disabled for all your frame definitions for this model?

None of the affected models use, or have ever used, the smoothduration def command.

Is there something else required to disable smoothing?

Also, why would smoothing cause the models to become giant?

[Plagman]

You can disable it altogether on your client using the command dood! mentioned. Can you reproduce those issues with smoothing disabled? The answer to that question will help me track the problem down.

[DeeperThought]

I'm having trouble reproducing it at all right now, because I eliminated the model that it happened with most often. Maybe someone else who is using an older (i.e. released) version of one of the affected mods can try this?

Open the console and type: r_animsmoothing 0

You can tell if smoothing is disabled by going to a mirror (such as the one in E1L1) and switching the jetpack on and off. Without smoothing, Duke's legs move instantly.

[ADM]

Thread too large. Part 2 available here:
http://forums.3drealms.com/vb/showthread.php?t=30513