[TerminX]

I just haven't committed the fix that initializes appicon to NULL and adds the null ptr check there yet.

[Danukem]

Does this mean that the crash will be less frequent after the next revision?

[TerminX]

That particular crash is Linux/SDL only.

[gt1750]

Some Polymer shots to spice up this thread:



I really like the shrinker/expander glowing effect. Places with normalmapped textures yet no dynamic lights can look awesome with it.

[TerminX]

Glad you like it... the shrinker and expander glowing was one of my 30 second "hey, I bet that would be cool" sort of things that turned out great.

[Fox]

Hey, should HRP use an ultra-realistic water texture?



The water that is used right now is based on Duke's water, okay, but it's far from being realistic.

And I see the water more like the skyboxes, which are not made to be an exactly copy of Duke, it takes a realistic look.

Imagine if we could make Derelict (or Crytical Mass ending) looks like this

[Piano Man]

Ilovefoxes, on Dec 1 2009, 11:37 AM, said:

Imagine if we could make Derelict (or Crytical Mass ending) looks like this

The only way to do that would be to build the maps using another game engine other than the Duke engine....

Technically I would love to see someone do it... Imagine the Crysis engine running Duke Nukem 3D and pumping up the textures....


DROOLS!!!!!

[Fox]

Since Derelict and Crytical Mass have very well defined water textures and skyboxes right after them, it is not impossible to add that effect.

Maybe that horizon thing would need an effect that would infinite stretch an sector visual floor and texture? That would be easy to apply using maphacks, including Derelict underwater.

[Piterplus]

TX, on Nov 30 2009, 11:19 PM, said:

Glad you like it... the shrinker and expander glowing was one of my 30 second "hey, I bet that would be cool" sort of things that turned out great.

Indeed good idea,
It was nice surprise when I saw it first time.

[Sangman]

Ilovefoxes, on Dec 1 2009, 04:37 AM, said:

Imagine if we could make Derelict (or Crytical Mass ending) looks like this

With good textures and models, talent and A LOT of patience, something somewhat like that should be possible I think.

[supergoofy]

I just compiled the latest svn build:

eduke32_win32_20091203-1551.zip
removed

[edit]
with some mods there are still con compilation errors when you run eduke32. For example, NR:IotM (Naferia's Reign: Invasion of the Dark Mistress) mod appears to have errors in enhance.con, but I don't see where. Probably a false error report.
The last SVN build that worked ok was 1543. All the SVN builds from 1544 and newer (up to 1551) report errors in enhance.con

Also some mods like Hellduke TC have problems with weapons selection (if you try to change weapon you will see only the fist or the kick) when you run them with newer eduke32 builds up to 1543. Builds from 1544 to 1551 report con errors like in NR:IotM mod.

[Plagman]

If you don't want to run an unknown binary downloaded from a dubious file sharing website, there's a real build there along with source snapshot and changelog:
http://dukeworld.duk.../20091203-1551/

<pre>-----------------------------------------------------------------
-------
 r1551 | plagman | 2009-12-02 23:37:34 -0800 (Wed, 02 Dec 2009) | 1 line
 
 Don't rely on repeating texture samplers to flip polymer sprites and request clamped textures for sprite textures. Fixes bug 2869362.
 ------------------------------------------------------------------------
 r1550 | plagman | 2009-12-02 22:21:39 -0800 (Wed, 02 Dec 2009) | 1 line
 
 Make polymer impervious to corrupt maps (for the tenth time..). Fixes bug 2869996.
 ------------------------------------------------------------------------
 r1549 | plagman | 2009-12-02 21:38:09 -0800 (Wed, 02 Dec 2009) | 1 line
 
 Don't save the OpenGL driver max texture size to the config file and remove some remaining instrumentation from rev 1546.
 ------------------------------------------------------------------------
 r1548 | plagman | 2009-12-02 20:41:09 -0800 (Wed, 02 Dec 2009) | 1 line
 
 Don't overflow scantoasc[] with high scancodes. This makes the 'T' key work in the console on my laptop.
 ------------------------------------------------------------------------
 r1547 | plagman | 2009-12-02 19:11:41 -0800 (Wed, 02 Dec 2009) | 1 line
 
 Don't change the CWD to the eduke32.exe directory on Windows. EDuke32 now searches the CWD for data files on all platforms.
 ------------------------------------------------------------------------
 r1546 | plagman | 2009-12-02 19:07:04 -0800 (Wed, 02 Dec 2009) | 1 line
 
 Fixes a compressed texture cache bug where the background menu tile would sometime get replaced with another tile.
 ------------------------------------------------------------------------
 r1545 | terminx | 2009-11-19 19:54:40 -0800 (Thu, 19 Nov 2009) | 2 lines
 
 some CON compilation fixes
 
 ------------------------------------------------------------------------
 r1544 | terminx | 2009-11-17 17:17:56 -0800 (Tue, 17 Nov 2009) | 2 lines
 
 Second gtk startup window patch from bioman, other random fixes I can't be bothered to remember
 
 ------------------------------------------------------------------------
 r1543 | terminx | 2009-11-15 19:57:31 -0800 (Sun, 15 Nov 2009) | 2 lines
 
 fix glaring oversight in new multiplayer random ID generation scheme
 
 ------------------------------------------------------------------------
 r1542 | helixhorned | 2009-11-15 12:01:33 -0800 (Sun, 15 Nov 2009) | 1 line
 
 fixes wall drawing (bug introduced in r1540)
 ------------------------------------------------------------------------
 r1541 | terminx | 2009-11-15 01:10:50 -0800 (Sun, 15 Nov 2009) | 2 lines
 
 lower memory use for tilefromtexture by way of lzf compression, untested
 
 ------------------------------------------------------------------------
 r1540 | terminx | 2009-11-13 18:30:47 -0800 (Fri, 13 Nov 2009) | 2 lines
 
 A bunch of crap I meant to commit days ago... gtk startup window patch from bioman, nedmalloc REPLACE_SYSTEM_ALLOCATOR fix (related code uses #ifdef REPLACE_SYSTEM_ALLOCATOR so defining it to 0 was still satisfying the condition), slightly lower CPU usage in Mapster 2D mode, fix an ancient bug causing the player to not make vent footstep sounds on sprite based vents, fix map save state cvars, add "skill" cvar to console, fix saving the game while CON_SHOWVIEW is in use
 
 ------------------------------------------------------------------------
 </pre>

[supergoofy]

@Plagman, I tried to help. Now the file is removed. The problems/bugs though remain.

[Lord Misfit]

SuperGoofy's right about NR not wanting to compile in ENHANCE.CON with recent snapshots. Additionally, I just downloaded your binary of 1551, and it still gives me the same compiling errors:

EDuke32 2.0.0devel 20091106
addsearchpath(): Added E:/Other Stuff/eduke32/
OS: Windows XP (5.1.2600) Service Pack 2
Large page support available
Loading opengl32.dll
Loading glu32.dll
Initializing DirectDraw...
Using config file 'eduke32.cfg'.
Scanning for GRP files...
Using group file 'duke3d.grp' as main group file.
Compiling: GAME.CON (1445610 bytes)
Resizing code buffer to 1048576*4 bytes
Including: DEFS.CON (69451 bytes)
Including: USER.CON (181078 bytes)
USER.CON:1631: warning: truncating quote text to 127 characters.
Including: ENHANCE.CON (15829 bytes)
ENHANCE.CON: At top level:
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
ENHANCE.CON:20: error: expected a keyword but found `'.
fatal error: too many errors: Aborted
Found 1 warning(s), 64 error(s).
UDP networking uninitialized successfully.
Uninitializing DirectDraw...

So, yeah...

And I know ENHANCE.CON hasn't been changed in ages on my end, so I honestly don't think I'm directly responsible for these compiling errors.

[supergoofy]

@Plagman, what do you mean by unknown binary? I compiled it. What's your point? Are you accusing me of something?

[Plagman]

I'm not; I'm just paranoid about running binaries I don't know hosted on places I don't know, and I assume a lot of people are like me in that regard. I don't really trust third-party hosting services like mediafire.

[Hendricks266]

I keep more frequently compiled binaries of EDuke32 here:

http://hendricks266.duke4.net/eduke32/

I use 7-Zip compression for smaller filesize, and the script I use incorporates more files.

Plagman/TX's synthesis script should be edited to include these files in the binary .zip, like mine: *.exe *.m32 *.sample *.txt *.hlp *.cfg *.map names.h
Because of this, it is currently missing: duke3d.def.sample enhance.con.sample SEHELP.HLP STHELP.HLP m32script_ex.map
Fortunately, these are mostly inconsequential except for m32script_ex.map.

[Gambini]

I asked this to TX sometime ago, but maybe it will make more noise here.

When Polymer wasn´t released, Sectoreffectors 50 and 49 (polymer light sources) worked as pure white lights when the xyzvel values was set to 0 0 0.

It doesn´t work anymore, but think it should be added again because:

1- White colour is 255 255 255, so 0 0 0 would be an useless black light (that doesnt make any sense).

2- It would make the mappers life way easier since inserting and placing a sectoreffector and setting its lotag/hitag values doesn´t take more than five seconds and can be done in 3d mode. while setting its velocity values requires to go to 2d mode, find the sprite among all the other ones surrounding it, setting its lotag/hitag values and then pressing F8 and setting its three velocity values.

3- It´s usually also the most used light, and therefore, the one that requires a quick application the most. Just as the most used sprite in a map, that happens to be the one appearing everytime one presses S.

4- Some of the first maps/mods featuring Polymer ligths used this method and now those lights aren´t working at all (well, yeah they still kill the frames per second). Examples: Fusion TC Redux, Dukeplus Community Build Project.

5- It´s easier to teach someone how to apply a lightsource the way i´m asking than the way it is now.

6- Some of the first maps/mods featuring Polymer ligths used this method and now those lights aren´t working at all (well, yeah they still kill the frames per second). Examples: Fusion TC Redux, Dukeplus Community Build Project.

[Stabs]

not having too many problems, the only thing i'd really like is to be able to edit the colours in 3d mode.

right shift and insert (good thing about this too is dosnt screw up the sprite placement for sectors in SoS situations) for mass copying in 2d mode works great and tab & enter for 3d mode, combined with using mouse, shift, and wheel to move them its pretty easy to do lots of lighting really quick.

[Gambini]

I do know about those shortcuts but the way it was before made things even easier. if it´s back, nobody suffers, both methods work. If it´s not back some maps dont work properly on polymer (as the mentioned examples) and there is only one method to work with.

[Stabs]

spose it wouldnt matter much as i always use a 250 / 250 / 250,

the only benefit of having a 0 0 0 black would be to make really custom looking light, or perhaps it could be used to counter light leaking from adjacent rooms.. maybe

[Gambini]

We´re not understanding each other man.

A black light does not exist because it casts 0 unities of red, 0 if green and 0 of blue. Neither it can absorb light from other sources.

A 250 250 250 ¨grey¨ light casts 250 unities of each primary colour, or what is the same: Is 97% effective regarding its reach.

So, Basicly a light with 128 128 128 and a hitag of 1024, and another one with 255 255 255 and a hitag of 512 would be identicall.


BTW i have a point despite the way you like to work your lights. There are about a millon maps done for duke and maybe 20/30 going to be done someday. We have to think about the maps that could be updated for Polymer more or at least the same we think in the maps that are going to be done for Polymer. Surely in a project like yours playing with light colours and powers is a pleasure. But in the maps i have done, and i want to update for polymer, setting every lightsource´s value manually is a pain in the butt not worth of the time.

[Todd]

DanM, on Dec 13 2009, 05:49 PM, said:

spose it wouldnt matter much as i always use a 250 / 250 / 250,

the only benefit of having a 0 0 0 black would be to make really custom looking light, or perhaps it could be used to counter light leaking from adjacent rooms.. maybe

There's also lights turned off by scripts--000 the natural setting to turn the light off.


A solution would be mapster setting the velocity values to 250 by default when you insert a sector effector. That of course would fail if you insert a sprite and then turn it into a sector effector.

[Gambini]

Ops that about the lights turned off drops by the ground my argument. I didn´t know it was possible.

Unfortunatelly, making every SE having the vel values on 255 by default would mess things more than the solution, there are some uses for those values apart from polymer lights, like some Dukeplus effects. I guess the best solution would be adding some command for editing vel values in 3d mode.

[Danukem]

Todd, on Dec 13 2009, 03:37 PM, said:

There's also lights turned off by scripts--000 the natural setting to turn the light off.

I disagree. If you turn off a light that way, then you have to save the RGB values somewhere in case it gets turned on again (if it's not going to get turned on again, then the best solution is to delete the SE, which saves resources). Rather than save the RGB values, just set the light radius to 0 and save the original radius (easier to save 1 number than 3).

[Plagman]

It seems like what you really want is the editor to set 255/255/255 as the default color for newly spawned light and/or easier ways to change the color in 2D/3D mode. I bet a trivial BUILDscript could allow you to do both, but helixhorned probably knows better.

[Gambini]

Something like that. The thing is that before, when it worked with 0 0 0, it was great. I wonder if you changed it for some particular reason?

[Plagman]

I didn't write/touch the code that binds these SEs to Polymer lights, so I don't think I had anything to do with that change (as far as Polymer itself is concerned, RGB of 0/0/0 has always meant no light). As convenient as 0/0/0 meaning 255/255/255 was, it doesn't really make sense and sounds like a hackish fix. I think Mapster32 should really just assign 255/255/255 when you change the lotag of an SE to 49 and 50 that has x/y/zvel of 0/0/0.

[Plagman]

Hendricks266: I updated synthesis.sh to package the missing files you mentioned; thanks for the heads-up. Having two sources of development builds is a little redundant and seems like it could possible incur some multiplayer/compatibility confusion, though.

[supergoofy]

but these files they are not needed