EDuke32 2.0 and Polymer! "talk about the wonders of EDuke32 and the new renderer"
#3528 Posted 16 March 2013 - 02:24 PM
#3529 Posted 16 March 2013 - 04:46 PM
This post has been edited by Trooper Mick: 16 March 2013 - 04:48 PM
#3530 Posted 16 March 2013 - 07:10 PM
Trooper Mick, on 16 March 2013 - 04:46 PM, said:
You'll still need to know what the different structs do and how they interact/relate with each other. It will just be easier to manipulate and should be more flexible.
#3531 Posted 17 March 2013 - 12:39 AM
#3533 Posted 17 March 2013 - 05:31 AM
http://wiki.eduke32....ember_functions
The specific ways of accessing them might change, but you'll have to know what things are still. All of the things you do in CON for the most part have analogues in the actual game code/engine, there's just some really terrible and somewhat ASM type ways of manipulating them. A trig function which should be one line takes 10 in CON (for example).
#3534 Posted 17 March 2013 - 09:08 AM
Trooper Dan, on 16 March 2013 - 02:24 PM, said:
Yes, transition to Lua will proceed gradually. You'll be able to load one root CON + modules like now, but in addition to this, multiple root Lua files which in turn may "require" other Lua modules. CON will probably never be removed completely because of stuff like defining game resources or gamestartup, for which there would be no point in duplicating on the Lua side.
#3535 Posted 17 March 2013 - 11:32 AM
Trooper Mick, on 17 March 2013 - 12:39 AM, said:
Nah, it's easier to the second languange, then the next is even easier... Start with one but be flexible, Lua is as good any, CON if your strictly planning on only ever using Eduke. C or C++ some say but really it's all the same principals. Learn the principals, syntax is not the end of the world. Just start and learn some basics. I'd tell you to write hello world problems in ten different languages before you pick one.
#3536 Posted 17 March 2013 - 04:51 PM
#3537 Posted 19 March 2013 - 08:54 PM
#3538 Posted 20 March 2013 - 10:58 PM
http://wolf3dredux.sourceforge.net/
The Wolfenstein 3-D Redux can up-scale the data from Wolfenstein 3-D and thus it greatly improves the visual outcome. Can this be implemented in EDuke32 ?
#3539 Posted 20 March 2013 - 11:05 PM
#3540 Posted 24 March 2013 - 08:45 AM
#3542 Posted 24 March 2013 - 01:22 PM
This post has been edited by Fox: 24 March 2013 - 01:23 PM
#3543 Posted 24 March 2013 - 01:29 PM
#3544 Posted 24 March 2013 - 01:32 PM
This post has been edited by Skulldog: 24 March 2013 - 01:34 PM
#3545 Posted 24 March 2013 - 02:02 PM
JonoF uploaded the source to his data installer to his forums but they are closed, so I made a mirror: http://hendricks266....datainst-r8.zip
A completely separate Mac data installer is on GitHub: https://github.com/jonof/datainst-osx
#3546 Posted 27 March 2013 - 10:43 AM
#3547 Posted 28 March 2013 - 10:57 AM
I have a map where there is a huge spotlight for a water caustics effect on a big area. If the light radius encompasses other areas from the map that are completely isolated from the area the light is on (separated by a closed door), it won't get 100% occluded inside the isolated areas. Sprites inside those areas will be affected by the light, but sectors will not.
This post has been edited by Diaz: 28 March 2013 - 11:02 AM
#3548 Posted 28 March 2013 - 12:05 PM
James, on 27 March 2013 - 10:43 AM, said:
Yes, this is a thing for the future.
#3549 Posted 29 March 2013 - 02:05 PM
http://hendricks266....est_20130329.7z
#3550 Posted 29 March 2013 - 02:32 PM
Diaz, on 28 March 2013 - 10:57 AM, said:
I have a map where there is a huge spotlight for a water caustics effect on a big area. If the light radius encompasses other areas from the map that are completely isolated from the area the light is on (separated by a closed door), it won't get 100% occluded inside the isolated areas. Sprites inside those areas will be affected by the light, but sectors will not.
Hi Diaz,
You're right, sprites get affected regardless of which sector they're in. When we keep better track of which sector(s) sprites are into, this will be solved. I added it to this:
http://wiki.eduke32....er_Deficiencies
#3552 Posted 29 March 2013 - 10:10 PM
Do you have any suggestions that I should check ?
By the way, many thanks for this.
#3553 Posted 30 March 2013 - 12:11 AM
#3554 Posted 30 March 2013 - 04:46 AM
supergoofy, on 29 March 2013 - 10:10 PM, said:
Nothing big comes to mind. Could you check if 32-bit and 64-bit savegames are labeled as such correctly in both builds? Also, one remaining item on the to-do list is to make readarrayfromfile/writearraytofile consistent between architectures (i.e. save in one; load in the other); at present I don't think it works correctly. Networking compatibility between architectures is something we'll also have to look out for.
supergoofy, on 29 March 2013 - 10:10 PM, said:
No problem. We've tried a couple times to move synthesis to a different toolchain that could build 64-bit binaries but each time it ends up making the 32-bit builds crash. It might be because of nedmalloc.
One way to help would be to help diagnose these crashes. This is the most recent build with "i686-w64-mingw32-gcc-4.7.2-release-linux64 + native NASM". Try it with and without nedmalloc in all three renderers.
#3555 Posted 30 March 2013 - 05:05 AM
I will check later with the build you told me (http://dukeworld.duk.../20121230-3350/)
[edit]
I have tested eduke32_win32_20121230-3350.zip
When nedmalloc.dll is present, then eduke32.exe doesn't even start (no crash message of eduke32.log)
When I deleted nedmalloc.dll, then eduke32 started ok.
This post has been edited by supergoofy: 30 March 2013 - 05:15 AM
#3556 Posted 03 April 2013 - 09:32 AM
This post has been edited by M210: 03 April 2013 - 09:33 AM

Help
Duke4.net
DNF #1
Duke 3D #1


