Current CVS errors

[kibora]

Why am i wearing a shield, instead of a bow in the last cvs?

[Schmurk]

Why am i wearing a shield, instead of a bow in the last cvs?

I don't know, works fine for me.

 

@Florian: is this a new patch or was it also in the patch that was on mantis?

[Florian]

@Florian: is this a new patch or was it also in the patch that was on mantis?

Patch for the patch, so it's a new one, the diff is against the latest cvs.

[0ctane]

Undefined symbols:
 "ec::GlowEffect::GlowEffect(ec::EyeCandy*, bool*, ec::Vec3*, ec::GlowEffect::GlowType, unsigned short)", referenced from:
  _ec_create_glow_level_up_def in eye_candy_wrapper.o
  _ec_create_glow_level_up_att in eye_candy_wrapper.o
  _ec_create_glow_level_up_oa in eye_candy_wrapper.o
  _ec_create_glow_poison in eye_candy_wrapper.o
  _ec_create_glow_harm in eye_candy_wrapper.o
  _ec_create_glow_remote_heal in eye_candy_wrapper.o
ld: symbol(s) not found
collect2: ld returned 1 exit status

Anyone else seeing this today? I searched through the code but can only find GlowEffect in eye_candy_wrapper.cpp. Am I missing a file somewhere?

[Florian]

Am I missing a file somewhere?

Yes, two new files, eye_candy/effect_glow.[ch]

Edited March 19, 2008 by Florian

[0ctane]

My bad. XCode did not properly update its database.

[ago]

Edit^2: Im using a compiled exe in src_dir and expect the exe to pickup all files required from data_dir

 

current CVS crashes in coreanimation.cpp:129

 

void CalCoreAnimation::scale(float factor)

{

// loop through all core track

std::list<CalCoreTrack *>::iterator iteratorCoreTrack;

for(iteratorCoreTrack = m_listCoreTrack.begin(); iteratorCoreTrack != m_listCoreTrack.end(); ++iteratorCoreTrack) <=== crashes here

{

(*iteratorCoreTrack)->scale(factor);

}

}

 

It works well without CACHE_ANIMATIONS

 

Edit: it crashes because the Object is NULL because in cal_animation_cache.hpp:37 anim_ptr.get() returns NULL because loadCoreAnimation fails

 

CalCoreAnimationPtr anim_ptr = CalLoader::loadCoreAnimation(fileName, NULL);

CalCoreAnimation_Scale(anim_ptr.get(), scale);

 

I looked into it a bit more (my girlfriend will teach me what is more important later). The code is missing the correct data_dir conversion. Cal3d is loading the file without taking data_dir into account. cal_animation_cache must use file_io functions to translate filenames (eg. ./animations/player_walk.caf) to filenames in the data_dir.

 

> Cal3D.dll!CalLoader::loadCoreAnimation(const std::basic_string<char,std::char_traits<char>,std::allocator<char> > & strFilename="./animations/player_walk.caf", CalCoreSkeleton * skel=0x00000000) Line 81 C++

elc.exe!CalAnimationCache::loadAnimation(const std::basic_string<char,std::char_traits<char>,std::allocator<char> > & fileName="./animations/player_walk.caf", float scale=1.0000000) Line 37 + 0x12 bytes C++

elc.exe!CalCoreModel_ELLoadCoreAnimation(CalCoreModel * self=0x01ccfe18, const char * strFilename=0x0012e1bc, float scale=1.0000000) Line 274 + 0x3a bytes C++

elc.exe!cal_load_anim(actor_types * act=0x01288c20, const char * str=0x0012e880, const char * sound=0x007bfb19, const char * sound_scale=0x007bfb19, int duration=-1) Line 229 + 0x23 bytes C

Edited March 20, 2008 by ago

[Schmurk]

Ok, I should have fixed it. Tell me if you still have the problem...

[Florian]

Patch

 

https://developer.berlios.de/patch/index.ph...p;group_id=1256

 

for bad reference on ongoing harvesting effect which causes a client crash.

[ago]

Ok, I should have fixed it. Tell me if you still have the problem...

 

Works fine now.

 

Memory usage is down from 180MB to 127MB.

[Entropy]

Patch

 

https://developer.berlios.de/patch/index.ph...p;group_id=1256

 

for bad reference on ongoing harvesting effect which causes a client crash.

 

Don't you have CVS access?

[0ctane]

Patch

https://developer.berlios.de/patch/index.ph...p;group_id=1256

for bad reference on ongoing harvesting effect which causes a client crash.

Don't you have CVS access?

lol. Nope, he does not. I recommend him.

[Entropy]

Ok, make a berlios account then let me know the user name.

[ago]

Judging from who submitted the patches I'd think the berlios name is superfloh

[Alberich]

build without NEW_ACTOR_ANIMATION fails

patch:

--- cal_types.h	26 Oct 2007 21:30:11 -0000	1.10
+++ cal_types.h	30 Mar 2008 13:05:58 -0000
@@ -15,9 +15,9 @@
	int anim_index;
	cal_animation_type kind;
	float duration;
-#ifdef	NEW_ACTOR_ANIMATION
+#if defined(NEW_ACTOR_ANIMATION) || defined(NEW_ACTOR_MOVEMENT)
	float duration_scale;
-#endif
+#endif // NEW_ACTOR_ANIMATION | NEW_ACTOR_MOVEMENT
#ifdef NEW_SOUND
	int sound;
	float sound_scale;

[Florian]

build without NEW_ACTOR_ANIMATION fails

already reported, but since NEW_ACTOR_ANIMATION is in the standard build config you're not supposed to build without it

[Alberich]

Them how about removing it when nobody is supposed to use the option? along with all the other defaults...

 

sorry for not checking if already reported

[Florian]

Them how about removing it when nobody is supposed to use the option? along with all the other defaults...

 

#define purges happen from time to time

 

I commited your patch anyway

[Learner]

Them how about removing it when nobody is supposed to use the option? along with all the other defaults...

 

#define purges happen from time to time

 

I commited your patch anyway

Yes, don't purge it too quickly tought. Keeping the option for a while after it becomes standard gives us a backup plan incase problems show up as it hits the wider range of users.

[rauch]

Please, whoever did it, could you make the fonts borders allover the gui a bit more decent? the black border is fine on the inv quickslots, but it looks just plain ugly in my opinion everywhere else.

 

Maybe a bit lighter color (dark gray?) could be used?

[ago]

I agree. At least in the statbars the text border does conceal parts of the statbar which was visible before.

[Ermabwed]

I agree as well. The stats in HUD/clock/material/ethereal points/food level/emu/experience can be seen very nicely without black borders. One thing which might be nice would be an alpha background for those parts when one uses the transparent side bar...

 

I have two other problems though:

* "show reflections" option looks like I see only reflections and nothing else

* when I turn off the "use animation program" option, I can't turn it back on, always get "Not using vertex program for actor animation." and according to start up information, I do have the GL_ARB_vertex_program extension.

 

I'm compiling with default options minus new_sound.

[Schmurk]

There are still some issues with the reflections, it's on work...

[Vegar]

There are still some issues with the reflections, it's on work...

Is this the issue you're talking about? (taken at beam)

 

 

Edit: I've found that reverting the following patch, which is extracted from the commit "Fixed some bugs and changed some values in the sky code.", cvs date "1 Apr 2008 21:20:32 -0000", fixes the above issue. Those that are having issues with latest CVS can patch -R this patch for now.

http://developer.berlios.de/patch/download.php?id=2429

Edited April 4, 2008 by Vegar

[Schmurk]

There are still some issues with the reflections, it's on work...

Is this the issue you're talking about? (taken at beam)

 

 

Edit: I've found that reverting the following patch, which is extracted from the commit "Fixed some bugs and changed some values in the sky code.", cvs date "1 Apr 2008 21:20:32 -0000", fixes the above issue. Those that are having issues with latest CVS can patch -R this patch for now.

http://developer.berlios.de/patch/download.php?id=2429

No this is not a bug. This is because you're compiling with the skybox and because I've changed the way the near and far clip plane work. Put a value of 1.0 for near plane and 100.0 for far plane and it should be ok.

 

EDIT: Anyway, I suppose it's the problem you have because it looks like a clipping plane problem but I don't see any link with your patch...

EDIT 2: Ok you are right, the commit I've done in this part of the code have broken the reflection if you don't use the frame buffer. I'll correct that asap...

Edited April 4, 2008 by Schmurk