Character and Weapon Systems

[beni]

Models:

Hey guys. Your stuff looks awesome. I especially like the design of the rockets. But I think you overdid it a bit. Since rockets will be quite small on the screen we will definitely not use that many details.
My suggestion is: Keep the detailed models because they're just cool and can be used in the game anyways (in the shop we might need closer views on weapons).
Now what you do is following: Make a less detailed model and put all the details on the texture. This makes no difference on the screen in the game, but it will make a difference in performance.
No mistakes in your models either: Keep up the good work.

[nicolasc]

No mistakes in your models either: Keep up the good work.

I know why I do the modeling by hand... (i.e inserting & removing faces, edges...)

@ high poly models: sigh... I knew that this was coming. Looking into it.
model, model, model, and then try to a question...
-- derived from wild wild west

cheers
nico

[patrick]

Just checked out our work and it looks very good!

But I got a segfault one, I wanted to shoot a secondary weapon (yet I have updated the data repos). I'm afraid I didn't get a backtrace.

Do the sounds on your system also seem a little noisy? Some of the weapons are sometimes very loud and then very calm...

[nicolasc]

I experienced the same (sound, and segfault)...

I have no explanation for the sound, but for the segfault, take a look at this

Code: Select all

Program received signal SIGSEGV, Segmentation fault.
SwarmProjectile::tick (this=0xa6b76600, time=0.0144990133)
    at ./lib/math/vector.h:91
91        inline float len() const { return sqrt (x*x + y*y + z*z); }
(gdb)
#0  SwarmProjectile::tick (this=0xa6b76600, time=0.0144990133)
    at ./lib/math/vector.h:91
#1  0x081453d3 in GameWorld::tick (this=0x8685538, entityList=@0xbfb03b60,
    dt=0.0144990133) at story_entities/game_world.cc:392
#2  0x08145909 in GameWorld::tick (this=0x8685538)
    at story_entities/game_world.cc:431
#3  0x081450e8 in GameWorld::run (this=0x8685538)
    at story_entities/game_world.cc:315
#4  0x08145395 in GameWorld::start (this=0x8685538)
    at story_entities/game_world.cc:249
#5  0x0814282b in Campaign::run (this=0x867d8e8)
    at story_entities/campaign.cc:154
#6  0x08142524 in Campaign::start (this=0x867d8e8)
    at story_entities/campaign.cc:85
#7  0x082c119c in GameLoader::start (this=0x867d5f8)
    at loading/game_loader.cc:163
#8  0x081691f4 in Orxonox::start (this=0x83fa438) at orxonox.cc:417
#9  0x0816b171 in startOrxonox (argc=1, argv=0xbfb03f64, name=@0xbfb03e60,
    port=-1) at orxonox.cc:573
#10 0x0816cb2c in main (argc=1, argv=0xbfb03f64) at orxonox.cc:481
(gdb)

maybe can make some sense out if this... I first thought that it was a division by zero... but add an if-clause didn't solve the problem (or it still occured)

cheers
nico

[patrick]

Didn't look at the code in detail, but looks as if there are serveral vectors.len() division that could be zero.
Try commenting the regions out if you are not sure. Also, if the length is zero, the vector must be zero too, check the vectors for this first.

Just a blind tip...

[beni]

trail.cc does not compile on my system. The compiler has obviously a problem with the variablenames rad0 and rad1. If I change it to radzero and radone it compiles fine.
Just wanted to say, that committed stuff should always compile, which it probably did on your systems anyway.

[Marc]

It worked on all our systems including the tardis system. Any chance your compiler is outdated or your project file has some special settings?

[nicolasc]

it worked on all our systems including the tardis system. any chance your compiler is outdated or your project file has some special settings?

Works like a charm here (Gentoo unstable)... your problem might be gcc-4.1 related

I managed to get the guiding system working on the swarm missile. It's plain simple, but it works.


cheers
nico

[beni]

it worked on all our systems including the tardis system. any chance your compiler is outdated or your project file has some special settings?

works like a charm here (Gentoo unstable)... your problem might be gcc-4.1 related

As I said: The problem is probably on my side only. I figure this out quickly.

I managed to get the guiding system working on the swarm missile. It's plain simple, but it works.

I tested it and it looks awsome

[patrick]

hey weapon-brothers!

Just tested your branch and I just wanted to tell you that you did...
...awesome work!!!

The trails already look very nice, looking forward to your presentation guys!

[nicolasc]

Out! Out!! OUT!!!
There are already too many people messing in that branch...

just kiddin' Glad you like it!

[Marc]

I'm looking forward to the presentation as well because I'm not sure that there will be anything left to show. Due to the "out team" additions to our branch the last 2-3 days, first some minor problems popped up and now the whole movement that was implemented by Michel is that badly broken that you can't control at all.

I would really like to see people first ask for the use of stuff instead of just changing it on other peoples branches! Thanks all.


EDIT: The whole thing compiles again and some further additions (new explosions, trail working and some other minor stuff) will be shown tomorrow at the presentation.

[beni]

Hey sorry, I probably left a not so stable version committed. Still I didn't change anything and with 2 or 3 comments the code would be like before. Not really a big deal.

But you're right. Should have at least told ya. Should there be any problem left I'm glad to help you to fix your branch again.

[hofzge]

Dear all: I must remind you that segfaults and the like belong into the bugfix forum.
I know that this reminder is kind of late, but I had little time lately to correct the forum until today

[Marc]

I know. But it is no segfault with the trunk ... so I'm not sure if it makes that much sense as nobody else is getting it.