Author Topic: Zathras 2.3 and Guardian of Light  (Read 6740 times)

0 Members and 1 Guest are viewing this topic.

Offline wesp5

  • 29
Zathras 2.3 and Guardian of Light
After finishing Dark Children I was testing my second beloved First Ones campaign, Guardian of Light by Darkmaster, when I got the following error, both in the missions against the Dilgar and the Humans:

Null vec3d in vec3d normalize.
Trace out of vecmat.cpp and find offending code.


[attachment deleted by a basterd]

 

Offline wesp5

  • 29
Re: Zathras 2.3 and Guardian of Light
Update: I continued with the release version engine and got stuck in the Dilgar mission, because it is much too hard and in a completely unrealistic way too. Even after setting the enemy fighter AI to rookie, the friendly fighter AI to elite and removing the EA missiles, which the Dilgar cruisers were firing, all Vorlon fighters die within minutes, myself included. This makes no sense whatsoever because the EA could have never beaten the Dilgar that way and they did in the show! What could be the reason?

Also the whole setup still looks very strange, because all ships are lighted red, although the nearby moon looks normal and the biggest nebula is blue. How can I change this? And last not least, why is the release version engine slower than the debug one?

 

Offline wesp5

  • 29
Re: Zathras 2.3 and Guardian of Light
Nobody knows something about the "Null vec3d in vec3d normalize" warning? I'm pretty sure I didn't get it with Zathras 2.0 and the older 3.6.12 builds...


Update: Searched the forum for it and found something similar in another thread, but without a solution. But this thread contained a interesting post from IPAndrews:

"At this late stage in the game it's better to have a single TBP 3.5 Final download with instructions to completely delete and uninstall any previous versions before installation."

Sounds like he allows a TBP 3.5 Final there, so why did we only get TBP 3.4b(eta)? Maybe we could at least fix very old annoying bugs like the Brakiri weapon warning like this, while keeping changes to models and stuff in Zathras.
« Last Edit: November 28, 2012, 02:47:35 am by wesp5 »

 

Offline FUBAR-BDHR

  • Self-Propelled Trouble Magnet
  • Moderator
  • 212
  • Master Drunk
    • 165th Beer Drinking Hell Raisers
Re: Zathras 2.3 and Guardian of Light
The problem with the null vec3d warnings is that they are just a symptom of the real problem and that problem could be anywhere.  The only way to track them down is to be running from the debugger or possibly (I've never got it to work) with the PDB files to get a stack trace.
No-one ever listens to Zathras. Quite mad, they say. It is good that Zathras does not mind. He's even grown to like it. Oh yes. -Zathras

 

Offline wesp5

  • 29
Re: Zathras 2.3 and Guardian of Light
The problem with the null vec3d warnings is that they are just a symptom of the real problem and that problem could be anywhere.

Yeah, I figured as much reading threads about them in the other FS2 boards. But I think they may be connected to the Dilgar fighter model, because as the mission was too hard for me, I decreased their number and the number of error messages went down as well! Maybe some of you model experts could take a look at them, I bet they are not used often.

The other time the warning appears in the mission against the EA, removing all fighters does not help, it pops up the moment the Vorlon dreadnaught destroys an EA Omega destroyer. And I guess if the latter was the reason, I would have gotten this warning in several other missions too, like Severed Dreams...

Anyway, why is the release build stuttering compared to the debug build? Is it framerate locked or forcing vertical sync or something?

 

Offline niffiwan

  • 211
  • Eluder Class
Re: Zathras 2.3 and Guardian of Light
actually - that stuttering sounds like a possible issue with the head animations, it doesn't show up in debug for an obscure reason, and it should have been fixed in 3.6.14.  Anyway, can you check and confirm if the stuttering only occurs when the head animations are playing?  (I'm assuming that TBP uses head ANIs?)
Creating a fs2_open.log | Red Alert Bug = Hex Edit | MediaVPs 2014: Bigger HUD gauges | 32bit libs for 64bit Ubuntu
----
Debian Packages (testing/unstable): Freespace2 | wxLauncher
----
m|m: I think I'm suffering from Stockholm syndrome. Bmpman is starting to make sense and it's actually written reasonably well...

 

Offline wesp5

  • 29
Re: Zathras 2.3 and Guardian of Light
Anyway, can you check and confirm if the stuttering only occurs when the head animations are playing?

No, at least on my system here I can't see a difference. But then strangely enough there is less stuttering on this old dualcore than on my new quadcore at home! Of course release builds only, debug builds run great. Any launcher setups that may make a difference between release and debug?

Also speaking of the launcher, I thought I used the same, but here it shows a cool Vorlon fighter image, at home only the normal FS2 one. Could this be the reason? Another more launch orientated issue, when starting in a window or at low resolution, sometimes the loading pictures get cut off or displayed wrong. Is there a way to make them size up correctly?

 

Offline niffiwan

  • 211
  • Eluder Class
Re: Zathras 2.3 and Guardian of Light
The processor affinity settings *may* have something to do with the stuttering, have a read of these posts for some background & suggestions.

http://www.hard-light.net/forums/index.php?topic=82620.msg1649312#msg1649312
http://www.hard-light.net/forums/index.php?topic=55990.0

I'm not sure what's causing the launcher issue.  Presumably you've got TBP launcher, and possible another one for normal FSO, I think they both should launcher the game in the same way.

I'm also not sure about window/low resolution thing, I haven't seen that sort of problem before, maybe a screenshot will help?
Creating a fs2_open.log | Red Alert Bug = Hex Edit | MediaVPs 2014: Bigger HUD gauges | 32bit libs for 64bit Ubuntu
----
Debian Packages (testing/unstable): Freespace2 | wxLauncher
----
m|m: I think I'm suffering from Stockholm syndrome. Bmpman is starting to make sense and it's actually written reasonably well...

 

Offline wesp5

  • 29
Re: Zathras 2.3 and Guardian of Light
Update: I think I found the reason for the Dilgar mission warnings: The Dilgar capital ship has no docking bay! So I get the warning when their fighters leave from there, but no warning when they come in from hyperspace. This is a thing the next Zathras should fix, I will repost this in the Zathras thread!

 

Offline wesp5

  • 29
Re: Zathras 2.3 and Guardian of Light
The processor affinity settings *may* have something to do with the stuttering, have a read of these posts for some background & suggestions.

If this is still true, those threads were quite old, there is an issue with the 3.6.15 release build because it obviously is handling the multicores much worse than the debug build. As you seem to know more about this, could you report it to the people actually improving the builds?

 

Offline niffiwan

  • 211
  • Eluder Class
Re: Zathras 2.3 and Guardian of Light
The Diaspora related thread isn't very old, about 1 month? :)  But, that's about the only recent issue I can recall seeing regarding stuttering - if you ignore the head ANI issue I referred to earlier... given that I know a lot of people don't have this issue I don't know if there's enough evidence yet to warrant a bug report about it.  Yet. 

Anyway, we should go through the normal support "dance" first, have you updated you video card drivers?  Can you provide a fs2_open.log file?
Creating a fs2_open.log | Red Alert Bug = Hex Edit | MediaVPs 2014: Bigger HUD gauges | 32bit libs for 64bit Ubuntu
----
Debian Packages (testing/unstable): Freespace2 | wxLauncher
----
m|m: I think I'm suffering from Stockholm syndrome. Bmpman is starting to make sense and it's actually written reasonably well...

 

Offline wesp5

  • 29
Re: Zathras 2.3 and Guardian of Light
Can you provide a fs2_open.log file?

I can, when I am at home, because the problem isn't evident here. But it can't really be my system alone, because the debug build runs absolutely smooth and only when I switch to the release build I notice the stuttering. Which should rather be the other way around if at all... 

 

Offline wesp5

  • 29
Re: Zathras 2.3 and Guardian of Light
Update: It must be some kind of Windows issue, because if it happens, I can simply remove the stuttering by Alt-Tabbing out of TBP and then back in. Still weird, that it doesn't happen with the debug builds though.

I have some other issues with this mission now. Both, Kosh's ship and another Vorlon transport don't get their orders after ships they fired on are destroyed.

Update: I fixed the problem above by using is-destroyed-delays instead of is-event-true-delays :)!

« Last Edit: November 29, 2012, 04:44:48 pm by wesp5 »

 

Offline wesp5

  • 29
Re: Zathras 2.3 and Guardian of Light
Okay, I'll move this here because the other thread had been derailed a bit ;). FUBAR wrote about the null vector issue of the EA mission:

"Mission could be tweaked by tuning the ships a bit but the bug can happen with any ship that happens to launch a fighter on a downward path."

What do you mean by "tuning" or was this supposed to be "turning"? One thing strange about this mission is that the enemy ships of the second wave appear upside down compared to the first wave or the player. Could this be the reason for the downward path issue or would we need to change all coordinates?

 

Offline Slasher

  • 29
Re: Zathras 2.3 and Guardian of Light
From what I gather, if the ships launch on a vector of (x, -y, z) when their initial coordinates are (x, y, z) then the game will crash?  In that case, adjust the launch ship so its fighters don't start out flying on a vector normal to the x/z-axis. 

This is weird because I've never encountered or even heard of this bug before.

 

Offline FUBAR-BDHR

  • Self-Propelled Trouble Magnet
  • Moderator
  • 212
  • Master Drunk
    • 165th Beer Drinking Hell Raisers
Re: Zathras 2.3 and Guardian of Light
Right now the ships have a pitch of exactly 90 degrees.  If that was changed to 89.9 0r 90.1 the issue might not happen.  This isn't a real solution as the ship turning for any reason and ending up facing back down at the time a fighter is launched would again cause the issue.  So it would just become random.  The best solution here is a code one so the problem doesn't happen at all. 

Also looking through that mission there are a number of other issues from invalid team assignments to the fighter waves launching on top of each other.  At times there are 12 fighters arriving on 4 paths resulting in 3 ships stacking on top of each other and beating the heck out of each other during launch. 
No-one ever listens to Zathras. Quite mad, they say. It is good that Zathras does not mind. He's even grown to like it. Oh yes. -Zathras

 

Offline FUBAR-BDHR

  • Self-Propelled Trouble Magnet
  • Moderator
  • 212
  • Master Drunk
    • 165th Beer Drinking Hell Raisers
Re: Zathras 2.3 and Guardian of Light
From what I gather, if the ships launch on a vector of (x, -y, z) when their initial coordinates are (x, y, z) then the game will crash?  In that case, adjust the launch ship so its fighters don't start out flying on a vector normal to the x/z-axis. 

This is weird because I've never encountered or even heard of this bug before.

Not quite but your on the right track.  It's the uvec (or up vector) that seems to be the issue.  The engine tries to calculate a uvec for the launch path which it normally succeeds in doing.  In the case of the first 2 paths pointing straight down the uvec it calculates happens to be 0,0,0 which is well a null vector.  At least I think that is what is happening.  My vector math skill are about 25 years out of date.   

Anyways it is quite amazing that this hasn't been caught before now as it is easily reproducible with even retail FS2. 
No-one ever listens to Zathras. Quite mad, they say. It is good that Zathras does not mind. He's even grown to like it. Oh yes. -Zathras

 

Offline wesp5

  • 29
Re: Zathras 2.3 and Guardian of Light
Anyways it is quite amazing that this hasn't been caught before now as it is easily reproducible with even retail FS2.

So is there are chance that this will be fixed in the next build? Otherwise I will try turning the ships and reducing the wing size as well. In that mission it doesn't really matter, becaue the Vorlons will win anyway ;). One last question though, where can I fin info about the number of fighters that can be launched from a ship? Because I still don't know if the Nova's or the Hyperion's can only launch 4 fighters at the same time. Maybe it should be added to the tech data description of each ship...

 

Offline FUBAR-BDHR

  • Self-Propelled Trouble Magnet
  • Moderator
  • 212
  • Master Drunk
    • 165th Beer Drinking Hell Raisers
Re: Zathras 2.3 and Guardian of Light
There are 2 ways.  Open the .pof and look at the number of bay paths or turn on show bay paths in FRED.  Unfortunately the FRED option is currently broke. 
No-one ever listens to Zathras. Quite mad, they say. It is good that Zathras does not mind. He's even grown to like it. Oh yes. -Zathras

 

Offline wesp5

  • 29
Re: Zathras 2.3 and Guardian of Light
Open the .pof and look at the number of bay paths or turn on show bay paths in FRED.  Unfortunately the FRED option is currently broke.

Doesn't sound easy for an amateur like me ;). So which one has only 4 bay paths, the Nova or the Hyperion?

Update: I'm just guessing that it must be the Hyperion. After all the Omegas seem to have 6 fighter wings...
« Last Edit: December 06, 2012, 04:36:22 pm by wesp5 »