[Release] GalNavy additions (was Adck's Behemoths Redux)

Discussion and information relevant to creating special missions, new ships, skins etc.

Moderators: winston, another_commander

User avatar
spara
---- E L I T E ----
---- E L I T E ----
Posts: 2691
Joined: Wed Aug 15, 2012 4:19 am
Location: Finland

Re: [Release] GalNavy additions (was Adck's Behemoths Redux)

Post by spara »

New facelift (version 1.3.2).

* patched two deprecation(?) bugs in the original GalNavy oxp's mission texts: shipname and score. See a couple of posts above for details.
* fixed one missing texture bug.

I did not do anything to the AI bug in the original Behemoth oxp. There are references to route1PatrolAI.plist in behemothFighterDockingAI.plist. This is a case error. It should be route1patrolAI.plist. It probably works in Windows, but it throws an error in Linux. It's a bit of a rare case, but someone ought to fix it.
User avatar
BuggyBY
Dangerous
Dangerous
Posts: 108
Joined: Thu Feb 09, 2012 9:03 am

Re: [Release] GalNavy additions (was Adck's Behemoths Redux)

Post by BuggyBY »

I'm still on v1.3 of the OXP, so it may have been fixed and not mentioned since then, but I noticed the Navy Transport's exhaust plumes don't quite start from the actual exhausts:
Image
Another picture, Smivs Cobra Mk3 exterior view in foreground for comparison
User avatar
spara
---- E L I T E ----
---- E L I T E ----
Posts: 2691
Joined: Wed Aug 15, 2012 4:19 am
Location: Finland

Re: [Release] GalNavy additions (was Adck's Behemoths Redux)

Post by spara »

BuggyBY wrote:
I'm still on v1.3 of the OXP, so it may have been fixed and not mentioned since then, but I noticed the Navy Transport's exhaust plumes don't quite start from the actual exhausts:
Thanks for reporting, I don't think I have fixed it. I'll check it out and fix it.
Duggan
---- E L I T E ----
---- E L I T E ----
Posts: 496
Joined: Sat Dec 31, 2011 2:58 pm

Re: [Release] GalNavy additions (was Adck's Behemoths Redux)

Post by Duggan »

Hello spara,

I have no idea whether this is to do with any of the Add ons in this thread, but just in case :)
17:46:08.758 [files.notFound]: ----- WARNING: Could not find texture file "adck_cobra3.png".
17:46:08.779 [files.notFound]: ----- WARNING: Could not find texture file "adck_cobra3_e.png".
17:46:08.800 [files.notFound]: ----- WARNING: Could not find texture file "adck_cobra3_n.png".
Flying Python Class Cruiser, Chapter & Verse IV
User avatar
spara
---- E L I T E ----
---- E L I T E ----
Posts: 2691
Joined: Wed Aug 15, 2012 4:19 am
Location: Finland

Re: [Release] GalNavy additions (was Adck's Behemoths Redux)

Post by spara »

Duggan wrote:
Hello spara,

I have no idea whether this is to do with any of the Add ons in this thread, but just in case :)
17:46:08.758 [files.notFound]: ----- WARNING: Could not find texture file "adck_cobra3.png".
17:46:08.779 [files.notFound]: ----- WARNING: Could not find texture file "adck_cobra3_e.png".
17:46:08.800 [files.notFound]: ----- WARNING: Could not find texture file "adck_cobra3_n.png".
Hmmm. Those errors should be fixed in facelift version 1.3.2. Which version are you using?
Duggan
---- E L I T E ----
---- E L I T E ----
Posts: 496
Joined: Sat Dec 31, 2011 2:58 pm

Re: [Release] GalNavy additions (was Adck's Behemoths Redux)

Post by Duggan »

My bad , I got 1.3.1 , I'll update now :)
Flying Python Class Cruiser, Chapter & Verse IV
User avatar
spara
---- E L I T E ----
---- E L I T E ----
Posts: 2691
Joined: Wed Aug 15, 2012 4:19 am
Location: Finland

Re: [Release] GalNavy additions (was Adck's Behemoths Redux)

Post by spara »

Finishing stuff here. I have uploaded version 1.3.3 of the facelift to the box. Fixes the exhaust glitch mentioned earlier. I feel I'm done here, this is probably the final version. Unless more bugs are found of course. And maybe ADCK wants to revive his project :wink: .
User avatar
Keeper
---- E L I T E ----
---- E L I T E ----
Posts: 273
Joined: Fri Feb 01, 2013 7:44 am
Location: Indian Hills, Nevada, USA

Re: [Release] GalNavy additions (was Adck's Behemoths Redux)

Post by Keeper »

spara wrote:
Finishing stuff here. I have uploaded version 1.3.3 of the facelift to the box. Fixes the exhaust glitch mentioned earlier. I feel I'm done here, this is probably the final version. Unless more bugs are found of course. And maybe ADCK wants to revive his project :wink: .
The Minesweeper again has been set to smooth=true, which looks really weird. I've set it back to false in my installation.
User avatar
spara
---- E L I T E ----
---- E L I T E ----
Posts: 2691
Joined: Wed Aug 15, 2012 4:19 am
Location: Finland

Re: [Release] GalNavy additions (was Adck's Behemoths Redux)

Post by spara »

Keeper wrote:
The Minesweeper again has been set to smooth=true, which looks really weird. I've set it back to false in my installation.
:shock: :roll: Oh my, that's not good. Not good at all. I'll better cross check it with an older version when I get home.
User avatar
spara
---- E L I T E ----
---- E L I T E ----
Posts: 2691
Joined: Wed Aug 15, 2012 4:19 am
Location: Finland

Re: [Release] GalNavy additions (was Adck's Behemoths Redux)

Post by spara »

Ok, sorry for the hassle, folks. Smooth lightning issues should be in order now (version 1.3.4). Minesweeper's smooth setting was wrong, it is now corrected. While I was looking at things, I checked the Neolite navy ships and compared them with the original Neolites. Found a couple of inconsistencies in smooth definitions that I corrected to match the originals. If something still looks wrong, keep on reporting and I'll fix it.
User avatar
Keeper
---- E L I T E ----
---- E L I T E ----
Posts: 273
Joined: Fri Feb 01, 2013 7:44 am
Location: Indian Hills, Nevada, USA

Re: [Release] GalNavy additions (was Adck's Behemoths Redux)

Post by Keeper »

Just saw this in my log (1.79-trunk):

Code: Select all

  16:52:20.518 [shader.compile.failure]: ***** ERROR: GLSL fragment shader compilation failed for ncc_carrier.fragment:
>>>>> GLSL log:
ERROR: 2:165: '*' :  wrong operand types  no operation '*' exists that takes a left-hand operand of type 'const int' and a right operand of type '4-component vector of float' (or there is no acceptable conversion)
ERROR: 2:165: 'assign' :  cannot convert from 'const int' to '4-component vector of float'


16:52:20.518 [shader.load.fullModeFailed]: ----- WARNING: Could not build shader oolite-tangent-space-vertex.vertex/ncc_carrier.fragment in full complexity mode, trying simple mode.
  16:52:20.520 [shader.compile.failure]: ***** ERROR: GLSL fragment shader compilation failed for ncc_carrier.fragment:
>>>>> GLSL log:
ERROR: 2:165: '*' :  wrong operand types  no operation '*' exists that takes a left-hand operand of type 'const int' and a right operand of type '4-component vector of float' (or there is no acceptable conversion)
ERROR: 2:165: 'assign' :  cannot convert from 'const int' to '4-component vector of float'


16:52:20.520 [shader.load.failed]: ***** ERROR: Could not build shader oolite-tangent-space-vertex.vertex/ncc_carrier.fragment.
User avatar
spara
---- E L I T E ----
---- E L I T E ----
Posts: 2691
Joined: Wed Aug 15, 2012 4:19 am
Location: Finland

Re: [Release] GalNavy additions (was Adck's Behemoths Redux)

Post by spara »

Keeper wrote:
Just saw this in my log (1.79-trunk):

Code: Select all

  16:52:20.518 [shader.compile.failure]: ***** ERROR: GLSL fragment shader compilation failed for ncc_carrier.fragment:
>>>>> GLSL log:
ERROR: 2:165: '*' :  wrong operand types  no operation '*' exists that takes a left-hand operand of type 'const int' and a right operand of type '4-component vector of float' (or there is no acceptable conversion)
ERROR: 2:165: 'assign' :  cannot convert from 'const int' to '4-component vector of float'


16:52:20.518 [shader.load.fullModeFailed]: ----- WARNING: Could not build shader oolite-tangent-space-vertex.vertex/ncc_carrier.fragment in full complexity mode, trying simple mode.
  16:52:20.520 [shader.compile.failure]: ***** ERROR: GLSL fragment shader compilation failed for ncc_carrier.fragment:
>>>>> GLSL log:
ERROR: 2:165: '*' :  wrong operand types  no operation '*' exists that takes a left-hand operand of type 'const int' and a right operand of type '4-component vector of float' (or there is no acceptable conversion)
ERROR: 2:165: 'assign' :  cannot convert from 'const int' to '4-component vector of float'


16:52:20.520 [shader.load.failed]: ***** ERROR: Could not build shader oolite-tangent-space-vertex.vertex/ncc_carrier.fragment.
A shader problem... :cry: . Not my area of expertise really, although I tweaked the default shader for this one. I'll see if I can figure out something. Haven't noticed the error myself though, which makes fixing harder. Could this have something to do with 1.79-trunk? Thanks for reporting anyway.
User avatar
cim
Quite Grand Sub-Admiral
Quite Grand Sub-Admiral
Posts: 4072
Joined: Fri Nov 11, 2011 6:19 pm

Re: [Release] GalNavy additions (was Adck's Behemoths Redux)

Post by cim »

Some graphics cards are more tolerant than others about trying to stick data of the wrong type into a variable, so you might not get the error yourself. It's unlikely to be specific to 1.79.

Check your datatypes in the .fragment file: somewhere you have an int on the left of an = sign, and a vec4 on the right. "2:165" is either "line:column" or "column:line" (I can't remember which offhand) if you want some hints about where the error might be.
User avatar
Keeper
---- E L I T E ----
---- E L I T E ----
Posts: 273
Joined: Fri Feb 01, 2013 7:44 am
Location: Indian Hills, Nevada, USA

Re: [Release] GalNavy additions (was Adck's Behemoths Redux)

Post by Keeper »

It's definitely not specific to 1.79. I just added the ncc_carrier ships to a demoships.plist and ran 1.77 then scrolled through the demo ships. The same error appeared in the log for each one -- and they all were using simple shaders, thus didn't have their nameplates or any decent lighting.
User avatar
spara
---- E L I T E ----
---- E L I T E ----
Posts: 2691
Joined: Wed Aug 15, 2012 4:19 am
Location: Finland

Re: [Release] GalNavy additions (was Adck's Behemoths Redux)

Post by spara »

Ok. The error message was quite clear and the culprit was easy to spot. I was multiplying a vector (vec4) with an integer and that operation seems to be a no-no in shader language. By chance it was working in my system.

Originally I was doing this:

Code: Select all

5.0 * CalcSpecularLight(lightVector, eyeVector, exponent, normal, gl_LightSource[1].specular)
After reading some shader language specs, I have concluded that the correct way to do it is this:

Code: Select all

vec4(5.0, 5.0, 5.0, 5.0) * CalcSpecularLight(lightVector, eyeVector, exponent, normal, gl_LightSource[1].specular)
Please correct me, if I'm wrong. Both work just fine in my system so I can't tell any difference.

New version is uploaded to the box and link updated to the wiki. Could you please test it, Keeper, and report back?
Post Reply