0.5.1 beta2 release (yep, beta 2)

Just downloaded Vega Strike and need help? Have a question but don't know where to start? Ask here.
klauss
Elite
Elite
Posts: 7243
Joined: Mon Apr 18, 2005 2:40 pm
Location: LS87, Buenos Aires, República Argentina

Re: 0.5.1 beta2 release (yep, beta 2)

Post by klauss »

Win XP, 32 bits, AMD Athlon 64, 1.5G ram, GeForce 6600.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
macghost
Star Pilot
Star Pilot
Posts: 7
Joined: Fri Apr 09, 2010 9:50 pm

Re: 0.5.1 beta2 release (yep, beta 2)

Post by macghost »

I have 0.5.1b2 working on my macboook under macos 10.6 . It seems to run fine. The planet doens't look so nice anymore but I dunno if that is a mac thing or an integrated graphics thing. Will see if I can come up with a dmg for other mac users.
klauss
Elite
Elite
Posts: 7243
Joined: Mon Apr 18, 2005 2:40 pm
Location: LS87, Buenos Aires, República Argentina

Re: 0.5.1 beta2 release (yep, beta 2)

Post by klauss »

What GPU are you using?

"No shader" settings are effectively looking worse, but no shader should be a last-resort fallback in even older hardware, as even rather old onboard GPU chips have some form of programmable shading.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
macghost
Star Pilot
Star Pilot
Posts: 7
Joined: Fri Apr 09, 2010 9:50 pm

Re: 0.5.1 beta2 release (yep, beta 2)

Post by macghost »

I have the much-hated Intel GMA 950.
I'm getting playable results under "verylowshader" and "GeomMedium" with all the planets and ships I have seen so far looking fine. Planets, the fighter base, and the exterior view of my llama.begin, do not appear correctly on higher shader settings. Any suggestion on how to make a binary that will run on a machine I didn't compile it on?
klauss
Elite
Elite
Posts: 7243
Joined: Mon Apr 18, 2005 2:40 pm
Location: LS87, Buenos Aires, República Argentina

Re: 0.5.1 beta2 release (yep, beta 2)

Post by klauss »

If you can post console output for a few shader settings, it will help in diagnosing.

For building a binary, as long as you link against system libraries it should be fine to run on a similar system (ie: same library versions). If you have libraries you had to build, rather than shipping them you should link those statically.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
klauss
Elite
Elite
Posts: 7243
Joined: Mon Apr 18, 2005 2:40 pm
Location: LS87, Buenos Aires, República Argentina

Re: 0.5.1 beta2 release (yep, beta 2)

Post by klauss »

Sooo.... what's left for the final release?

I see only 0.5.2 stuff on the trackers. Am I missing something?
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
loki1950
The Shepherd
Posts: 5841
Joined: Fri May 13, 2005 8:37 pm
Location: Ottawa
Contact:

Re: 0.5.1 beta2 release (yep, beta 2)

Post by loki1950 »

Well according to the wiki task list it appears ready to package.So lets get it out the door so the users can point us to the bugs we missed.


Enjoy the Choice :)
my box::HP Envy i5-6400 @2Q70GHzx4 8 Gb ram/1 Tb(Win10 64)/3 Tb Mint 19.2/GTX745 4Gb acer S243HL K222HQL
Q8200/Asus P5QDLX/8 Gb ram/WD 2Tb 2-500 G HD/GF GT640 2Gb Mint 17.3 64 bit Win 10 32 bit acer and Lenovo ideapad 320-15ARB Win 10/Mint 19.2
pheonixstorm
Elite
Elite
Posts: 1567
Joined: Tue Jan 26, 2010 2:03 am

Re: 0.5.1 beta2 release (yep, beta 2)

Post by pheonixstorm »

There are a few completed models to try adding int, but other than that I don't see anything else left on the roadmap that needs to go in.
Because of YOU Arbiter, MY kids? can't get enough gas. OR NIPPLE! How does that mkae you feeeel? ~ Halo
energyman76b
ISO Party Member
ISO Party Member
Posts: 445
Joined: Tue Feb 11, 2003 8:04 am

Re: 0.5.1 beta2 release (yep, beta 2)

Post by energyman76b »

looks good with an ati 5770 so far and everything cranked up. Except for the segfaults.
klauss
Elite
Elite
Posts: 7243
Joined: Mon Apr 18, 2005 2:40 pm
Location: LS87, Buenos Aires, República Argentina

Re: 0.5.1 beta2 release (yep, beta 2)

Post by klauss »

energyman76b wrote:looks good with an ati 5770 so far and everything cranked up. Except for the segfaults.
We want stacktraces and bug reports.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
energyman76b
ISO Party Member
ISO Party Member
Posts: 445
Joined: Tue Feb 11, 2003 8:04 am

Re: 0.5.1 beta2 release (yep, beta 2)

Post by energyman76b »

klauss wrote:
energyman76b wrote:looks good with an ati 5770 so far and everything cranked up. Except for the segfaults.
We want stacktraces and bug reports.
something like this I presume:
*** Python Warning 16!
Warning Traceback 16:
File "m/privateer.py", line 34, in Execute
i.Execute()
File "m/random_encounters.py", line 288, in Execute
dynamic_battle.UpdateCombatTurn()
File "m/dynamic_battle.py", line 34, in UpdateCombatTurn
if (not SimulateBattles()):
File "m/dynamic_battle.py", line 215, in SimulateBattles
if (not attackFlightgroup (ally[0],ally[1],enemy[0],enemy[1],enemy[2])):
File "m/dynamic_battle.py", line 565, in attackFlightgroup
SimulatedDukeItOut (fgname,faction,enfgname,enfaction)
File "m/dynamic_battle.py", line 448, in SimulatedDukeItOut
ApplyDamage(enfgname,enfaction,enemy,envictim,enstats,HowMuchDamage(ally,vsrandom.uniform(0,enstats[1])),fgname,faction)
File "m/dynamic_battle.py", line 418, in HowMuchDamage
stats=faction_ships.GetStats(i[0])
File "m/faction_ships.py", line 577, in GetStats
debug.error( 'cannot find '+name)
Message: cannot find Entourage

ship destroyed-no flightgroup
ship destroyed-no flightgroup
ship destroyed-no flightgroup
ship destroyed-no flightgroup
ship destroyed-no flightgroup
zsh: segmentation fault ./vegastrike

not very helpfull I guess. But all I got.

Will see what happens in the next couple of days. Got just home from work.
klauss
Elite
Elite
Posts: 7243
Joined: Mon Apr 18, 2005 2:40 pm
Location: LS87, Buenos Aires, República Argentina

Re: 0.5.1 beta2 release (yep, beta 2)

Post by klauss »

Run with gdb:

gdb --args ./vegastrike --debug=3

That will provide more helpful info even without a stacktrace, which you get after the crash with:

(gdb) bt full

Just be sure not to run VS in full-screen, as it won't be happy about the debugger then.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
klauss
Elite
Elite
Posts: 7243
Joined: Mon Apr 18, 2005 2:40 pm
Location: LS87, Buenos Aires, República Argentina

Re: 0.5.1 beta2 release (yep, beta 2)

Post by klauss »

I'm preparing the release on SF's servers. I'll wait a bit to be able to diagnose this segfault, but if necessary we can make a maintenance release later so I'm not waiting thaaaat long ;)
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
energyman76b
ISO Party Member
ISO Party Member
Posts: 445
Joined: Tue Feb 11, 2003 8:04 am

Re: 0.5.1 beta2 release (yep, beta 2)

Post by energyman76b »

ok, that was painfull - recompiled without the release setting, didn't strip the binaries, start a new game in gdb and flew to the az-17 jumppoint (because shooting around there triggered the last segfault).

After a while the game locked up.
Then the keyboard
Then I had to use sysrq keys to reboot.

So three crashes in three games. Once wildly changing around between the viewpoints, 2 times turning a lot. Hm.

Next step: redo, but have to pipe the output somewhere.
energyman76b
ISO Party Member
ISO Party Member
Posts: 445
Joined: Tue Feb 11, 2003 8:04 am

Re: 0.5.1 beta2 release (yep, beta 2)

Post by energyman76b »

and again - this time 'rawed' quickly enough and saw gdb and vs still running. But attaching&stuff did not work.

So, whenever gdb is involved, I can't do anything, because on crash I am not dumped back to the desktop?

Oh, and stderr did not include anything interessting at all:

pages and pages of this stuff:
Beam collide with 0x7fffc85f97f0, distance 57.315315
Drawn 175388 vertices in 32 batches
Drawn 175388 vertices in 32 batches
Beam collide with 0x7fffc2236a30, distance 94.556427
Beam collide with 0x7fffc85f97f0, distance 57.315315
Beam collide with 0x7fffc85f5b10, distance 15.931334
Beam collide with 0x7fffc85f5b10, distance 14.995034
Drawn 175388 vertices in 32 batches
Drawn 175388 vertices in 32 batches
energyman76b
ISO Party Member
ISO Party Member
Posts: 445
Joined: Tue Feb 11, 2003 8:04 am

Re: 0.5.1 beta2 release (yep, beta 2)

Post by energyman76b »

ok, windowed mode worked fine. Started new game, flew to 17-ar, got killed, reloaded, flew to Relay --> crash:
Program received signal SIGSEGV, Segmentation fault.
0x00007fffdffe4c04 in ?? () from /usr/lib64/dri/fglrx_dri.so

bt full
#0 0x00007fffdffe4c04 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#1 0x00007fffe00042ca in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#2 0x00007fffe0006074 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#3 0x00007fffdf75319e in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#4 0x00007fffdf753c96 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#5 0x00007fffdf60c8b9 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#6 0x00007fffdf60d3f1 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#7 0x00007fffdf61d7fb in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#8 0x00007fffdf617448 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#9 0x00007fffdf75257d in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#10 0x00007fffdf8c4cad in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#11 0x0000000000b568a9 in GFXActiveTexture(int) ()
No symbol table info available.
#12 0x0000000000b5695f in GFXDisable(STATE) ()
No symbol table info available.
#13 0x000000000093f4ca in GameCockpit::DrawTurretTargetBoxes(Radar::Sensor const&) ()
No symbol table info available.
#14 0x0000000000944bbe in GameCockpit::Draw() ()
No symbol table info available.
#15 0x00000000009018b6 in GameStarSystem::Draw(bool) ()
No symbol table info available.
#16 0x00000000009053f9 in GameUniverse::StartDraw() ()
No symbol table info available.
#17 0x00000000008f0d71 in main_loop() ()
No symbol table info available.
#18 0x0000000000b5d936 in winsys_process_events ()
No symbol table info available.
#19 0x0000000000b61898 in main ()
No symbol table info available.

oh looky.

This is 8.96-rc1 of AMD's driver package for linux (yes, I know, unreleased). It is working fine with ut2004 and desktop is stable as rock.

More info: amd64, gentoo, kernel 3.0.25, xorg-1.10.6, glibc 2.14.1
klauss
Elite
Elite
Posts: 7243
Joined: Mon Apr 18, 2005 2:40 pm
Location: LS87, Buenos Aires, República Argentina

Re: 0.5.1 beta2 release (yep, beta 2)

Post by klauss »

I'm thinking it's your drivers, at least that's where it died according to the backtrace.

Try tracing again, to see whether it crashes at the same spot.

In any case, if it's the drivers, it's probably related to shaders, so trying lower shader settings should help work around the bug.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
energyman76b
ISO Party Member
ISO Party Member
Posts: 445
Joined: Tue Feb 11, 2003 8:04 am

Re: 0.5.1 beta2 release (yep, beta 2)

Post by energyman76b »

this time SIGBUS:
rogram received signal SIGBUS, Bus error.
0x00007fffe04f6990 in ?? () from /usr/lib64/dri/fglrx_dri.so
(gdb) bt full
#0 0x00007fffe04f6990 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#1 0x00007fffe05cdfd4 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#2 0x00007fffe05c7a71 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#3 0x00007fffe00042a8 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#4 0x00007fffe0006074 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#5 0x00007fffdf75319e in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#6 0x00007fffdf753c96 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#7 0x00007fffdf60c8b9 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#8 0x00007fffdf60d3f1 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#9 0x00007fffdf61d7fb in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#10 0x00007fffdf617448 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#11 0x00007fffdf75257d in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#12 0x00007fffdf8c4cad in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#13 0x0000000000b568a9 in GFXActiveTexture(int) ()
No symbol table info available.
#14 0x0000000000b5695f in GFXDisable(STATE) ()
No symbol table info available.
#15 0x000000000093f4ca in GameCockpit::DrawTurretTargetBoxes(Radar::Sensor const&) ()
No symbol table info available.
#16 0x0000000000944bbe in GameCockpit::Draw() ()
No symbol table info available.
#17 0x00000000009018b6 in GameStarSystem::Draw(bool) ()
No symbol table info available.
#18 0x00000000009053f9 in GameUniverse::StartDraw() ()
No symbol table info available.
#19 0x00000000008f0d71 in main_loop() ()
No symbol table info available.
#20 0x0000000000b5d936 in winsys_process_events ()
energyman76b
ISO Party Member
ISO Party Member
Posts: 445
Joined: Tue Feb 11, 2003 8:04 am

Re: 0.5.1 beta2 release (yep, beta 2)

Post by energyman76b »

what I will try next: different driver version. But not in the next *lookatwatch* 8h at least.
klauss
Elite
Elite
Posts: 7243
Joined: Mon Apr 18, 2005 2:40 pm
Location: LS87, Buenos Aires, República Argentina

Re: 0.5.1 beta2 release (yep, beta 2)

Post by klauss »

dNice. Same place. Sigbus is
similar to segfault. We might be able to come up with a workaround.
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
energyman76b
ISO Party Member
ISO Party Member
Posts: 445
Joined: Tue Feb 11, 2003 8:04 am

Re: 0.5.1 beta2 release (yep, beta 2)

Post by energyman76b »

ok, next driver - same problem. Will report this problem on the appropriate mailing list.

Program received signal SIGBUS, Bus error.
0x00007fffe051c580 in ?? () from /usr/lib64/dri/fglrx_dri.so
(gdb) bt full
#0 0x00007fffe051c580 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#1 0x00007fffe05f3d74 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#2 0x00007fffe05ed711 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#3 0x00007fffe004a3a8 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#4 0x00007fffe004bc94 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#5 0x00007fffdf773b99 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#6 0x00007fffdf7746c4 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#7 0x00007fffdf636039 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#8 0x00007fffdf636a61 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#9 0x00007fffdf64738b in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#10 0x00007fffdf640bc8 in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#11 0x00007fffdf772f6d in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#12 0x00007fffdf90282d in ?? () from /usr/lib64/dri/fglrx_dri.so
No symbol table info available.
#13 0x0000000000b568a9 in GFXActiveTexture(int) ()
No symbol table info available.
#14 0x0000000000b5695f in GFXDisable(STATE) ()
No symbol table info available.
#15 0x000000000093f4ca in GameCockpit::DrawTurretTargetBoxes(Radar::Sensor const&) ()
No symbol table info available.
#16 0x0000000000944bbe in GameCockpit::Draw() ()
No symbol table info available.
#17 0x00000000009018b6 in GameStarSystem::Draw(bool) ()
No symbol table info available.
#18 0x00000000009053f9 in GameUniverse::StartDraw() ()
No symbol table info available.
#19 0x00000000008f0d71 in main_loop() ()
No symbol table info available.
#20 0x0000000000b5d936 in winsys_process_events ()
---Type <return> to continue, or q <return> to quit---
No symbol table info available.
#21 0x0000000000b61898 in main ()
No symbol table info available.
klauss
Elite
Elite
Posts: 7243
Joined: Mon Apr 18, 2005 2:40 pm
Location: LS87, Buenos Aires, República Argentina

Re: 0.5.1 beta2 release (yep, beta 2)

Post by klauss »

What's your configuration like? (ie, detail level, texture resolution, shader level)
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
klauss
Elite
Elite
Posts: 7243
Joined: Mon Apr 18, 2005 2:40 pm
Location: LS87, Buenos Aires, República Argentina

Re: 0.5.1 beta2 release (yep, beta 2)

Post by klauss »

You know, that particular stacktrace points to a really unlikely culprit. It basically points to an API call to set some clientside state on OpenGL - it shouldn't by any means crash like that. Something funky is happening earlier that de-stabilizes the driver.

If you can fiddle around with config settings to work around the bug and post the details, we might figure out what is it. At this point, I'm hunting blind.

Whatever you do, I'd ask that you create an issue on the bug tracker with all this information so we can track it down properly (this forum is not the ideal means of tracking bugs).
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
energyman76b
ISO Party Member
ISO Party Member
Posts: 445
Joined: Tue Feb 11, 2003 8:04 am

Re: 0.5.1 beta2 release (yep, beta 2)

Post by energyman76b »

Geometry: extreme detail
textures: unlimited
factions textures: on
shaders: extreme shaders
accelerated visual: on
autodocker: on.
resolution: eitehr 1920x1200 fs or 1400x1050 windowed

xorg-server 1.10.6, amd64, amd drivers. hd5770 card with 1gb ram. 8gb system ram. AMD drivers are prereleases. KDE 4.8 desktop. If something 'destabilizes' the driver it should:
also crash kwin
crash ut2004
or
do funky stuff on resume from suspend to ram.

and all those work just fine.

I will open a bug on the tracker soon.
klauss
Elite
Elite
Posts: 7243
Joined: Mon Apr 18, 2005 2:40 pm
Location: LS87, Buenos Aires, República Argentina

Re: 0.5.1 beta2 release (yep, beta 2)

Post by klauss »

energyman76b wrote:If something 'destabilizes' the driver it should:
also crash kwin
crash ut2004
or
do funky stuff on resume from suspend to ram.
Not necessarily. We've hit lots of bugs nobody else hit, in Mesa, nvidia drivers, everywhere.

Did you try lower shader settings?
Oíd mortales, el grito sagrado...
Call me "Menes, lord of Cats"
Wing Commander Universe
Post Reply