BETA Release OGLAClient Beta 100511: Bug hunt continues

[Auzar]

You need orbiter100503beta.

Excuse, I did not see that there was a new Orbiter beta, with it to this error is not present )
My English is very very bad, sorry )))
Прошу прощения, не видел что вышла новая бета Орбитера, с ней этой ошибки нет.
с моим английским трудно что то понять даже с онлайн перводчиком, к сожалению )))

 

[tl8]

Artlav, a dumb user POV, I have an ATI card and the only thing that seems to of happened is that the yellow text at the top of the screen is missing.

I don't know what setting I should be trying or anything like that though

 

[Artlav]

Surface MFD still have an garbled look, almost to the point of unreadability. See attached screenshot.

To be fixed eventually. Can't tell what exactly causes it.

No PAPI/VASI lights at the SLF

The landing lights along the runway that helps one line up?
No idea where they are in the core's data, would fix once found.

No air-shade with ray-traced atmosphere. No way to have both?

Should be some, but it's safe to say that this is not yet implemented properly.

Artlav, a dumb user POV, I have an ATI card and the only thing that seems to of happened is that the yellow text at the top of the screen is missing.

I don't know what setting I should be trying or anything like that though

What happens when you cycle thru cockpit modes in the shuttle-A or DG?
Any crashes if you do the same in two provided scenarios?
Will the text reappear if you update to OGLAClient 100504?

With the new version, everything works *except* the raytraced atmosphere.

Lucky, because i just swatted a bug that causes ISS to vanish and Shuttle to burst into triangles on some systems.

 

[tl8]

In a nutshell nothing unusual happens

Ok, Cockpit of the DG seems ok, but the current lack of ISS is concerning

Shuttle-A seems to work, some text on the 2D panel is offset(ripped) still

The MFD problem I had before is gone

The yellow text is back but looks rubbish, should I use Vista Boost (clear text off)

General Question, are the Function Key menus implemented yet? Only F1, F7 and F8 work

 

[DaveS]

Two more bugs to report:

• The new Map MFD doesn't display right. Where the map display is supposed to be, there's only a light-blueish rectangle
• Using the target selection menu in the Map MFD causes a CTD

---------- Post added at 10:54 AM ---------- Previous post was at 10:50 AM ----------

The landing lights along the runway that helps one line up?
No idea where they are in the core's data, would fix once found.

They're part of the RUNWAYLIGHTS definition in base cfgs. So look for those.

 

[Artlav]

but the current lack of ISS is concerning

Already fixed.

The yellow text is back but looks rubbish, should I use Vista Boost (clear text off)

Screenshot please?
The text in date/time accel area is generated in the core, so it might be affected by Windows. If it looks anti-aliased (like MFDs used to look like before), then it's Vista Boost job to fix that.

General Question, are the Function Key menus implemented yet? Only F1, F7 and F8 work

F2 doesn't work? Or, do you mean menus not being visible in full-screen mode?
The keys are processed by Orbiter core directly, so it should be either none work (an old bug that happens if you switch window while Orbiter loads), or everything works.

The new Map MFD doesn't display right. Where the map display is supposed to be, there's only a light-blueish rectangle

Confirmed, investigating.
Would it fix itself if you switch MFDs a few times (~1-10)?

Using the target selection menu in the Map MFD causes a CTD

Only in spaceports submenu or at once?

 

[DaveS]

Confirmed, investigating.
Would it fix itself if you switch MFDs a few times (~1-10)?

Only in spaceports submenu or at once?

1: Cycling it(once) cleared the display problem.
2: Only spaceports sub-menu, the rest are fine.

 

[Artlav]

OGLAClient 100505:
orbides.1gb.ru/orbf/oglaclient-100505_beta-100503.zip
Requires Orbiter beta 100503.
Raytraced haze tables: http://orbides.1gb.ru/orbf/ogla_earth_scatter.zip

Changes:
*Fixed mesh issue from 100504
*Fixed Atlantis animations bug
*Fixed surface filling issues (Surface MFD)
*Fixed variety of 2D rendering bugs
*Fixed variety of add-on related crashes

Surface MFD now fixed.
Bad thing is, GDI emulation got broken as a result, still waiting for Martins clarifications.

Known bugs and near-term release plans:
-GDI emulation broken (Map MFD, all legacy MFDs, old panels)
-MFD transparency issues (Arrow only?)
-Something crashes OFSS scenarios
-Lightweight Lv9+ planet rendering (non-Orulex)
-Switch for raytraced air into F7 menu
-Smooth night side light/dark transition, sunset lighting effect
-Documentation!
-Make planetary shine angle-dependent (dawn, sunset)
-Add particle streams
-Get the trash out
-Shadows problem on Nvidia (?)
-F* key menus problems (?)
-Terrain tiling blinking issues (?)
-Fix visibility sequence for terrain (?)
-Proper tangent storage and handling (?)
-Animate tangent (?)
-Fading exhaust (?)

1: Cycling it(once) cleared the display problem.
2: Only spaceports sub-menu, the rest are fine.

1. Broken, to be fixed soon.
2. No idea what causing it, to be fixed eventually.

 

[tl8]

Ok, With Vista Boost, the text is better

F2 now works, I don't know if the other F keys are not working or not showing up

All the textures seem to be missing

The MFDs in the HUD only mode are broken, no black background and no bottom green line

Edit: Found where the function fey menus are coming up, small error there no buttons are on the outline (Close, Help etc)

Edit 2: The Texture problem seems to of fixed itself

 

[Artlav]

OGLAClient 100506:
http://orbides.1gb.ru/orbf/oglaclient-100506_beta-100503.zip
Requires Orbiter beta 100503.
Raytraced haze tables: http://orbides.1gb.ru/orbf/ogla_earth_scatter.zip

A clean-up release.
Looks like it's a go for more in-depth testing, as i runned out of petty bugs to fix.


If you know or found an OGLAClient bug that is not on the list below, can provide details on ones marked with (?), or tested the beta and found that some of the bugs llisted as fixed are actually not, please provide feedback.

Be advised that i can't fix bugs that i'm not constantly reminded of.

Changes:
*Straightened out 2D rendering (MapMFD, HUD, lines)
*Fixed GDI emulation (Map MFD, all legacy MFDs, old panels)
*Fixed MFD transparency issues (Arrow, non-transparent MFDs)
*Added a switch for raytraced air into F7 menu
*Updated the manual

Known bugs and near-term release plans:
-Something crashes OFSS scenarios (multistage2.dll ?? )
-Lightweight Lv9+ planet rendering (non-Orulex)
-Smooth night side light/dark transition, sunset lighting effect
-Make planetary shine angle-dependent (dawn, sunset)
-Add particle streams
-Shadows problem on Nvidia (?)
-F* key menus problems (?)
-Terrain tiling blinking issues (?)
-Fix visibility sequence for terrain (?)
-Proper tangent storage and handling (?)
-Animate tangent (?)
-Fading exhaust (?)

The MFDs in the HUD only mode are broken, no black background and no bottom green line

Fixed.

 

[DaveS]

Artlav: Does OGLA support dynamic textures yet? Seems like SSU have a problem with several of the VC panels without dynamic texture support, namely that the panels that make sure of the dynamic texturing is blank white!

 

[Quick_Nick]

Alright, I'm very recently jumping into use of Orbiter beta, and I've only tried to use OGLA a couple times. And for some reason, while it works fine (although VERY slow) on my old single-core, it just won't run on my decent duel-core.
Windows Vista x64
Pentium Duel-Core @ 2.50GHz
Unfortunately using a builtin graphics controller, so maybe this is the problem.

I'm getting this in the ogla log:

5/6/2010 3:52:11 PM|    INIT| OGLA v100506 (GLGR)
5/6/2010 3:52:41 PM|    INIT| GPU supports OpenGL 1.2
5/6/2010 3:52:41 PM|    INIT| GPU supports OpenGL 1.4
5/6/2010 3:52:41 PM|    INIT| GPU supports OpenGL 2.0
5/6/2010 3:52:41 PM|    INIT| glgruva=1, usevbo=1, vboav=1, glgr_stensh_aupd=0, gvsync=0, gl_comp_sup=1, gl_shm4=0
5/6/2010 3:52:41 PM| OGLADBG| OGLA 100506 Debug.
5/6/2010 3:52:43 PM|  OGLTEX| Error: (oglc_blit) Access violation at address 046CF918 in module 'ig4dev32.dll'. Write of address 00000000 (hc=0)
5/6/2010 3:52:43 PM|  OGLTEX| Error: (oglc_blit) Access violation at address 046CF918 in module 'ig4dev32.dll'. Write of address 00000000 (hc=0)
5/6/2010 3:52:43 PM|  OGLTEX| Error: (oglc_blit) Access violation at address 046CF918 in module 'ig4dev32.dll'. Write of address 00000000 (hc=0)
5/6/2010 3:52:43 PM|  OGLTEX| Error: (oglc_blit) Access violation at address 046CF918 in module 'ig4dev32.dll'. Write of address 00000000 (hc=0)
5/6/2010 3:52:43 PM|  OGLTEX| Error: (oglc_blit) Access violation at address 046CF918 in module 'ig4dev32.dll'. Write of address 00000000 (hc=0)
5/6/2010 3:52:43 PM|  OGLTEX| Error: (oglc_blit) Access violation at address 046CF918 in module 'ig4dev32.dll'. Write of address 00000000 (hc=0)
5/6/2010 3:52:43 PM|  OGLTEX| Error: (oglc_blit) Access violation at address 046CF918 in module 'ig4dev32.dll'. Write of address 00000000 (hc=0)
5/6/2010 3:52:43 PM|  OGLTEX| Error: (oglc_blit) Access violation at address 046CF918 in module 'ig4dev32.dll'. Write of address 00000000 (hc=0)
5/6/2010 3:52:43 PM|  OGLTEX| Error: (oglc_blit) Access violation at address 046CF918 in module 'ig4dev32.dll'. Write of address 00000000 (hc=0)
5/6/2010 3:52:43 PM|  OGLTEX| Error: (oglc_blit) Access violation at address 046CF918 in module 'ig4dev32.dll'. Write of address 00000000 (hc=0)
5/6/2010 3:52:43 PM|  OGLTEX| Error: (oglc_blit) Access violation at address 046CF918 in module 'ig4dev32.dll'. Write of address 00000000 (hc=0)
5/6/2010 3:52:43 PM|  OGLTEX| Error: (oglc_blit) Access violation at address 046CF918 in module 'ig4dev32.dll'. Write of address 00000000 (hc=0)
5/6/2010 3:52:43 PM|  OGLTEX| Error: (oglc_blit) Access violation at address 046CF918 in module 'ig4dev32.dll'. Write of address 00000000 (hc=0)
5/6/2010 3:52:43 PM|  OGLTEX| Error: (oglc_blit) Access violation at address 046CF918 in module 'ig4dev32.dll'. Write of address 00000000 (hc=0)

And orbiter.log has this error at the bottom:

---------------------------------------------------------------
>>> WARNING: Obsolete API function used: oapiBlt
Colour key argument not supported by graphics client
---------------------------------------------------------------

I've atached the dxdiag if it helps anything.

 

[Quick_Nick]

Okay, I ran a different scenario. Orbiter actually loaded up, but then immediately crashed with similar errors in ogla.log and no errors in orbiter.log
EDIT: And trying the OGLA-setscene scenario, it runs with the first obvious problem being that the information at the top of the screen is replaced by big white rectangles. But it doesn't crash until I go into cockpit, where I see all of the graphics jumbled all over the place.
My suspicions are: I may need to get a real graphics card. Norton freaked out when I tried to install the runtime libraries the first time, but testinstall now says OK. Maybe Vista freaked out about something.

 

[Artlav]

Artlav: Does OGLA support dynamic textures yet? Seems like SSU have a problem with several of the VC panels without dynamic texture support, namely that the panels that make sure of the dynamic texturing is blank white!

If you see the name on the DG wing, then dynamic textures work, as they should be since ages ago.
My wild guess is that SSU uses GDI to draw them, and then something mistranslates in GDI emulation.
Will investigate.
What exactly does not work, and where to look for it?

while it works fine (although VERY slow) on my old single-core

Strange, it should work at least as fast as DX7 client on old hardware (unless you want to see a load of advanced graphics, in which case it's logical).

Unfortunately using a builtin graphics controller, so maybe this is the problem.
...

5/6/2010 3:52:41 PM|    INIT| GPU supports OpenGL 1.4
5/6/2010 3:52:43 PM|  OGLTEX| Error: (oglc_blit) Access violation at address 046CF918 in module 'ig4dev32.dll'. Write of address 00000000 (hc=0)

Intel GPU.
The problem is that it reports OGL 1.4 as supported, yet it does not support FBO somehow.
Will have to make feature testing more aware of unusual cases.

Could you please run it again with this module:
http://orbides.1gb.ru/orbf/ogla-100506-1.zip
(put into modules directory)
EDIT: as well as http://orbides.1gb.ru/orbf/glgrcln.zip , posting glgr.txt
And post the log after it crashes?

My suspicions are: I may need to get a real graphics card.

Quite likely the case. Intel GPUs are flirting with standards like crazy, besides being generally underpowered.

 

[DaveS]

If you see the name on the DG wing, then dynamic textures work, as they should be since ages ago.
My wild guess is that SSU uses GDI to draw them, and then something mistranslates in GDI emulation.
Will investigate.
What exactly does not work, and where to look for it?

Well, the C3 panel is texture-less and is white(as the basic material color is white). Same thing with the aft panels.

It isn't a case of missing textures as they load fine in both the inline and the external DX7 clients.

Also for some reason the payload bay isn't drawn in the interior views. Just blank, empty space.

 

[tl8]

Looking alot better

MFD buttons in the 2D panel still look rubbish

When I 1st load up with Advanced Atmosphere all the textures are gone, quitting and switching to Ray traced Atmosphere fixes the problem

When I had Advanced Atmosphere, The 3d terrain was blocky and was quite visible

Some text on the DG 2D panel looks rubbish, I think that could be the texture but not sure

Yellow text looks better though some letters are short changed pixels on the right hand side

Cycling the Yellow text (F7 I) produced a line of pixels below the top left

 

[Artlav]

When I 1st load up with Advanced Atmosphere all the textures are gone, quitting and switching to Ray traced Atmosphere fixes the problem

When I had Advanced Atmosphere, The 3d terrain was blocky and was quite visible

Could you provide screenshots for both?

 

[tl8]

It seems to only happen to a new cfg file (I had to reinstall it)

This also shows all 3 errors

 

[Artlav]

Well, the C3 panel is texture-less and is white(as the basic material color is white). Same thing with the aft panels.

Also for some reason the payload bay isn't drawn in the interior views. Just blank, empty space.

Confirmed, althrough the extra MFDs light up if turned on.
There also seems to be some sort of a crash during the loading, do you get it?
Investigating.

It seems to only happen to a new cfg file (I had to reinstall it)

Sorry, that means what exactly?

This also shows all 3 errors

Could you provide the ogla.log please? Sounds like shaders don't compile.

 

[DaveS]

Confirmed, althrough the extra MFDs light up if turned on.
There also seems to be some sort of a crash during the loading, do you get it?

No but, I'm not using v1.25 but the latest code revision in the SSU SVN repository.

Is it any specific scenario that CTDs or just a special one.