New Release D3D9Client Development

[kuddel]

Why are you trying a different approach every time?
But that aside, let me try to help you:

1. You've installed the DirectX (9) runtimes => good
2. You should unzip Orbiter 2016 in a folder of your choice (http://orbit.medphys.ucl.ac.uk/download.html). Here you might have chosen the "msi Installer". That is also an option. I usually chose the ZIP and unpack it wherever I like, but the msi-Option should work as well (I assume that is what you meant when you said "[...]Unzipped into a blank folder and ran setup[...]", right?
3. You extract the content of "3D9Client for Orbiter 2016" into the folder you've filled at step 2. (the download you mentioned - D3D9ClientR3.8-forOrbiter2016(r1137).zip - is O.K). Make sure you keep the sub-folder structure when extracting.
4. Run ortiter_ng.exe, select D3D9Client under "Modules"
5. Setup the video tab (resolution etc.)
6. Start a scenario

After that you might like to install (extract into your Orbiter folder) HiRes Textures etc. also to be found on http://orbit.medphys.ucl.ac.uk/download.html

 

[gattispilot]

So I have done 1-5. But when I run a default DG scns it ctd.

; === ORBITER Master Configuration File ===
EchoAllParams = FALSE
LPadRect = 1138 325 1688 899

; === Visual parameters ===
EnableLocalLights = TRUE
AmbientLevel = 20
CSphereBgImage = Visible (DSS2/Wikisky)
CSphereBgPath = csphere\dds2
CSphereBgIntensity = 0.05

; === Visual helper parameters ===
Bodyforces = 188 1 1

; === Device settings ===
DeviceIndex = 0
ModeIndex = 0
NoVSync = TRUE
WindowWidth = 400
WindowHeight = 300

; === User interface parameters ===
MenubarMode = 0
SelVesselTab = 3

; === Window positions ===
LpadScnListWidth = 204
LpadModListWidth = 213
LpadExtListWidth = 213

; === Active plugin list ===
ACTIVE_MODULES
  ScnEditor
  LuaConsole
  LuaMFD
  InterMFD57
  Multistage2015_MFD
  CameraMFD
  CamControl
  Attachment Manager
  UnivPTG
  GPCMFD
  STSGuidanceMFD
  Glideslope
  D3D9Client
END_MODULES

 

[kuddel]

So I have done 1-5.

No you didn't!
You've installed several add-ons! Please be patient and do it one step after the other!
Any add-on could potentially break your setup. When installing several things at once no one can tell who's to blame.

By the way, unfortunately I've edited my procedure once ore twice, so you might not have gotten the "final" version sorry if that was the case

 

[gattispilot]

Thanks. I will take a look

---------- Post added 04-19-19 at 05:58 AM ---------- Previous post was 04-18-19 at 06:15 PM ----------

Still get the dxruntime error:

; === ORBITER Master Configuration File ===
EchoAllParams = FALSE
LPadRect = 539 240 1089 814

; === Visual parameters ===
EnableLocalLights = TRUE

; === Device settings ===
DeviceIndex = 0
ModeIndex = 0
NoVSync = TRUE
StencilBuffer = TRUE
WindowWidth = 900
WindowHeight = 562

; === User interface parameters ===
SelVesselTab = 3

; === Window positions ===
LpadScnListWidth = 204
LpadModListWidth = 213
LpadExtListWidth = 213

; === Active plugin list ===
ACTIVE_MODULES
  D3D9Client
END_MODULES

**** Orbiter.log
000000.000: Build Aug 28 2016 [v.160828]
000000.000: Timer precision: 6.83946e-007 sec
000000.000: Found 0 joystick(s)
000000.000: Module AtlantisConfig.dll .... [Build 160828, API 160828]
000000.000: Module AtmConfig.dll ......... [Build 160828, API 160828]
000000.000: Module DGConfigurator.dll .... [Build 160828, API 160828]
000000.000: ---------------------------------------------------------------
BaseDir    : C:\Orbiter2016\
ConfigDir  : C:\Orbiter2016\Config\
MeshDir    : C:\Orbiter2016\Meshes\
TextureDir : C:\Orbiter2016\Textures\
HightexDir : C:\Orbiter2016\Textures2\
ScenarioDir: C:\Orbiter2016\Scenarios\
000000.000: ---------------------------------------------------------------
000000.000: Module D3D9Client.dll ........ [Build 190406, API 160828]
000000.000: 
000000.000: **** Creating simulation session
000000.000: D3D9: [DirectX 9 Initialized]
            D3D9: 3D-Adapter.............. : AMD Radeon HD 8330
            D3D9: MaxTextureWidth......... : 16384
            D3D9: MaxTextureHeight........ : 16384
            D3D9: MaxTextureRepeat........ : 8192
            D3D9: VolTexAddressCaps....... : 0x3F
            D3D9: NumSimultaneousRTs...... : 4
            D3D9: VertexDeclCaps.......... : 0x3FF
            D3D9: MiscCaps................ : 0x3FCCF2
            D3D9: XNA Math Support........ : Yes
            D3D9: Vertex Texture.......... : Yes
            D3D9: Shadow Mapping.......... : Yes
            D3D9: D3DFMT_A16B16G16R16F.... : Yes
            D3D9: D3DFMT_A32B32G32R32F.... : Yes
            D3D9: D3DFMT_D32F_LOCKABLE.... : Yes
            D3D9: D3DFMT_A2R10G10B10...... : Yes
            D3D9: D3DFMT_L8............... : Yes
            D3D9: D3DDTCAPS_DEC3N......... : Yes
            D3D9: D3DDTCAPS_FLOAT16_2..... : Yes
            D3D9: D3DDTCAPS_FLOAT16_4..... : Yes
            D3D9: Runs under WINE......... : No
            D3D9: Available Texture Memory : 4092 MB
000000.000: D3D9: [3DDevice Initialized]
D3D9: ERROR: D3D9Surface.cpp Line:2045 Error:-2147024770 D3DXCreateEffectFromFileA(pDev, name, 0, 0, 0, 0, &FX, &errors)
D3D9: ERROR: Failed to create an Effect (Modules/D3D9Client/CKBlit.fx)
D3D9: ERROR: D3D9Pad.cpp Line:87 Error:-2147024770 D3DXCreateEffectFromFileA(pDev, name, 0, 0, 0, 0, &FX, &errors)
D3D9: ERROR: Failed to create an Effect (Modules/D3D9Client/Sketchpad.fx)
D3D9: ERROR: D3D9Effect.cpp Line:341 Error:-2147024770 D3DXCreateEffectFromFileA(pDev, name, macro, 0, D3DXSHADER_NO_PRESHADER|D3DXSHADER_PREFER_FLOW_CONTROL, 0, &FX, &errors)
D3D9: ERROR: Failed to create an Effect (Modules/D3D9Client/D3D9Client.fx)
D3D9: ERROR: PlanetRenderer.cpp Line:193 Error:-2147024770 D3DXCreateEffectFromFileA(pDev, name, macro, 0, 0, 0, &pShader, &errors)
D3D9: ERROR: Failed to create an Effect (Modules\D3D9Client\Surface.fx)
D3D9: ERROR: Scene.cpp Line:2997 Error:-2147024770 D3DXCreateEffectFromFile(pDev, name, 0, 0, 0, 0, &FX, &errors)
D3D9: ERROR: Failed to create an Effect (Modules/D3D9Client/SceneTech.fx)

 

[jarmonik]

If I recall errors like that are related to missmatch of DirectX-runtimes. Official distributions are here and the link to required runtimes [June-2010] http://users.kymp.net/~p501474a/D3D9Client/


In a case of non-official builds the required runtimes may wary. The latest runtimes package doesn't contain older runtimes against which the client may have been linked.

 

[gattispilot]

Thanks That did it. Got 3.8 running. I was using the link in the doc.

 

[Ripley]

... I was using the link in the doc.

Time to update it then.

 

[kuddel]

Time to update it then.

They have been (2017-08-31). I believe the 2016-branch didn't get that merge.

Everyone that finds an error in the documents, please tell us! Documenting an ever changing environment is not without flaws.

Nevertheless, this Runtimes issue comes up every couple of weeks and the forum is full of hints about this.

---------- Post added at 14:24 ---------- Previous post was at 13:19 ----------

Merged some "logging features" from trunk to 2016-branch:

Merged revision(s) 861,1042-1047 from trunk:
- Documentation update
  - updated DirectX runtimes URL to link to the current (June 2010) version.
- fixed incorrect report of spacecraft.dll used
  when starting another scenario without closing the Launchpad dialog.
- Added logging of loaded D3D9 DLLs and their versions.
- added EditorConfig file
  Visual Studio 2017 supports this natively, Visual Studio 2015 might need a plugin
  (see https://EditorConfig.org)

I did expect that to have happened already :facepalm:

 

[patio]

gcSetupCustomCamera shadow

hello

I am using gcSetupCustomCamera in an MFD

it works but I don't get vessel shadows

Is there something to do to have shadows activated ?

Thanks

 

[jalexb88]

I have a minor issue with the self-shadows: When I set the terrain mapping to projected this works very well for the most part, except that the shadow is not visible from internal views. This makes it quite challenging for scenarios such as landing the LM on the moon as one relies a lot on the shadow for height perception. Using stencil mode is the only way right now to see your shadow from internal views.

 

[Antares14]

DX9 to Vulkan wrapper

Don't know if you guys would take an interest in this but I came across a wrapper called VK9 which apparently runs DirectX 9 applications on Vulkan. Given by what I've read on other forums, it's to help increase frame rate, etc. I installed the latest pre-release available on Orbiter and it didn't have any conflicts. However, I can't tell if it's a) running and b) if there's any improvement to Orbiter as I have an GTX 1080 and Orbiter always runs smoothly, even when I push it at DSR.

Anyway, link is below if you want to check it out:

https://github.com/disks86/VK9

 

[DaveS]

- Cloud shadow intensity over water has been increased.

Where is this set? I'd like to tweak it for myself.

 

[jarmonik]

Where is this set? I'd like to tweak it for myself.

It was in the shader files but since it didn't really worked very well it got disabled. Line 845 in Surface.fx. The real problem is the blue atmospheric haze that makes the shadows less visible against dark blue water. Maybe the haze is too thick ?

There is also "OrbitalShadowMult = 0.85" parameter added in D3D9Client.cfg that will control ground shadow intensity.

//float fBlue = saturate(cTex.b - (cTex.g + cTex.r)*0.5f)*3.0f + 0.5f;
fShd = saturate(1.0 - fShd*fAlpha); // *fBlue);

---------- Post added at 16:49 ---------- Previous post was at 16:39 ----------



New Build of the Client Is Out

The name of the beta build is messed up for some reason but it's supposed to be for r.88 of the Orbiter.

- Vessel ground shadows should be now visible from the cockpit view.
- Some other ground shadow fixes been made. Overlapping stencil shadows shouldn't be as common anymore but can still occur with very large structures.
- Planet night lights intensity fixes has been made and the runway lights should be brighter.
- An attempt is made to address the VC over-saturation issue in SSU. (Haven't been able to test it) (feedback required)
- There is also "OrbitalShadowMult = 0.85" parameter added in D3D9Client.cfg to control cloud shadow intensity (you may need to run the client once for the line to appear)

 

[DaveS]

- An attempt is made to address the VC over-saturation issue in SSU. (Haven't been able to test it) (feedback required)

Thanks for this and things are much better now, everything is perfectly usable again, no over-saturation anywhere.

 

[clipper]

Is there any update on the smaller sized MFD font issue I posted about a while back? It's not a major concern but I just wanted to point it out as it still remains the way it used to be even with the latest D3D9 build and GPU drivers (I also had an OS reinstallation in the meantime and the issue persisted with a clean W10 install).

 

[jarmonik]

Is there any update on the smaller sized MFD font issue I posted about a while back? It's not a major concern but I just wanted to point it out as it still remains the way it used to be even with the latest D3D9 build and GPU drivers (I also had an OS reinstallation in the meantime and the issue persisted with a clean W10 install).

I'll forward this to Martin.

 

[DaveS]

I think I have discovered a bug in the D39DClient's handling of the UNDERSHADOWS base element parameter. I was playing around with the SSU Canaveral.cfg file and when I applied the UNDERSHADOWS parameter to the MESH elements, I got what is shown in the first attached screenshot. When I removed UNDERSHADOWS from the VAB MESH entry, I got the proper looking second screenshot.

 

[jarmonik]

I think I have discovered a bug in the D39DClient's handling of the UNDERSHADOWS base element parameter.

The UNDERSHADOWS is supposed to be applied to a flat runways and landing pads. The object will be rendered without z-buffer beneath/below any other building or vessel. So, it looks ok to me.

 

[kuddel]

The name of the beta build is messed up for some reason but it's supposed to be for r.88 of the Orbiter.

My fault ... :blush:

That's fixed now with the new version of the over.exe tool. It used to look at the "old" Orbiter repository path to figure out the revision of Orbiter.lib.
So after an update[1] the build script should work again and identify the version to be r88!

[1] Also merged into branches "2010-P1" and "2016"

 

[DaveS]

Is there any work on going to improve the atmosphere rendering? I've tried my best to get the current implementation pretty accurate but it's just not getting there, no matter how much I play with the sliders in the Atmospheric Controls dialog, I'm not getting anywhere close to an accurate color.