Project D3D11Client Development

[asmi]

Screenshots were taken after the light stopped adapting. And stars are always like in those shots, even if I look towards the sun. It doesn't care about my stars settings in "Parameters".

OK, I will check it out tonight. Will let you know my findings.

 

[dumbo2007]

Well it works now. I did that step of making server/orbiter.exe run with the gpu. The only thing is orbiter_ng starts up ok and then I need to unload and reload d3d11client module at least once. After that it starts up ok with 4x MSAA and post processing effects.

With post processing effects i get frame rates of abut 30, without I touch about 200 to 300. I am sure the NVidia card is turning on now, because when I specifically choose intel then with post processing effects and MSAA, I get extremely low fps while with nvidia it runs smooth.

So now that I am not going to return my laptop to amazon .....lets see why we need to unload/reload this client. Is it because when the client is loaded once it saves some setting which are not removed even if Orbiter_ng is shutdown ? Could it be that some dll is loaded which is not unloaded by windows till long after orbiter_ng is shut down ?

There is also this wierd error with texture colors that always happens. It results in the landing pads being completely dark, or the textured models being sme random color probably due to a texture color issue. The client runs ok the first time, no color problems. If I exit and go back to launch pad and run another scenario then all these colors show up on the textures :





Goes away if I go back to the launchpad and unload/reload the client on modules tab. Then after I load a scenario, I see the green textures for a split second before they switch to their normal grey colors. Kind of makes me think that they are getting stored in video memory and are not cleared from there or something.



Also I can see that as Jarod reported, the lighting ramps up initially with post processing effects enabled, but quickly ramps down and settles. Now if I am close to the surface and looking almost parallel to it, the brightness of the surface increases. Its probably some bug in the shader lighting calculation for the surface. Like as my angle with the surface of the moon decreases(i.e. I look increasingly parallel to it, towards the horizon) and I am close to the ground, the brightness always goes up.

By the way, if you are ok with remotely accessing my laptop, then I can setup the environment on my laptop and you can debug the crash on my machine anytime during the weekend . I have got visual c++ express on my laptop and I can get your code and setup a debug build. I have also got the directx sdk june 2010 installed and a fast connection. I have a huuuuge ship in mind and I am really banking on the client with all possible bump maps to make it look good, especially the bridge and lower decks

 

[buznee]

CTD for me with clean install.

I was wondering if someone could help me.

I am currently running Windows 7 64 bit. Nvidia Geforce 470 GTX. Latest nVidia drivers, downloaded and installed directx_Jun2010_redist.exe.

Installed a clean copy of orbiter100830.zip, updated it to 111105 with the file orbiter111105-100830diff BETA.7z

Installed asmi-ovp-5c0b74564ccf.zip

Opened Orbiter_ng.exe, selected D3D11Client for the graphics engine under modules. When I click on Video, I get a CTD.

The following on the log file.
"**** Orbiter.log
Build Aug 30 2010 [v.100830]
Timer precision: 3.01755e-007 sec
Found 1 joystick(s)
Module AtlantisConfig.dll .... [Build 100830, API 100830]
Module AtmConfig.dll ......... [Build 100830, API 100830]
Module DGConfigurator.dll .... [Build 100830, API 100830]
Module D3D11Client.dll ....... [Build 120419, API 111105]
Error loading module Modules\Plugin\D3D11Client.dll (code -1073741674)"

I noticed that the build shows up as 100830 for some modules. Any help would be appreciated.

 

[asmi]

I was wondering if someone could help me.

I am currently running Windows 7 64 bit. Nvidia Geforce 470 GTX. Latest nVidia drivers, downloaded and installed directx_Jun2010_redist.exe.

Installed a clean copy of orbiter100830.zip, updated it to 111105 with the file orbiter111105-100830diff BETA.7z


The following on the log file.
"**** Orbiter.log
Build Aug 30 2010 [v.100830]
Timer precision: 3.01755e-007 sec
Found 1 joystick(s)
Module AtlantisConfig.dll .... [Build 100830, API 100830]
Module AtmConfig.dll ......... [Build 100830, API 100830]
Module DGConfigurator.dll .... [Build 100830, API 100830]
Module D3D11Client.dll ....... [Build 120419, API 111105]
Error loading module Modules\Plugin\D3D11Client.dll (code -1073741674)"

I noticed that the build shows up as 100830 for some modules. Any help would be appreciated.

I've marked things that are wrong here. Looks like your Orbiter didn't get updated after all...

Here is a log file like it should be:

**** Orbiter.log
Build Nov 5 2011 [v.111105]
Timer precision: 4.46656e-007 sec
Found 0 joystick(s)
Module AtlantisConfig.dll .... [Build 111105, API 111105]
Module AtmConfig.dll ......... [Build 111105, API 111105]
Module DGConfigurator.dll .... [Build 111105, API 111105]
Module AeroBrakeMFD.dll ...... [Build ******, API 100830]
---------------------------------------------------------------
>>> WARNING: Obsolete API function used: oapiRegisterMFDMode
At least one active module is accessing an obsolete interface function.
Addons which rely on obsolete functions may not be compatible with
future versions of Orbiter.
---------------------------------------------------------------
Module OrbiterSound.dll ...... [Build 120328, API 100830]
Module ScnEditor.dll ......... [Build 111105, API 111105]
Module transx.dll ............ [Build 100905, API 100905]
Module RendezvousMFD.dll ..... [Build ******, API 050206]
Module BaseSyncMFD.dll ....... [Build 100616, API 100603]
Module InterMFD55.dll ........ [Build 100826, API 100704]
Module LuaMFD.dll ............ [Build 111105, API 111105]
Module D3D11Client.dll ....... [Build 120420, API 111105]

 

[buznee]

Thanks asmi, is this the correct file to update orbiter to the correct version? orbiter111105-100830diff BETA.7z

 

[asmi]

Thanks asmi, is this the correct file to update orbiter to the correct version? orbiter111105-100830diff BETA.7z

I was actually doing an incremental upgrades myself (I've downloaded all diff files and applied them one by one). As a general rule, before trying to run D3D11Client please make sure stock client works fine - because if it doesn't - there is no way NG version will.
Your log clearly tells that it wasn't properly upgraded, so please try again (fresh install, upgrade to P1 (that is, 100830), finally aforementioned diff. Try running stock client and check log file. If it still says 100830, you've gotta apply all incremental upgrades one by one in proper order.

Let me know if any of that helped, or if you run into some other issues...

 

[buznee]

Apparently Winrar doesnt like working with 7z type of files. Downloaded 7-zip and it extracted correctly. Now it works =)

 

[asmi]

Apparently Winrar doesnt like working with 7z type of files. Downloaded 7-zip and it extracted correctly. Now it works =)

Awesome! :thumbup:

 

[asmi]

Below is another preview of what's coming - terrain & atmosphere rendering. It's just behind the corner
All credit for this goes to Glider!

 

[Cras]

that is awesome.

 

[Mattyv]

Nice work!:thumbup:

 

[Frogisis]

Amazing! I'm crazy excited for a release candidate.

 

[Glider]

Below is another preview of what's coming - terrain & atmosphere rendering. It's just behind the corner
All credit for this goes to Glider!

4 more previews of normal mapped Earth from orbit and sunsets/sunrises rendered with recently implemented E.Bruneton's atmospheric scattering:

 

[dumbo2007]

Awesome work both of you, cant wait


So I guess the header for declaring the functions used to get the altitude at any place is still present ?

 

[Frankynov]

That's very beautiful !!

How is the framerate ?

 

[asmi]

How is the framerate ?

There is FPS on my screenshots, but please ignore first one as it was made in debug build. I've got Alienware R17 M3 laptop with nVidia 460M video. As you can see it's in very comfortable 60+ range.

 

[dumbo2007]

Ah you got Alienware , I would probably have gone for it but for the same price of $900 I was getting i7 and nvidia 630m + 6 gb with asus

 

[asmi]

Ah you got Alienware , I would probably have gone for it but for the same price of $900 I was getting i7 and nvidia 630m + 6 gb with asus

I've got i7 and 16 GB RAM, along with 1.5Gb 460M and 3D-monitor And I've got unique Alienware's color backlighting, which can be controlled by software
The only problem - mine was ~$3700...

 

[dumbo2007]

So I have been wondering, the question which has been there in my mind all this while ...all this beautiful capabilities you guys have been adding, I guess, if somebody wanted to, then all the bump maps and stuff could also be used to really texture decks inside a ship

Like this maybe :

Oh no...another trek fanboy !!

is it possible to make the long extended lights and multiple lights for a scene ? I am trying to pull together a scene graph which can handle huge numbersof objects, allocating and deallocating them on the fly as the camera moves around.

 

[asmi]

So I have been wondering, the question which has been there in my mind all this while ...all this beautiful capabilities you guys have been adding, I guess, if somebody wanted to, then all the bump maps and stuff could also be used to really texture decks inside a ship

Like this maybe :

We CAN create an SDK for that, but it will be up to vessel devs to model/use SDK to make all this happen. Personally I think that would be pretty cool, and probably once I get some spare time I'll model something like that for default DG (since it's open source so I can modify it) :tiphat:

Oh no...another trek fanboy !!

is it possible to make the long extended lights and multiple lights for a scene ?

Absolutely possible!