New Release D3D9Client Development

[N_Molson]

Tested, excellent ! :thumbup:

 

[rodion_herrera]

Just tested RC39 with my Brighton Beach and the results are just stunning. 4X antialias with framerates always above 130 FPS, and this is just an old nVidia 8800GT Now this thing is making me read up on Tangent Space norm maps, and getting tempted to put norm maps in my Brighton and I really shouldn't or else I'll spend ages making norm maps for all the textures LOL

-RODION

 

[flyer]

Energia is indeed working now for me.

The HUD requires the GDI compatibility mode to be selected. This then results in low framerates when in the cockpit view only (external view framerates remain high).

I understand that this is to be expected and I thought I would just mention it for your information.

Thanks again for your help!

Flyer

 

[David413]

RC39 - Perfect!

Thanks for this; I'm finishing up the next Shuttle Fleet and it works great with this release.

Dave

 

[Cras]

Energia is indeed working now for me.

The HUD requires the GDI compatibility mode to be selected. This then results in low framerates when in the cockpit view only (external view framerates remain high).

I understand that this is to be expected and I thought I would just mention it for your information.

Thanks again for your help!

Flyer

Do you see this issue?
http://www.orbiter-forum.com/showthread.php?t=25336

---------- Post added at 06:44 PM ---------- Previous post was at 06:44 PM ----------

Thanks for this; I'm finishing up the next Shuttle Fleet and it works great with this release.

Dave

Oh man, now I am getting excited, looking forward to it.

 

[astrosammy]

Oh man, now I am getting excited, looking forward to it.

:hesaid:

The low frame rates with Energia are normal for most computers with Windows Vista and 7, the workaround is to disable "Transparent indicator" in Extra/Vessel configuration/Energia configuration.

 

[jarmonik]

Do you see this issue?
http://www.orbiter-forum.com/showthread.php?t=25336

I read the report and I presume the CTD is fixed. There is nothing I can do about the autopilot failure. It's probably some kind of timestep/physics anomaly in the Orbiter_ng. The Energy or energia is unlikely the only one having the problem. My hands are already full of work with the client so I would be very happy if someone could try to reproduce the results and after that it should be filed in the official bug tracker.

EDIT: Of course, the problem could be caused by a bad autopilot and in that case there is not much Martin can do about it. But since there exists the braking performance problem as well it would be good to pring this issue in Martin's attention as well, since, they could be connected.

 

[DaveS]

I am not exactly sure what you mean by reflection maps. I presume you mean that there is a spherical environment map around the vessel and the color of the reflected light would be taken from the environment map.

If it is something like this for HST, then yes.

 

[Cras]

I read the report and I presume the CTD is fixed. There is nothing I can do about the autopilot failure. It's probably some kind of timestep/physics anomaly in the Orbiter_ng. The Energy or energia is unlikely the only one having the problem. My hands are already full of work with the client so I would be very happy if someone could try to reproduce the results and after that it should be filed in the official bug tracker.

EDIT: Of course, the problem could be caused by a bad autopilot and in that case there is not much Martin can do about it. But since there exists the braking performance problem as well it would be good to pring this issue in Martin's attention as well, since, they could be connected.

I confirm the CTD is fixed, and I am thinking that the Energy project itself shows some rotten parts under the conditions I am running at. I tried the same steps to limit the frame rate that you told me about with the landing bug, and they did not have any effect.

The thing is just odd, because the Energy just has enourmous amounts more thrust, and it just jumps off the pad and gets itself into a position that the autopilot just can't deal with.

I don't know if it is thrust, or somehow Energy's weight is not properly being considered, or what is going on. I firmly believe D3D9 is not at fault here, but Energy just acting badly in the NG enviornment.

 

[flyer]

Cras:

I'm not ignoring you, I have to install the XR2 to test out this bug on my machine, but I will post my results once I have them :thumbup:

 

[Evil_Onyx]

I have uploaded the RC39 in the first post.

- Energia CTD should be fixed.
- Surface base shadow offset issue should be fixed.

Using RC39 i am getting the offset issue, but using RC38 i don't.
I have reinstalled Orbiter for other reasons and the problem still persists using RC39. I will post screen shots of it later if required.

I am running in GDI compatibility on and windowed

---------- Post added at 23:07 ---------- Previous post was at 22:49 ----------

Just tried it with and without GDI in fullscreen with the same result.

 

[Felipi1205]

I need help!
When I'm loading the scenatrio file, my Orbiter closes when loading the earth_cloud.tex. Why?

 

[jarmonik]

Using RC39 i am getting the offset issue, but using RC38 i don't.

Yes, that's possible I suppose. There is a flag "MapObjectsToSphere" in a base configuration file. If it's set to "TRUE" then the shadows will wotk in RC39 and if it's set to "FALSE" then the shadows will work in RC38. Unfortunately, I am not aware of any means to read the state of the flag so I can't choose a proper mapping method for the shadows.

---------- Post added at 02:50 ---------- Previous post was at 02:45 ----------

I need help!
When I'm loading the scenatrio file, my Orbiter closes when loading the earth_cloud.tex. Why?

- Go in the advanced setup menu and set the debug level to "2".
- Run a scenario so that it closes/crashes.
- Zip the "D3D9ClientLog.html" that's located in /Modules/D3D9Client/ and post it here.

earth_cloud.tex could be old or corrupt, so, you could try to download and install latest files.

 

[Felipi1205]

(318: 135.71s 1133163us)(0x910) RegisteringVisual (Earth) hVessel=0x0, hObj=0x180D340, Vis=0x6050E58, Rec=0x6014F40, Type=4
(319: 135.77s 50858us)(0x910)[ERROR] Exception Code=0xC0000005, Address=0x681D329A
(320: 135.77s 50931us)(0x910)[ERROR] EAX=0x0000000A EBX=0x0923CBA0 ECX=0x00168142 EDX=0x091BDBA0 ESI=0x045AA9C0 EDI=0x00000000 EBP=0x009FF0A8 ESP=0x009FF06C EIP=0x681D329A
(321: 135.78s 54409us)(0x910)[ERROR] C:\Windows\system32\d3d9.dll EntryPoint=0x68041811, Base=0x68040000, Size=1847296
(322: 135.78s 59036us)(0x910)[ERROR] Exception in clbkRenderScene()
(323: 135.79s 61690us)(0x910)[ERROR] !!! Abnormal Program Termination !!!
(324: 135.79s 61773us)(0x910) --------------ExitModule------------
(325: 135.79s 62146us)(0x910) Log Closed

 

[jarmonik]

The CTD seems to occur in the core D3D9 library. There is nothing that would indicate it has anything to do with the Earth or the Clouds. Complete log would have been more useful.

- Does the problem occur in clean installation ?
- What addons/scenario are you using ?
- What kind of hardware you have ? Hardware specs ?

- Install latest drivers

 

[Felipi1205]

-I'm using the latest drivers...
-My hardware isn't good at all, but he can do it!
-This bug happens with any scenario!
-I never tried with a clean installation, maybe I'll try...

 

[Turbo]

-I'm using the latest drivers...
-My hardware isn't good at all, but he can do it!
-This bug happens with any scenario!
-I never tried with a clean installation, maybe I'll try...

It would help him more, if you gave him some specific computer specs. (processor / OS / video card / etc)
Also, always try addons (including graphics clients) in a fresh install when getting problems to see if they are working correctly. Some addons will not work 100% correctly with certian other addons.

 

[jedidia]

I ran over a bug I cannot quite pin down while testing Orbiter Galaxy with D3D9 client. The bug is not directly related to orbiter galaxy, it could be a general problem with custom systems in D3D9, or it could be a lack of information on my part.

Anyways, the deal is this: In the systems I'm exporting, (most) planetary textures don't show up on planets when running Orbiter in D3D9. They do show up in Vanilla Orbiter, however.

The problem is not related to the textures themselfes: I can rename the textures to planets of the solar system, load, and they show up. Equally, I can rename default orbiter textures to fit to planets in my custom system, and they don't show up. Here are simplified configs of the system (only one planet remaining, all you need to reproduce):

359.7h0.1g103.9fSystem.cfg:

; === Configuration file for the 359.7h0.1g103.9f solar system ===
Name = 359.7h0.1g103.9f System

Star1 = OGalaxySystems/359.7h0.1g103.9f/359.7h0.1g103.9f
Planet1 = OGalaxySystems/359.7h0.1g103.9f/359.7h0.1g103.9f a

359.7h0.1g103.9f.cfg:

; === Configuration file for star 359.7h0.1g103.9f ===
Name = 359.7h0.1g103.9f
Module = Sun
ErrorLimit = 1e-6
SamplingInterval = 1497
Mass = 1.59114e+030
Size = 7.3776e+008

359.7h0.1g103.9f a.cfg:

; === Configuration file for Planet 359.7h0.1g103.9f 0
Name = 359.7h0.1g103.9f a
EllipticOrbit = TRUE
HasElements = TRUE

; === Orbital Parameters === 
Epoch = 2000
SemiMajorAxis = 5.26156e+010
Eccentricity = 0.0506015
Inclination = 0.0155404
LongAscNode = 5.55839
LongPerihelion = 1.98133
MeanLongitude = 1.59189

; === Physical Parameters === 
Size = 1589166.271982
Mass = 89440431855634733000000.000000

; === Rotation Elements === 
SidRotPeriod = 7359527.230513
SidRotOffset = 0 
Obliquity = 1375.095493
LAN = 1

; === Visualization Parameters === 
MaxPatchResolution = 8
MinCloudResolution = 1
MaxCloudResolution = 8

I posted these in text so you can take a look at it, in case it might have something to do with the configs.
If you want to reproduce, there's a zip attached. Drop it into your Orbiter folder, take any default texture and rename it to "359.7h0.1g103.9f a.tex" and start TestScn. In vanilla Orbiter you should see the planet with the texture you renamed, in D3D9 Orbiter you should see the planet black. If not, then the issue is even more complicated than I thought.

 

[jarmonik]

I ran over a bug I cannot quite pin down while testing Orbiter Galaxy with D3D9 client. The bug is not directly related to orbiter galaxy, it could be a general problem with custom systems in D3D9, or it could be a lack of information on my part.

Anyways, the deal is this: In the systems I'm exporting, (most) planetary textures don't show up on planets when running Orbiter in D3D9. They do show up in Vanilla Orbiter, however.

I was able to reproduce the bug and it will be fixed in RC40. It was caused by a black cloud layer. If you disable clouds from the visual effects tab you should see the planet properly.

It seems that oapiGetObjectParam (hObj, OBJPRM_PLANET_HASCLOUDS) returns true for the planet even if it doesn't have a clouds.

 

[Astronut25]

Glowing Spacecraft

Sorry about the delay in response.

I noticed that most objects were no longer glowing, but some were, and after much searching I think I've narrowed it down to the normal textures, if I remove the normal texures from the items that are glowing, they will show correctly (however, some show correctly with them anyway)

this one shows the only exception (the Shuttle ET), and the inflatable module without normals

inflatable modules with normals