New Release D3D9Client Development

[Ripley]

...Are those users sharing the same desktop? If they deleted something from their own desktop, it's their fault. If many users share the same user profile on a computer, and the same desktop, it's administrator's fault.

We have one pc per user.

 

[n122vu]

Ok so my understanding of how Win7/Win8/UAC works was just a little off, but I think my diagnosis is still a plausible explanation, even if it isn't the right one.

From the technet article "UAC Processes and Interactions":

When an administrator logs on, two separate access tokens are created for the user: a standard user access token and an administrator access token. The standard user access token contains the same user-specific information as the administrator access token, but the administrative Windows privileges and SIDs are removed.

This is the part that I had wrong:

The standard user access token is used to start applications that do not perform administrative tasks (standard user applications). The standard user access token is then used to display the desktop (Explorer.exe). Explorer.exe is the parent process from which all other user-initiated processes inherit their access token. As a result, all applications run as a standard user unless a user provides consent or credentials to approve an application to use a full administrative access token.

A user that is a member of the Administrators group can log on, browse the Web, and read e-mail while using a standard user access token. When the administrator needs to perform a task that requires the administrator access token, Windows 7 automatically prompts the user for approval. This prompt is called an elevation prompt, and its behavior can be configured by using the Local Security Policy snap-in (Secpol.msc) or Group Policy.

I think I was also confusing the UAC prompt with doing a right-click, Run as Administrator. That is a separate issue that can be discussed further in another thread if anyone wants, and unless you're doing that or have Run as Administrator checked in the compatibility settings of the desktop shortcut, is not relevant to this problem. Sorry if I unintentionally muddied the water here.

 

[Cras]

FWIW, I was having some wicked issues with CTD upon loading the sim under R7. I find that when the sim isnt the active window it will crash during loading, and it just had a hard time making sure that was the case.

Since R10, I havent seen it once, so something seems different. Good for me, bad for Ripley.

 

[Ripley]

Me? Why? :shrug:
Never had CTDs or anything...(crossing fingers right now)

 

[Frankynov]

Sorry to trouble the party but I've met a CTD with Amso
It's not the first one. In general, it happens at some critical situations such as undocking the LEM+CSM from SIV, final undocking the LM from CSM around the moon...

Here is the orbiter.log :

D3D9Client: [Scene Initialized]
Orbiter Version 100830
D3D9Client Build [Mar 18 2013]
Exception Code=0xC0000005, Address=0x0044F425
EAX=0x0000000C EBX=0x000000EE ECX=0x00000000 EDX=0x00000012 ESI=0x109C7400 EDI=0x0000003C EBP=0x00000174 ESP=0x009FFA84 EIP=0x0044F425
C:\Users\Franky\Desktop\orbiter100830\modules\server\orbiter.exe EntryPoint=0x004ACFAC, Base=0x00400000, Size=2097152
Exception in clbkRenderScene()
!!! Abnormal Program Termination !!!

And D3D9 log in attachment
View attachment D3D9Client Log.zip

 

[Phoenix]

I have been getting an occasional CTD at scenario start up. Starting a scenario with pause enabled may fix the situation - it seems to with my computer.

 

[jarmonik]

I have been getting an occasional CTD at scenario start up. Starting a scenario with pause enabled may fix the situation - it seems to with my computer.

Are you using Windows 8 by any chnage ?

 

[Cras]

Me? Why? :shrug:
Never had CTDs or anything...(crossing fingers right now)

I dunno. Dunno why I said your name. Brain was throttled down I guess. I meant the other guy, not you.


Oops



Have a cold one on me.

 

[Phoenix]

Are you using Windows 8 by any chnage ?

No, Windows 7.

It might be a resource loading issue because it seems to only affect an installed version of Orbiter I have that has all the high resolution planetary textures. This version takes longer to start up and may be getting out if sync. with the rest of Orbiter. I understand multiple threads are used. Pausing at start-up gives the other parts of Orbiter time to catch up. Just a thought at what might be causing it.

Having said that, I have just started up several scenarios with the hires planetary textures - AMSO, DG-S, XR-2 - and without pausing, and they started fine. I don't know. It seems to be intermittent. It could be my computer running out of contiguous memory or something after it's been on a while.

Thanks.

 

[Felix24]

Here are some ideas for a slightly modified Fresnel integration:

cRefl.rgba = cRefl.rgba * (1.0f - fresnel) + !(gUseRefl && (cRefl.rgba == 0)) * fresnel;

This is the original:

cRefl.rgb = cRefl.rgb * (1.0f - fresnel) + fresnel;

Two things at work here- 1) using/changing cRefl.rgba instead of cRefl.rgb prevents over-saturation of the Fresnel reflection by keeping the levels consistent between the rgb and alpha channel, and 2) the term "!(gUseRefl && (cRefl.rgba == 0)) * fresnel" allows a reflection map, if it is defined, to mask off any Fresnel reflections along with ordinary reflections. This would be handy for gaps between windows in a single texture, or between adjacent solar panels.

 

[jarmonik]

Here are some ideas for a slightly modified Fresnel integration:

cRefl.rgba = cRefl.rgba * (1.0f - fresnel) + !(gUseRefl && (cRefl.rgba == 0)) * fresnel;

This is the original:

cRefl.rgb = cRefl.rgb * (1.0f - fresnel) + fresnel;

Two things at work here- 1) using/changing cRefl.rgba instead of cRefl.rgb prevents over-saturation of the Fresnel reflection by keeping the levels consistent between the rgb and alpha channel

I tried that and every surface that has a fresnel reflection are becoming completely black when viewed from a shallow angle. Why should I trade a minor saturation issue to even a greater problem ? Considering that there is the multiplier value that can be adjusted to remove the over-saturation.

2) the term "!(gUseRefl && (cRefl.rgba == 0)) * fresnel" allows a reflection map, if it is defined, to mask off any Fresnel reflections along with ordinary reflections. This would be handy for gaps between windows in a single texture, or between adjacent solar panels.

I need to think about that... Is it worth of breaking the 64 instruction boundry ?

 

[Felix24]

I tried that and every surface that has a fresnel reflection are becoming completely black when viewed from a shallow angle.

I'm not able to get it to do that. On orbit, when viewed from a shallow angle, the surface becomes dark, but that is the reflection of space. With any other background, it reflects properly, and does not darken too much.

It sounds similar to the alpha mipmap issue that you fixed earlier. Have you tried it in your newest version?

 

[romanasul]

Hello everyone, I don't know what the hell is going on with my computer but I'm getting crashes all the time when trying to launch thorton's Progress M1 to his Mir station. After spacecraft separation the screen flashes a few times and then freezes, followed by a windows saying that ALT+TAB is not supported. I have not ALT tabbed or pressed any key whatsoever. Someone please help this is really frustrating.

 

[Felix24]

This might be a dumb question, but does it crash in windowed mode as well as full-screen mode?

 

[asbjos]

Hello everyone, I don't know what the hell is going on with my computer but I'm getting crashes all the time when trying to launch thorton's Progress M1 to his Mir station. After spacecraft separation the screen flashes a few times and then freezes, followed by a windows saying that ALT+TAB is not supported. I have not ALT tabbed or pressed any key whatsoever. Someone please help this is really frustrating.

I don't know why it is happening, but as a workaround, you can go to the Video tab in the Launchpad, and switch Full Screen Mode from "True Full Screen" to "Full Screen Window".

 

[jarmonik]

On orbit, when viewed from a shallow angle, the surface becomes dark, but that is the reflection of space.

Fresnel reflection is not supposed to reflect black space like that, only metallic reflections are.

 

[n122vu]

Hey, jarmonik, just want to offer a well-deserved thanks for all your hard work on the client. Without it, the EAS Cortez would likely be unusable on many lower-end machines.

I've been getting back into development on the Cortez, and when I started up under D3D9client R10 I noticed some colors missing on the bridge dome. Then I realized there were some missing on the gravity wheel as well. These colors are defined in the material properties, so I'm not sure why they are not showing up. Here is a comparison of the inline client vs. R10:

Gravity wheel - inline client:

The same shot of the gravity wheel in R10:

Here are the material properties for these blue sections of the gravity wheel:

MATERIAL ColPansBlue
0.271 0.471 0.627 1
0.271 0.471 0.627 1
1.000 1.000 1.000 1

It only seems to be the blue sections that are affected. Other sections of the ship seem to be appearing with the correct color/shading/material properties. Any thoughts?

Thanks,
n122vu

 

[orb]

It only seems to be the blue sections that are affected.

I can see also red material missing at the far right and at left in inner section (most likely the same reason as with blue material in the center).

 

[n122vu]

Ah, I see that now. Confirmed just now in the client. Good catch.

 

[Notebook]

To Jarmonik and the others, thanks for the client. Makes a big difference. Also solved the cloud shadow problem I had:

Original:

http://i89.photobucket.com/albums/k207/Notebook_04/Westcott_Default_zps3356f8d4.png

D3D9R10 Client:
http://i89.photobucket.com/albums/k207/Notebook_04/D3D9R10_zps91489c39.png

N.