[BB25187]

Hi all,

Thanks Carlo for the fix.

There is still something strange though.

If I go to the following passenger view with Key 5, I get this sorting of materials.
http://imageshack.com/a/img163/2987/lw61y.jpg

Then I press key 8 to switch to the free camera: I get something different (similar to what we got before #2131 and with MSTS). So the sorting algorithm seems to be different with the different cameras now.
http://imageshack.com/a/img706/3491/svly.jpg

Regards
Vincent

[James Ross]

BB25187, on 14 April 2014 - 12:04 PM, said:

If I go to the following passenger view with Key 5, I get this sorting of materials.

Then I press key 8 to switch to the free camera: I get something different (similar to what we got before #2131 and with MSTS). So the sorting algorithm seems to be different with the different cameras now.

Firstly, the interior shape you see from camera 5 isn't necessarily the normal vehicle model. When you switch to another camera, we immediately flip back to the normal model (instead of the interior one). That alone could be the reason for the difference, but additionally the near-plane of camera 5 is closer than other cameras to allow you to get a better view of the interior, so the exact rendering behaviour will be different because of the different near planes too.

[Csantucci]

James,
I did not get crashes with your patched runactivities. I tested both critical trainsets with both the standard and LAA runactivity. No crash.

[James Ross]

Csantucci, on 14 April 2014 - 12:17 PM, said:

James,
I did not get crashes with your patched runactivities. I tested both critical trainsets with both the standard and LAA runactivity. No crash.

Great, thanks! I'll commit and close the bug then. :)

[Csantucci]

Thanks James,
that's nice.
Pls. check the attached patch referring to what BB25187 says: I overlooked to modify this when making the rain modification
 RenderFrame.cs.patch.zip (441bytes)
Number of downloads: 351

[James Ross]

Csantucci, on 14 April 2014 - 12:38 PM, said:

Pls. check the attached patch referring to what BB25187 says: I overlooked to modify this when making the rain modification
RenderFrame.cs.patch.zip

Ah, this is what's causing Vincent's issue? I'll pop that in now.

[James Ross]

Fixed in X2179.

[BB25187]

Hi,

James Ross, on 14 April 2014 - 12:16 PM, said:

Firstly, the interior shape you see from camera 5 isn't necessarily the normal vehicle model. When you switch to another camera, we immediately flip back to the normal model (instead of the interior one). That alone could be the reason for the difference, but additionally the near-plane of camera 5 is closer than other cameras to allow you to get a better view of the interior, so the exact rendering behaviour will be different because of the different near planes too.

Thanks for your reply.
In this case the shape used for passenger and external view is the very same.
I will recheck the materials used to build the different layers of "glass" and walls - it might happen that they were chosen to get something correct with MSTS, rather than according to rational criteria. In any case, a succession of semi-transparent layers should be visible from inside and outside. So, I hope the choice of the materials (in either one or two 3D) can rely on similar rules whatever the camera used to display them!

Regards

[Csantucci]

Vincent,
pls. try version X2179 and check if you still have the layers problem.

[BB25187]

Hi,

Marvelous!
Rev #2179 provides the same rendering of semi-transparent parts with cam 5 and 8. I noticed that very near planes are displayed differently in passenger view, as evoked by James. This is a good point for "imersion".
Besides this, the result is also similar to what we got with MSTS and with OR before the passenger view is handled in a specific way. So, I won't have to rework my 3D.
Great thanks to you and James for the fix in the repository! :)

Regards
Vincent