SSU Development thread (4.0 to 5.0) [DEVELOPMENT HALTED DUE TIME REQUIREMENTS!]

[Wolf]

The glass OV I've seen before, and if I remember it had to do with the GC file settings... DaveS handles that so he should know.

The shadows on panel C3, that's news to me... can you check the other panels for that? If it's only C3, can you go back to r3018 and see if it still happens?

Only C3 is affected.
How can I revert back to 3018?

 

[GLS]

Only C3 is affected.
How can I revert back to 3018?

Right-click >> TortoiseSVN >> Update to revision...

Just checked the mesh and those shadows/lights are on triangles, but the normals all point in the same direction so lighting should be uniform. :shrug:

 

[Wolf]

Ok, the glossy SSU was due to a bad install of the latest D3D9 GC revision (fixed now)
As far as the C3 panel issue I think it might be related with something at my end since I have also some weird visual effect on the SRBs sometimes (dependin on the lighting conditions).
I'll post some pics later

 

[DaveS]

The glass OV I've seen before, and if I remember it had to do with the GC file settings... DaveS handles that so he should know

Checked in an updated GC cfg file that will fix this.

 

[GLS]

Slidewires are now animated!
The structure currently collides with the PLBD handrails. Checked the diagrams and things don't seem to be in the correct place, and also the PLBDs axis isn't correct.

Updated ticket 177 and added these issues there.

---------- Post added at 04:56 PM ---------- Previous post was at 04:06 PM ----------

Also, the RMS_EE_VERN_TRANSLATION_SPEED const was changed to 0.61... where did it come from? I'm asking because it isn't a "round" number.

Is this and ticket #193 related?
What is the explanation to close the ticket?

 

[DaveS]

What is the explanation to close the ticket?

It isn't a bug as seems to be mission-dependent, hence the discrepancy I noted.

 

[GLS]

It isn't a bug as seems to be mission-dependent, hence the discrepancy I noted.

And the RMS_EE_VERN_TRANSLATION_SPEED change?

Are you working SLC-6? I just finished cleaning the RMS and was thinking of doing the SLC-6 meshes next.

 

[DaveS]

And the RMS_EE_VERN_TRANSLATION_SPEED change?

I don't know know where it came it from. I didn't even know about it until you pointed it out. The only RMS-related things I have ever worked on is the animations, never the actual mechanics. And no, I'm not working on anything besides the orbiter right now.

---------- Post added at 07:30 PM ---------- Previous post was at 07:02 PM ----------

Speaking of SLC-6: Do you have the same terrain corruption as I am there? I just want to know if I should file an official bug ticket about it since it seems that something has changed in the post-2016 betas when it comes to modified terrain that is cause of the terrain corruption.

 

[GLS]

Speaking of SLC-6: Do you have the same terrain corruption as I am there? I just want to know if I should file an official bug ticket about it since it seems that something has changed in the post-2016 betas when it comes to modified terrain that is cause of the terrain corruption.

Yes, the pad is hovering... :facepalm:

 

[DaveS]

Yes, the pad is hovering... :facepalm:

Thanks for confirming this. I've filed an official bug ticket regarding this as it seems to be something going on core-wise and out of our hands.

 

[DaveS]

GLS: Your presence is requested in the official terrain bug ticket: https://www.orbiter-forum.com/project.php?issueid=1408 as you were the one who did the terrain modifications as the tileedit tool refused to work for me.

 

[GLS]

GLS: Your presence is requested in the official terrain bug ticket: https://www.orbiter-forum.com/project.php?issueid=1408 as you were the one who did the terrain modifications as the tileedit tool refused to work for me.

Sorry I missed the party, but I was sleeping.:lol:

---------- Post added at 10:31 AM ---------- Previous post was at 08:16 AM ----------

On the hidden triangles in SLC-6: the TSM umbilicals and piping cost almost 100k triangles! Which IMO is a bit too much, especially when most of it is hidden inside the TSMs. Just deleting the hidden triangles cuts about 10% of the original size (at no loss for detail), but we could delete the sections of pipes that run after the "back support" (a bend and vertical section), and much more could be gained. This could be done in time for the release. In the future, the pipes could probably be simplified: a slight decrease in their curving (so less sections) and also decreasing their detail by using, e.g., 10 instead of 20 vertices around.

Again for the record, I'm not against detail. But this alone costs almost as much as the entire OV, and it's small and mostly hidden.

 

[Urwumpe]

Again for the record, I'm not against detail. But this alone costs almost as much as the entire OV, and it's small and mostly hidden.

Also, some smaller details could be done as textures, since we really rarely get close enough to notice the difference.

 

[GLS]

Ok, after losing a good chunk of the changes, because half-way thru saving the mesh blender decided it would write gibberish, and moments later my pc when kaput, I got half of the work done on the SLC-6_STS_LaunchMount.msh file.
Results so far:
>> original
v 114920
t 142321

>> current WIP with all groups cleaned except LOX umbilical (which is on the LH2 side :uhh
v 97747 (-14.94%)
t 125261 (-11.99%)

If 10-15% isn't impressive, then 20000 vertices and triangles is. And probably there is still another 20K inside the other TSM.

 

[DaveS]

I've managed to reduce the LOX TSM umbilical triangle count by 84.8%. So down from 58,791 triangles to 8,947 triangles. The vertex count is down by 71.4% from 27,128 to 7,762. Note these numbers may not be correct once exported to the msh format.

I'm sure I could get the LH2 umbilicals down a similar amount.

 

[Donamy]

If you are reducing in AC3D. When you export as .msh the vertices stay welded. When you import to AC3D is when the vertices get unwelded for some reason.

 

[GLS]

I've managed to reduce the LOX TSM umbilical triangle count by 84.8%. So down from 58,791 triangles to 8,947 triangles. The vertex count is down by 71.4% from 27,128 to 7,762. Note these numbers may not be correct once exported to the msh format.

I'm sure I could get the LH2 umbilicals down a similar amount.

Good! :thumbup:
So there is no point in me continuing the work in the umbilicals at SLC-6, right? :shrug:
I'll commit it anyway tonight, as there were changes in other groups, and then I'll move to the MST, as the grid structure there seems to have fat.

 

[Wolf]

I think there are still some issues with the latest mesh.

1) There is an area located on the upper wing close to the leading edge where the texture looks wrong. This is difficult to detect when viewing Atlantis/Discovery or Endeavour but it is quite evident looking at Columbia and Challenger original since they have white tiles with lettering in that area.


2)There is also a problem where the black metal strip inside of the RCC merges with the wing white tiles.


3) Last is the old Side Hatch issue which was fixed time ago but now shows up again.

 

[Donamy]

I think there are still some issues with the latest mesh.

1) There is an area located on the upper wing close to the leading edge where the texture looks wrong. This is difficult to detect when viewing Atlantis/Discovery or Endeavour but it is quite evident looking at Columbia and Challenger original since they have white tiles with lettering in that area.
View attachment 16590

2)There is also a problem where the black metal strip inside of the RCC merges with the wing white tiles.
View attachment 16591

3) Last is the old Side Hatch issue which was fixed time ago but now shows up again.
View attachment 16592

Needs some fine-tooth remapping.

 

[GLS]

In image 2, it's the gear wheel well poking out, and I noted that in the OV mesh issues ticket.