SSU Development Thread (2.0 to 3.0)

[GLS]

I would still prefer leaving mesh editing to the graphics department - they should know what they are doing.

I agree with you about fixing the top/down error, as that needs changes on 2 axis, and it must be done very well to keep the keyboard on the panel surface. What I'm proposing is subtracting a certain fixed amount from one of the axis (the Z axis), and because of the orientation of the panel, this would shift the keyboard aft, in a parallel plane to the panel surface. IMO it's safe, but if there are concerns then I'll just go do something else.

How about an new release candidate today?

Sounds good, but I think we should wait for the TAA (shouldn't be long now).

 

[Donamy]

I propose having 3 meshes.

1. TAA with hatch and cover
2. ExAirlock no ODS would have the covers and ring handrail
3. ISSAirlock would have the top cover and ring handrail removed. That way the ODS could be added, without changing it's mesh or animations.

 

[Urwumpe]

I agree with you about fixing the top/down error, as that needs changes on 2 axis, and it must be done very well to keep the keyboard on the panel surface. What I'm proposing is subtracting a certain fixed amount from one of the axis (the Z axis), and because of the orientation of the panel, this would shift the keyboard aft, in a parallel plane to the panel surface. IMO it's safe, but if there are concerns then I'll just go do something else.

Well, then just put my concern on the record: I think that such editing is, regardless how simple it is, an unnecessary error source. We have people who can do this, without having to edit numbers. Just one small typo there and we have new mesh errors. And it also takes you much more time to do this.

Sounds good, but I think we should wait for the TAA (shouldn't be long now).

Its OK, then I will delay it and work on the SSUWorkbench. XAML looked way easier when everything worked. Can somebody tell me why a "3*" column is smaller than a "1*" column?

 

[GLS]

I propose having 3 meshes.

1. TAA with hatch and cover
2. ExAirlock no ODS would have the covers and ring handrail
3. ISSAirlock would have the top cover and ring handrail removed. That way the ODS could be added, without changing it's mesh or animations.

If we need different meshes for the ExAL, couldn't the ODS already be attached to its ExAL? This way there would be just 2 meshes instead of 3.
BTW, for Mir and ISS dockings wasn't the ODS/ExAL the same? If so then it shouldn't be called "ISSAirlock" but maybe "ExAirlock_ODS" or something similar.

---------- Post added at 02:14 PM ---------- Previous post was at 02:04 PM ----------

Well, then just put my concern on the record: I think that such editing is, regardless how simple it is, an unnecessary error source. We have people who can do this, without having to edit numbers. Just one small typo there and we have new mesh errors. And it also takes you much more time to do this.

I put the idea forward as I think I could do it safely. It wasn't a complete fix (that I think could be unsafe to do manually), but it would look much better than it is now. It was simply a single-axis subtraction on 128 vertices, but as most share the same value, there would be just 5 different numbers to deal with.
But like I said, I'll not go against the will of the group, so this is a mute discussion now.

 

[Donamy]

The TAA is the 3rd mesh. If you attach the ODS to the airlock, then there would be a need, to update the animations. For Mir the blanket material for the ODS would have to be changed to orange.

You could call it StnAirlock

 

[DaveS]

The TAA is the 3rd mesh. If you attach the ODS to the airlock, then there would be a need, to update the animations. For Mir the blanket material for the ODS would have to be changed to orange.

You could call it StnAirlock

Another change between the two is the External Airlock truss. For the Mir missions it was outfitted with a second Trajectory Control Sensor (TCS) as well as two cameras, one for the Soyuz docking target on Kristall and one for the Buran docking target on the same module.

 

[GLS]

The TAA is the 3rd mesh. If you attach the ODS to the airlock, then there would be a need, to update the animations. For Mir the blanket material for the ODS would have to be changed to orange.
OK, do what you have to do.

You could call it StnAirlock

As in Standard Airlock?

 

[Donamy]

Station Airlock

 

[GLS]

Another change between the two is the External Airlock truss. For the Mir missions it was outfitted with a second Trajectory Control Sensor (TCS) as well as two cameras, one for the Soyuz docking target on Kristall and one for the Buran docking target on the same module.

Couldn't those extra parts be added to the Mir ODS mesh? That way they would only show when the Mir ODS was used.

EDIT: Oh wait, we don't have/need a Mir ODS mesh.... could it then be added as a separate group on the single ODS mesh then? We would show/hide that group as needed. And the ODS texture change for Mir could be done via code, right?

 

[Donamy]

So five meshes ?

1. TAA
2. ExAL
3. StnAL
4. ODS
5. MirODS

 

[DaveS]

Here's the best photo I have managed to find that shows the additional TCS as well as the two cameras on the starboard side of the truss:

 

[GLS]

So five meshes ?

1. TAA
2. ExAL
3. StnAL
4. ODS
5. MirODS

No, I just realized it wasn't needed. Please read the edit to my post.

 

[Urwumpe]

Can't we just extend the external airlock with the necessary ODS/Mir DS components, and keep the baseline mostly constant?

 

[Donamy]

Can't we just extend the external airlock with the necessary ODS/Mir DS components, and keep the baseline mostly constant?

There are issues with handrails and fabric covers.

---------- Post added at 02:00 PM ---------- Previous post was at 01:55 PM ----------

We could just release a bare-bones Airlock for now, with the TAA.

 

[Urwumpe]

There are issues with handrails and fabric covers.

What issues? Is it more than "Remove handrail group" and "add different cover group"?

 

[GLS]

There are issues with handrails and fabric covers.

---------- Post added at 02:00 PM ---------- Previous post was at 01:55 PM ----------

We could just release a bare-bones Airlock for now, with the TAA.

How long do you estimate it will take to finish all? I'll need maybe 2 hours tops to finish the code: the TAA part is done, I'm now moving the ODS ExAL mesh code from Atlantis.cpp to their own existing files, thus loading the meshes only when needed (as opposed the current loading then always and then hide/show them).

 

[Donamy]

What issues? Is it more than "Remove handrail group" and "add different cover group"?

The fabric covers need to be animated also.

---------- Post added at 02:25 PM ---------- Previous post was at 02:23 PM ----------

I can do the five meshes in about 45 min.

The Mir extras, alittle longer.

 

[GLS]

The fabric covers need to be animated also.

---------- Post added at 02:25 PM ---------- Previous post was at 02:23 PM ----------

I can do the five meshes in about 45 min.

The Mir extras, alittle longer.

Can't the Mir cameras and stuff be added to the ODS mesh, so we only need 4 meshes? We could hide/show those groups as needed. And the ODS textures you make the ISS version the default, and for Mir we change the texture in the code when needed.

 

[Urwumpe]

The fabric covers need to be animated also.

No big problem at all for us coders. You could add the covers to the hatch and make them a separate group. We animate the groups of the hatch including the cover. (Unless the cover is moved differently, then we just need two animations).

If there is no fabric cover, we remove it from rendering by changing its visibility flags.

 

[Donamy]

So how many meshes total ?