Jump to content

Mesh Buildings w/Physics Determining LI?


Summer Logan
 Share

You are about to reply to a thread that has been inactive for 2894 days.

Please take a moment to consider if this thread is worth bumping.

Recommended Posts

Hello

I have a couple questions about Meshing...the main question being about Meshing with a Physics File and how to determine what the LI will be when switching from Convex Hull to Prim before uploading.

So some back story to my question, firstly I use a Mesh Generator in-world to convert Prim to Mesh. I made a Mesh Building with a Physic File and when I uploaded it, it was 4 LI on Convex Hull. I then switched it to Prim and it still said 4 Prims...By the time I added the windows and doors it was 5 Prims total. Which was great! I had a small texturing flaw but figured I would rebuild the building...same size, same way, same textures even...just fixing the flaw...however this time I uploaded it and it went from 4 LI to 7 Prims just from switching from Convex Hull to Prim...then after adding windows and doors it went from 7 Prims to 9!!! ARG! I did everything the same exact way...so I'm wondering what happened before that 4 LI stayed 4 and how can I recreate that? 

So basically on the upload window I see where it will say what the LI will be...but I know that's the Convex Hull value of the LI...how can I determine the LI when I'm using a Physics File and I switch from Convex Hull to Prim. Why do some things amazingly stay the same and other things the prim count jumps way up? Even if it's the same exact build? Very strange.

My second Mesh question is...as stated I use an in-world Mesh Generator...however recently I discovered I don't need to use that anymore and I can export to Blender from SL and then upload back into SL...However I've been having problems with the texturing...For example when I used the Mesh Generator it would remember all the texturing surfaces and if they were horizontical or vertical also if I lined up textures using planar it what somewhat retain this information. Now using my new export method even though it's textured perfectly in the world when I go to upload and retexture it the textures are going opposite directions and not lining up at all...I don't know much about Blender...I only know a few self taught things...however I would love to use my new exporting Prim to Mesh method if I could figure out why my textures are messing up on reupload and how I can fix them. 

Thank you so much for reading...You can reply here or if easier message me in SL...Thank you!

Link to comment
Share on other sites

One possibility for the unreproducible LI: The LOD generator used by the uploader is non-deterministic. That means that identical input can produce different results on different occasions. The differences can sometimes affect the download weights. If the LOD meshes are used for the physics, then the physics weight can also change.

You didn't tell us whether you used automatic LODs. So we can't tell whether this might be the source of the effect. Also, you haven't said whether you used triangle-based (no Analyze) or hull-based (Analyzed) physics. If it was triangle-based, then there are some strange characteristics of the uploader that can produce quite unexpected results. If it was hull-based, then if it used auto-LOD, the physics weights could differ because of the non-determinism. Was it the download weight or the physics weight that changed?

Differences in UV mapping can also have quite large effects on the download weight. So if your texture changes involved changes to the UV mapping, that's another possibility. This would be reflected in different uploader vertex counts in the uploads before and after. Did you record these?

There is no way to obtain the Prim-type physics weights before uploading and rezzing. If you need to experiment, you might want to do this on the beta grid (Aditi), so that you don't waste real L$ on upload fees.

Can't help with the downloading to Blender, as I have no idea how that works.

 

Link to comment
Share on other sites

Ok I am very sorry I didn't give enough information. I am still new to Meshing and don't know a lot of the lingo or what everything means.

The building I had initially made from prims was like a little house...had a bay window...open concept kitchen/living and 1 bedroom. I made everything you weren't going to see invisible and generated that (I call that my main view, as that is what you will see and what is textured)...then I removed the roof and tops of doorways and made only the parts you collide with visible while everything else invisible to make the physics file (named physics file) and I exported that out too...Now when it came to uploading I selected my main view Mesh file...I didn't have to change any of the numbers on high/medium/low/lowest and they were fine as they were. I didn't remember the exact numbers, but I just checked and it says download 4.3

Then I went to the physics tab and chose physics from file and selected my second file named Physics file and I'm pretty sure I hit analyse and I thought I always had too. Plus it shows after analysing if there was any problems in the physic file like for example with the doorways etc. physic shows 1.1 and server 0.5

That is all the information I know right now. I didn't keep the original mesh files on my computer (dumb) and I don't have the original prim version in my inventory (dumb again!) lol I rebuilt it once and that when it went to 9 Prims...what makes me so sad is the fact I had it at 5! FIVE 5! Why I can't do it again I don't know =\

Link to comment
Share on other sites

A couple of pictures of the model, the physics mesh in Blender, the importer window, and the More Info floater with the weights might help, to get an idea what's going on.

Also what Drongle mentioned about analyzing the physics mesh or not. Try both methods and see if you don't  get the lower result you had at first without analyzing. It's common to forget to hit one, or the other button once in a while.

Link to comment
Share on other sites

The house is one mesh when I upload it.

 

Also I don't know the difference between triangle-based (no Analyze) or hull-based (Analyzed) physics. As I stated I'm still new to meshing and don't know a whole lot.

However I'm a fast learner. 

Link to comment
Share on other sites

They use different kinds primitives in the physics engine to detect collisions. Triangle-based shapes are just that, a collection of triangles. These don'r work very efficiently, especially if (any of) the triangles are small. So their physics weights get larger the smaller the mesh is. When you click "Analyze", the uploader generates a set of convex hulls that are supposed to approximate the behaviour. The physics engine can work more efficiently with convex hulls than with triangles, and since tTheir size makes no difference, the physics weights are independent of size (mostly). There is a crossover point, at about the size of walls of houses, where the triangle-based shape can start to have a lower physics weight than the hull-based shape. The hull generator can't deal very well with complex shapes. So it is usually best to present it with a set of blocks that are already convex hull. In other words, do the work before it gets the chance to mess it up. There is a lot of discussion on optimising either kind of shape in previous threads in this forum.

 

Link to comment
Share on other sites

  • 2 weeks later...

Depending on the physics behavior you want, you can sometimes dramatically reduce the physics-based LI result even in prim mode by simplifying the physics. For example, for the walls that have windows, set the physics on that wall as if the window wasn't there. This will drop the physics rendering cost in prim mode because the computer's view of the wall physics shape is simpler. An avatar won't be able to climb through the window, but you probably don't need that anyway. If an interior wall has 2 texturable faces but no windows or doorways, and it's just a texture difference, pretend it's only a single wall with a single face in the physics file. This won't change the physics behavior for avatars, it merely makes the prim shape calculation a little simpler and therefore slightly lower-LI.

You can also ignore some small details, ex: if there's a very thin texturable rail at the top of the wall for decorative purposes, pretend it's not there when making your physics file, since the shape of a rail 0.05m thick (for instance) four meters up isn't going to make a difference during normal building use. Unless you want people to be able to bank thrown objects against it and take the rail position into account, the physics shape of the rail is unnecessary. At some point, there's a trade-off between more realistic physics and the complexity and LI cost of the physics file; before that, there are almost always some simplifications you can make beyond "what won't be seen" and after that point, it's a judgment call, considering what your typical use will be and, if you mean to sell the item, what market segment you're aiming for.

The calculated physics LI when uploading is still going to be inaccurately low, and it'll go up when you switch to prim physics. With a little luck and a lot of keeping your physics and mesh files as simple as possible while retaining all the features and faces/verts you need, you can still keep it low and possibly even prevent the prim physics from exceeding the download cost.

You can also set the window physics to "None" after linking them to the mesh walls, especially if the wall physics don't have a hole where the window is. Openable doors do need their own physics and can be set to either convex hull or prim; their shape is typically simple.

On the texturing issue: it can be a bit of a pain in the butt, but review the "sync selection" feature in the UV mapper of Blender: http://blender.stackexchange.com/questions/27298/where-is-the-sync-selection-option This will allow you to check the alignment of UV texture maps, and then you can go back to regular mode and rotate the UVs as needed. I'm not sure if there's an easier way to do that or not; hopefully yes, but if not, that does work.

Link to comment
Share on other sites

You are about to reply to a thread that has been inactive for 2894 days.

Please take a moment to consider if this thread is worth bumping.

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
 Share

×
×
  • Create New...