First Mesh Build - Warehouse

[Greg Stella]

Hey guys, first of all, be nice please :) lol

 

This is my first mesh build, its a warehouse, mostly would be used for a store? Just wanted people opinions on this? Is it something you would buy? If so, how much would be an okay amount to pay for something like this? Put it all together on the beta grid tonight and it came out with a total of 54 prims/LI, which I don't see is too bad.

 

click for the picture of the warehouse :)

[IvanBenjammin]

Good first build. The LI might be a little high if its mostly flat surfaces, but that depends on the scale of it. Also, depending on how many mesh pieces its made from, the physics LI might be a bit high.

You need to fix the texture on the brown trim areas. It looks like a tiling texture but it would be better off without the lines in it.

[Greg Stella]

The LI is 54 with the pysics done already. And yeah, I think it would be better with some better textures

[IvanBenjammin]

Yeah, what I meant was, if its 54 LI because of physics (meaning the physics weight is higher than the download weight), you could probably reduce it.

[Charlotte Bartlett]

Hi Greg

It's a great start.  I think you can do some minor tweaking to your textures and you will be on your way.  I think you can sell it once ready.  You may want to consider an introduction price etc to get it up and running - tbh it's hard to give a price - personally I price my items at a range dependent on the level of effort,design,uniqueness etc.   Perhaps look at marketplace to see what a range of merchants price at and choose something you feel comfortable with.  I would do it mod and copy.

Additionally, below you may find some other mesh build information (perhaps useful, rubbish or things you are already considering):

1 - You can bring that LI to a lower number by playing with separate meshes , with custom physics (don't use the generated from SL loader) etc and using separate meshes for the interior walls to external walls with LODs.  Low LI is very important for stores as merchants will tend to need to use a lot of their prim allowance for all their vendors.  

2 - Materials- not sure what application you are using but recommend baking the Ambient Occlusion to give some shading to your textures.   If you want to take it a step further add lights to the interiors and bake the illumination too.  I think most people would expect for mesh to have AO baked in.  

3 - To keep the build efficient remove unused faces before uv unwrapping and uploading.  You will already have done this if you model with planes (but I know some SL builders like to model like they do in SL).

4- Consider the scale of the building and rendering cost (number and size of textures will be part of that).  Texture lag is becoming epic in SL as more builders put on multiples of 1024 unique texures over a build (even long established builders).   This is a really poor practice for game/SL assets.   You can google Penny Patton she writes up a clear approach for builders/designers about this at Digital Pastures.

 

 

 

[Kwakkelde Kwak]

---

Charlotte Bartlett wrote:

not sure what application you are using but recommend baking the Ambient Occlusion to give some shading to your textures.

Texture lag is becoming epic in SL as more builders put on multiples of 1024 unique texures over a build (even long established builders). 

---

I'd say your recommendations are good, but in the above there's one (somewhat understandable) flaw in your logic.

The best way to keep the texture load down, is to repeat textures throughout a build and throughout a collection of builds. Repeating  a 1024x1024 texture five times, causes less load than using five unique 512x512 textures.

Repeating textures with baked in AO is a lot harder than repeating basic diffuse textures, using baked in lights/shadows is even harder. It's certainly impossible to do it a lot of times in the warehouse as displayed. So using baked in effects will in most cases, including this build, result in bigger textures and more of them to keep the same texel size (resolution).

[Chic Aeon]

---

Repeating  a 1024x1024 texture five times, causes less load than using five unique 512x512 textures.

---

Agreed that you can't have your cake and eat it too but also wanted to note that a 1024 texture is FOUR times bigger than a 512 texutre ( I know YOU know that but some folks aren't that good at the math thing) so that while there is a difference, it isn't all that great. Strategic planning helps.  What I have done in the past is to build modularly so that the 1024 texture (with ambient map) is reused. Also, for the most part ambient maps do very little good on the outside of a building (flat wall) and can be used on windows and interiior areas instead, using repeating textures on the flat surfaces.

 

 

***********************

 

Looking at the building and guessing on its size, I suspect the land impact could be lowered a LOT. As mentioned, uploading sections of the build and linking inworld usually helps. And a very plain physics model (much info on that in past threads) may help. At the moment, I can't see much advantage to having that build as mesh. It pretty much looks like a prim build (not a bad thing but much easier to do LOL) and the land impact is about the same or maybe even more than it would be as prims.

A very good start though!   It all takes time. If you have the physics worked out, much of the battle is already won. Yeah!

[Drongle McMahon]

I thought the same about LI. So I just made a similar model, a bit more detailed, with 12 step staircase inside, and at about 38x35x21 it's LI was 18 (download weight, default CH physics=1.7, but no good for real use, of course)).  That's all in one piece,default medium LOD (you won't ever see anything lower). So that would be a good maximum target.

More important, I was impressed by the inworld AO, This is the latest viewer. It seems much stronger than I remember last time I made a big building. It alsi scales with the ogject. Anyone know if there have been changes to this? If it's like this, thgen there is indeed no need for baked AO, so that tiling textures can be used without sacrificing AO. That will mean a huge saving in texture download and gpu memory consumption. Here's a picture, inside and out, with blank white texture (I haven't UV mapped it, and I'm not going to, as it was just an experiment).

Of course, only those with good enough GPUs to activate AO will see this. So maybe this is for the future.

 

[Kwakkelde Kwak]

---

Chic Aeon wrote:

---

Repeating  a 1024x1024 texture five times, causes less load than using five unique 512x512 textures.

---

Agreed that you can't have your cake and eat it too

but also wanted to note that a 1024 texture is FOUR times bigger than a 512 texutre ( I know YOU know that but some folks aren't that good at the math thing) so that while there is a difference, it isn't all that great.

---

The difference between a single 1024 and five 512's isn't that great, but the resolution is twice as high, that's why I used it as an example:)

With a texture repeat (per surface) the differences can be far bigger of course. A 10x10 meter wall will look perfectly fine with a 256x256 brick texture repeated twice both horizontally and vertically. To get the resolution just as sharp with baked in AO, you'll need a single 512. (which uses, as you say, four times more memory). You can repeat this baked texture throughout the build, but it still needs four times the memory. If you can get away with a 4x4 repeat on a surface without a tiling effect, the difference will be 16 times.

[Charlotte Bartlett]

I think you misunderstood my post as I was not very clear apologies.

I was referring to those who use say 50 x 1024 UNIQUE textures.  Not those who repeat the same 1024 texture.  Hopefully that makes more sense!

I think it's a balancing act - and it's just something to take into account as I think for the OP it would help add value to his build to not have flat textures.   Worst case, use for the walls a texture with "shading" takes 2 seconds to do in photoshop and repeat that.

I am not sure how many  people can use the AO in world as it's requires advanced lighting - so against a balancing act.   I think the difference to me is he is asking what to price it  / whether we would buy.  So my feedback relates to a commercial approach I guess.

.

[Kwakkelde Kwak]

I understood your post just fine:) Maybe my answer wasn't clear, I don't know.

You can overload someone's computer. People who can't see the realtime occlusion, will have a relatively underpowered computer. So it's in their best interest to keep the texture load to an absolute minimum. If it's possible to replace a single diffuse texture in a build with a single baked one, the load will be the same. But as I explained, you have repeats per face, repeats per object and repeats among several objects. The repeats per face can almost never be done with baked in lights or shadows. So if you have to replace a 4x4 repeated texture with a single texture, you either lose 16 times the resolution or will end up with 16 times the memory use. That's your worst case scenario.

I understand the commercial point of view: "everybody does it so why can't my build look that good with nice lighting and all no matter what people's setting are?", but it is just bad practice from a logical point of view: "slow computer, keep the load as low as possible" and "fancy computer, fancy graphics."

[Aquila Kytori]

Hi

Something I notice in many "noob" mesh models is that there are often many vertices that could be removed without effecting the appearance of the mesh. Removing these by merging to others will help lower the LI count.

[Drongle McMahon]

Here's another example with a filled frame, illustrating some other mistakes/options. Four stages: 1. Left everything from loop cuts made to insert the walls without any hidden surfaces. 2. Remove redundant vertices, as Aquila said.  3. Delete faces at the bottom that will be underground and pointing down. 4. Sacrifice perfect joints by making the cross beam and the walls unconnected pieces (like the bottoms of Aquila's pillars). This saves the extra loops in the main pillars. Note that these pieces are not complete cubes - the edges of the walls and the ends of the cross beams are deleted because they are buried inside the main beams. This works, but you can run into problems with baked textures where one piece penetrates another unless you are very accurate. It's not as aesthetically satisfying as the properly jointed mesh ... but nearly halves the triangle count. (Tr = triangles; Vt = vertices).

[Aquila Kytori]

I've been using fitted mesh for over a year but I can't seem to get the new BUTT bone working for me.

 

Here's a white mesh rigged solely to the BUTT bone with avatar shape at default Ruth values. (The avatar skin is a alpha grid style so you can see the mesh)

 

 

If I push the avatar butt size up to the maximum 100, the mesh doesn't get bigger at all, although the avatar shape definitely does.

 

Even stranger, if I set the butt size to values below 25, the mesh becomes wider! (Although it does get flatter too so it's partially good)

 

If my scale was off, I would expect the mesh to be too small or too big or rotated but it should still scale relative to the avatar's butt. And I know it's reacting to the BUTT bone because, when I use avatar physics, I can make the mesh bounce and bounces in correct synch with the avatar mesh.

 

Has anyone else tried using the new BUTT bone? Am I misinterpreting the purpose of the BUTT bone?