Kwakkelde Kwak

So after a day (and putting on my laced shoes a couple of times) I noticed the guy in the video still has it all wrong. You don't have to loop the other way, that leaves the loops perpendicular to the tie just as much as how he (and I) did it at first. The trick is not to loop the second loop completely around the first one, but to stick it through the hole right away. My god this is complicated, even for veterans. (But at least I didn't have to retie my laces today) ((and he is right about the laces looking better!))

I am pretty sure that is your issue. Unfortunately I can't help with any Blender specifics, but now that you have the probable cause, I'm sure some Blender veteran can show you how to unify the normals.

So Vivienne was right afterall. Not minutes, not hours, but decades it took me to learn to tie my shoelaces, at least correctly.

I don't know how Blender "does it", but in 3ds Max missed rays are represented by bright red. Your picture shows "strange colours", at least strange for a normal map. Maybe some rays hit your high poly model from the wrong side, resulting in negative normals with matching colours. I can imagine something like that with all the creases and folds.

Vivienne Schell wrote: Fact is, you did not learn to tie your shoes within one hour, I learned it in a couple of minutes, before my swimming exam. My left handed father tought me, so I do it backwards.

Not to be nitpicking, but since Sketchup changed owner a couple of times, I can imagine the 3rd party sites aren't allowed to offer it. If Trimble offers Sketchup 8, that's not an issue of course. Again, it doesn't seem to sound like a very good tool for the job.

Penny Patton wrote: Your only restraint is a 256LI cap on any single attachment. But you can wear up to 38 attachments. This means a single avatar can wear more than half a sim's worth of content. It's far worse than that. As you said, there is no LI calculation for attachments. The only restriction is a 256 cap on amount of objects in a linkset. Since a single object can be "as big as a sim" and you can link 256 of those objects together, wear that thing 38 times, the impact is beyond comprehension (both for humans and computers). I'm not sure what the exact triangle limit is, but the grand total is in the 1 billion range. (130k triangles x 256 x 38 = 1.3 billion) The amount of texture memory used is even worse. Every mesh object can have 8 textures, so then we're at a maximum of 10 billion unique textures. Looking at certain avatars/attachments, I sometimes get the idea people are actually trying to reach these numbers.

If you already have it, I suppose. I can't find the old terms of use, but from what I read you were at some point allowed to use Sketchup for commercial purposes. I don't know if there are legal ways to obtain such an old version though. The answer to that would be in the terms of use as well. More importantly, I don't think sketchup is the program to use if you want to make geometrical 2D images.

Could you explain what you mean by "small UV's" ? The UV data itself isn't altered by mudbox, so going by your post I assume you mean the output textures are smaller than 1024x1024 (which sounds awfully big for a fingernail btw). In mudbox you can set the texture size when you create a new paint layer. To my best knowledge the default is 2048, but if that's not the case, you can always change it yourself.

IvanBenjammin wrote: If that doesn't help, then what Dain said above - you definitely don't want overlapping UVs. The other likely culprit, looking at your images, is that your Low Poly mesh is intersecting with itself in those folds. Any kind of overlap, be it UVs or faces, is going to cause problems. I'd put my money on that one. Overlapping UV's or geometry are a nightmare for baking textures, but if you project a high poly onto a low poly object, that shouldn't be an issue. The ray should pick up the first geometry it hits, so that's always just one. Look at the picture below, the lower (intersecting) geometry is projected onto the upper plane, which shows the resulting normal map

PersistentDigger wrote: I followed your link for LibreCAD. I know from past experience of "free" software that often they are in effect demo versions. They might work nicely in their own right, but it is using their file format that can be problematic. LibreCAD isn't a demo, it's completely free. The dxf format is an autodesk format, but it's only used internally. If I want to extract a picture from AutoCAD, I use a virtual pdf printer and import the print into GIMP (Yes I do use free software myself:) ) My best guess is that would also work just fine in LibreCAD. (I use PDFCreator for a couple of years now, but there are plenty around on the internet, for example CutePDF) Anyway, as you said yourself, just download LibreCAD and try it. I haven't done so myself but it looks like a promising alternative to AutoCAD if you don't need all the functions and don't want to spend any/a lot of money.

You could try LibreCAD. It's free en completely open source. Sketchup is not the tool you want for the job, since even if it's for your "personal doodlings", you're not allowed to use their free version (Sketchup Make) for any commercial endeavours. The Second Life TOS demand you to have full rights over your creations, which you simply won't have using Sketchup Make. From the Trimble TOS: Trimble Navigation Limited and/or its affiliates ("Trimble") gives you a personal, worldwide, royalty-free, non-assignable and non-exclusive license to use the executable version of the Software for non-commercial use only. Non-commercial use means: you may not sell, rent, lease or lend the output of the Software or the Services. If you are a for-profit organization of any kind, or an employee of a for-profit organization using the Software or Services in that capacity, you are engaged in commercial activity; therefore, in order to use the Software and Services, you must purchase a SketchUp Pro license.

Although what you suggest will work just fine, I'd suggest strongly NOT to do that for performance sake. Four unique 1024x1024 textures just for some AO effect will take up an unreasonable amount of memory, let alone nine tiles or more. Making an overlay with a single plane a fraction above the floor for the AO sounds like a better plan, that way you can tile the wooden floor in SL. In photoshop you can invert the AO map (so black and white are flipped), then copy it onto the alpha channel. I'd use a white texture, so you can change the colour in SL after uploading.

Looks like the version for the Mac is a beta. I suspect once it leaves that state, you'll have to buy a licence for that as well.

arton Rotaru wrote: Because you can achieve the desired texel density with small textures even on large meshes as well, with seamless tiling. And this way you can maintain a consistent look across all of your assets. Which makes them look even better. I just wanted to highlight this. For example, my latest brick walls use 256x256 textures, with a normal map. Especially with a couple of local lights, this gives a very satisfactory effect. A benefit of this setup is that people with an older or slower computer can tune down the graphics level and will only have to download/render a single 256x256 texture. People with a better computer can have all the extra effects of the normal map and lighting. The hard part is to find a balance between the texture not becoming too bland and the tiling effect not being too strong. For me the trick to decide "how low I can go" is to start with a small texture, see how it looks, then if neccecary scale up a level. I'm still often surprised on how little is needed to make my objects look the way I want. I can't remember the last time i uploaded a 1024x1024.

I don't use Blender, but if I'm not mistaken you can reset the values by hitting ctrl-A in object mode with the object selected. Alternatively, you could make a new object, join your mirrored object to it, then delete the new object.

I guess a lot has got to do with personal preferences, so I can only tell you about my own. I have a touch screen laptop and I use a wacom pen for certain tasks on my desktop. For precision work and speed both the touch screen and pen are pretty much useless for 3ds Max which I use (so I suspect the same for Blender). The touch screen I like for browsing and other simple tasks, the pen I use for texture work only, both 2D and 3D (Gimp and Mudbox). I absolutely couldn't work without a mouse when modeling or UV mapping, but for 3D painting in Blender (which I suspect you want to do) I'd prefer a pen, you get a lot more control with the pressure sensitivity and ability to draw like one would on paper.

Coby Foden wrote: I noticed that Autodesk just released free software which can make a 3D model from a bunch of photos. They didn't just release it, they're working on it for many years. First release was in 2009. Anyway, as I said earlier and as someone replied and as you stated, it will certainly NOT produce SL ready models. On top of that you'll need a good amount of good photographs from all angles, so besides a lot of "post-123D Catch" work, it will take a good amount of "pre-123D Catch" efford too, not to mention you will have to have access to a physical model.

Dellon Omizu wrote: I have another question, its possible to test the castle in SL beta Grid? I logged in there, and said I dont had permission for upload mesh, but I have for normal grid. For some obscure reason you have to take the test for both the beta grid and the main grid. You can do it here.

SInce you're talking about a castle, it's very possible your object is simply too big to be one linked object in SL. So, how big is it? You might also want to merge some objects before uploading (maximum size is 64x64x64 meter). I think the term in Blender is "join". One object can have up to eight separate materials, so that can really bring down your amount of objects, resulting in a lower server weight and therefor possible lower land impact. Joining can however increase download weight. Too many blanks to make a fair assessment. Good luck and don't be shy to post more information or questions.

Autodesk has a free app/program that will convert a series of photo's into a 3d object. I haven't really looked into it for years, so I don't know about its current state. Might be worth looking into. You'll still need Blender, 3ds Max, Maya or whatever polygonal program to make the models SL ready of course. It's called 123D Catch.

There are some free alternatives to Shadermap ($40), bitmap2material ($135) and crazybump ($299). The ones I've tried aren't giving quite as good results as the ones you pay for, but either Awesomebump or Insanebump (to name two) can be very helpful. The latter also has a GIMP plugin version.

I would say go for Blender for the Mac. If the choice is between Mudbox and 3ds Max, definitely go for 3ds Max. Mudbox is great for fooling around, making high poly models which you can "trace" in a program like 3ds Max, editing your 3ds Max creations in high poly for normal map baking and (which is what I use it for mostly) 3d painting. For creating game assets it is pretty much useless by itself.

Gaia Clary wrote: ChinRey wrote: Flat (or sharp) normals doesn't necessarily result in higher download weight than smooth ones. In many cases smooth normals gives considerably higher weight than flat ones. Can you give an example for this ? All i learned about normals indicates that flat shaded faces need 3 times more coordinates (one verterx normal for each vertex of the flat face) which would reslt in a raise of LI. Which would fit quite nicely in the scenario as described. from 16 to 46 or what was it, is pretty much that factor 3. In the UV editor in 3ds max, you want as little islands as possible, this results in the least amount of double vertices. By the looks of the land impact all triangles are their own seperate island, although I can't understand how one would get such a UV layout. Maybe a "flatten" with a very strict threshold? @Chin... I am also curious about an example where smoothing raises the landimpct instead of lowering it.

Why don't you install it and let the rest of us know? I have my copy of 2015 in the big black box, along with 2014. I was planning on updating my 2013 to 2016 next april or may when it comes out.