@thomthom said: Your lines are 380km !!! away from ORIGIN. SketchUp doesn't handle such large lengths. Import the model and untick "Preserve drawing origin" in the import options.

HEY...!! This tutorial is exactly as mine, I did it on july, but mine includes v-ray tutorial too...!!!!!!!! Check it out on http://forum.gfxdomain.com/others-tutorials/5391-xfrog-plants-tutorial-sketchup-v-ray-spanish-only.html

Quad core, 4gb and dedicated GPU and 64bit OS. The number of parameters around best for SU is subjective. it's a 32bit app so can't avail of more than 4gb of RAM on 64bit systems. But a good processor, enough RAM and a healthy GPU should yield good response time.

@unknownuser said: @jgb said: The "paperless office" has about as much chance of success as the paperless toilet! This is why most offices cannot do it. They think like this. You just have to change your attitude and approach. You're probably still writing checks ... I just wonder how long it will be before the last part reads: "You're probably still drafting drawings". Thanks for the info about the site office btw.

A draw call is a function used by a program to instruct the GPU to draw something. In games this is done once per frame, normally lots and lots of draw calls per frame (sort-of roughly analogous to your monitors refresh rate). In general you have one draw call per texture file used per object, but it may be more than that if you specify particular material types, ahh, check that, let's call it "qualities" so we SU folks are clear it's something different from SU materials, to influence the appearance of the texture. Such "qualities" might be gloss, brightness, various types of finish, like metal, rubber, wood, etc., and then whether there is an alpha channel present of not (transparency) So given one model, 4 texture files, 2 of which have an area of transparency, 3 of which have areas where those "qualities" apply, the draw calls would be 4 for the four texture file as applied to a basic appearance, 2 for the transparencies, 3 for the "qualities, for a total of 9 drawcalls. For a game, that's 9 per frame for the one model. In a complex game there might be a couple of hundred models in view so that may add up to several hundred or even a couple of thousand draw calls, depending on complexity. Because there is hardware involved each draw call is a CPU interrupt -- hello, this is long distance from America for Mr. Floyd... is Mr. Floyd there?. In addition to letting the GPU know it is supposed to do something, a game will also provide the mesh for each object that is to be drawn, which is all of the xyz data that forms the vertexes of the model as well as the UV mesh which tells the GPU how to align the artwork. For each model for each frame. In my experience it starts to get hard to keep frame rate equal to or above the monitor refresh rate when the draw call count exceeded 500 per frame. On the front side of the glass sits the game player, who, quite naturally, does not want to see any slow-downs/speed ups. The net effect is to place a burden on model builders and artists to make their stuff so it requires the fewest draw calls possible. SU is not particularly good at this but at least has the benefit that things can be done very quickly and so it can have a proper place for both prototyping ideas and doing fairly basic game models.

Hi Jeff, Did you geolocate your model? If so, it should have imported two things: a flat snapshot and a 3D terrain. Open your Layers dialog (Window menu) and try the toggle terrain button now. You should see how the layer visibility changes in the layer dialog. Is there anything like that there?

Hi Lee, The PM function is typically for box shaped buildings. In our historic downtown area, we hardly have a building with a rectangular plan so I cannot even use PM for those - not to speak about organic shapes like valleys and rocks. I doubt that it could ever be successfully used for those. There are other techniques/applications but generally they involve more photos taken from different angles so that the software (like TGi3D PhotoScan) can create the 3D shape from those different shots. Just like our two eyes that see two, different images of the same object and our brain puts those together into a 3D image (or something)

Hi Joe, hi folks. I did that a while ago : Horn2.skp However, since we now have access to plugins, I would do it like that: 1 - Create a cone; 2 - Bend it. To create a cone, draw a triangular profile and use a circle for the path with the Follow metool. For a ram horn, you may have to bend in steps and scale to create a helix form since these can have more than one complete turn Just ideas.

I second TIG's question why you're removing the Susan definition? Why is that essential for the plugin to work?

Yeah, that hint to the other "genre" is completely relevant in this case!

Hi Jase (Jason?) and welcome! @jasemilly said: Hoping to learn lots and not ask to many silly questions, my modeling knowledge is as close to non-existant as possible. There's practically no "silly" question here. This is why we set up this section: all newbies feel free to ask whatever they are not certain about.

mwm5053, many thanks for you reply! The link with the sample is really very useful andit gives a good impression of all the renders.

That's it! Thanks mate. Knew there had to be a simple way.

The 'colors' are set within the 3ds file. The 3ds file will be looking for any textures' image files in a certain place. So either keep the relevant folder of images with the 3ds file OR put the images in the same folder as the 3ds file. If the 3ds file has an absolute path set inside it, then I'm unclear how we might find that out...

Brilliant. Success. Thank you.

there are a few tanks in this forum that are really quite good and describe their methods. search for tanks.

Gai....Very nice! Going to try that. Chris....Never heard of that before. Will check it out. Thanks to both of you.

Pretty much the same rules as making any SU model smaller. If it's a repetitive model you can turn geometry into instances of eachother. It depends entirely on what's making the file size so big though. Maybe you can post a screenshot of the model. -Brodie

That's wonderful! Thanks Hieru!

True, low polygon model is "faceted" with the silhouette. Like in examples... there is nothing that can be done for it with selfshadowing. Only subdivision will help for that.