alright..see what i can do heres how it came out i think it came out alright.. [image: yUFH_petemirrors09.jpg]

Hummm, why not...I just have to immerse myself in sketchyphysics...I'm gonna try it, thanks for the idea pilou.

@unknownuser said: i might be misunderstanding so... Nice tutorial! You understood. Actually, I'm working on my tutorial's Ruby section and I want to suggest that this is a good example of a deceptively simple drawing that is really helped by having a Ruby. Using this is easier: [image: PdVW_roofster.gif]

And in the Shadows dialog, check "Use sun for shading". Now the amount of shading can be controlled by the Light and Dark sliders. Dark=100%>no shading. Anssi

@unknownuser said: i'm on a quad mac and do not have the slowness either.. SU is very responsive until i get into very high polycount territory.. when you say it runs slowly, are you talking only about how long it takes to open a file or are there other issues as well? most of my skps open instantly (no noticeable pause at all).. files with a lot of textures/poly (25MB+) take maybe a second or two to open.. you may have something set up weird or SU may be running normally.. it's hard to tell because i'm not quite sure what you mean by 'slow' I may well have something wrongly setup. It took me quite a while to discover that the Open GL option "Use fast feedbaback" was troublesome with the nVidia video card on the Windows machine. But stopwatch in hand, importing a 76 KB skp component into a 34 MB drawing takes 30 seconds on the Mac against 7 seconds on the laptop. Opening the 34 MB file on the Mac is faster than on the Windows machine (40 sec, against 50 sec), probably due to the newer, faster hard disk. Movement lag when orbiting is about the same on both machines. And sometimes drawing on the Mac seems to get somewhat sluggish (no measurements yet). I am evaluating the possibility of using large single files for projects, originating any number of views and cuts. Modifications of the base file can be automatically updated in Layout, avoiding a lot of work and the possibility of errors.

i think you would need to make your ground plane the same proportions as your image ie: 1024x768, which is 4:3 so it could be 100' x 75' for example. then when you create a new material in SU make sure to type these dimensions into the create material hieght width boxes before you select your image. paint the ground plane the you can scale it (make sure you scale uniform) and move it to fit your scene.

@honoluludesktop said: I was thinking about b-splines, loft, and other similar 3d modeling tools. if you need feature based NURBS modeling go for ViaCAD 3D which is avail. for a handful of bucks even for OSX and with SU import too. hth, Norbert

Hi Solo, Just trying to be creative, and didn't check if the following was previously posted. In the first model, create the desired scene, delete all, and save as a new file name. Then open the new file, find the drawing origin, and import the second model into it (on the origin), and go back to the scene as necessary. Worked on a small example, didn't try with anything big.

@solo said: Another example of how I do it [image: terrainpasting.jpg] hi!..a have a little question. The top view of the mountain, that texture, you did´t all whit Photoshop? Tks for the help!

Don't worry, I didn't know the answer until a couple weeks ago when I saw a post by Gaieus linking to it.

I worked CA with Mortenson as the contractor on a couple of projects way back in the day (actually had a hand in the hand drawings for the project it was so long ago :`) and they were quite progressive back then...this extensive use of SU doesn't surprise me a bit.

@gaieus said: Hi Al (and Rich), I don't think 3D PDF would become outdated soon. PDF is one of the most cross-platform and even cross-browser format (when embedded), can be easily sent by email or carried on a small drive since you can (almost) make sure that wherever you go and take (send) it, the other end will be able to see it. thanks for the comments Gai. PDF will probably survive, of course. But they have replaced the U3D format we use with a proprietary format. However, we suspect they will always have to display U3D, because otherwise legacy PDF files would stop working. We struggle with U3D because it never really worked properly. But Rich has made some improvements this past month and we will continue working with it. I think will will add an embedded html page creator to the existing plugin - using PDF first. The advantage it that it will be fairly easy for people to modify the page we create for their own needs. (It is much easier to modify a web page than to create one from scratch). This should work pretty well with embedded PDF and PDF is well supported. Also, we never learned as much about Flash as we should have. Perhaps a Flash/Collada option would work as an option. The O3D/Collada route is interesting, if Google gets behind O3D. Unfortunately, my experience is that Google often releases an interesting software application - usually in Beta mode - and then never decides to continue it. (Witness Google Desktop). On the other hand, if they do get excited about something they throw a lot into it - (e.g. YouTube). We are hoping they see a future for O3D and want to go comewhere with it.

Try running it under virtual machine - such as VirtualBox.

hi mead, thanks for you answer. i bought it in December 2007, it's a white one. it's an OS-X 10.5.8, 2.2 GHz Intel Core 2. Ram 2GB 667MHz DDR2 and the monitorchip is a GMA X3100. I have 36 GB available. So as you said, it's more than probably that's monitor-chip fault. it's weird that su 7 (this is the version) doesn't work whilest cinema4d works ok. well, i should change the chip for working smooth with it. thanks a lot

Since you're talking about animation export, I did a small test yesterday with Pibuz's warehouse. I made this video to answer a question about post-production motion blur in another forum, but I got an interesting result in render times: [flash=640,505:1e3wj8kn]http://www.youtube.com/v/iNVqKh2k164[/flash:1e3wj8kn] The reason for this seemingly absurd result (Twilight being almost as fast as SketchUp) is the same you noticed: During animation export in SketchUp, CPU use was just 25%. This is because the Core i5 has 4 physical cores. Who knows, perhaps when desktop CPUs with 8 physical cores arrive, we'll see biased raytracers become faster than SketchUp's native renderer because of its lack of multi-core support. Please note that I'm not complaining about SketchUp; as has been already mentioned, it's a modeler, and has never been marketed as a render solution. BTW, in the Twilight test I used an HDR as the sole light source, that's why you don't see hard shadows. Of course, you can use both HDR and Sun at the same time in Twilight. The variation in frame height was my mistake, before you ask.

This would be a case of think like a carpenter. The carpenter doesn't need the hole to exactly fit the sink, the sink just has to fit in the hole and since it has a return around the top edge, it will cover the parts of the hold that don't match the sink. The lazy man's way: First, hide the sink. Double click on the top until you get through the different groupings to where the top most surface is highlighted. Draw a rectangle on the surface of the top. Push pull the rectangle through the depth of the top. Turn on Hidden Geometry under the view Menu without exiting the sink top group, you will now see the ghosted view of the sink. Using the move tool or the push pull tool, move the sides of the hole until they are under the apron of the sink edge. Once that is done click out of the group, right click on the hidden sink and unhide it, turn off the hidden geometry under view and you are done. This uses very little geometry and represents what the carpenter will really do (they may put a radius in the corners but not always). Hope that was clear.

@hpw said: When it is possible to write a small ruby to draw text in SU, then it should be a easy to make an AutoLisp-utility to write such ruby out from AutoCAD. I used such approach to restore block-infos on SU components which get lost during import. It is indeed relatively easy for Ruby to parse the 'Text' data out of a DXF [then you won't need AutoLisp] and then to make bits of '2D Text' from it in your SKP... It's just having time to do write it that's the pain...

@remus said: If you go window->preferences->openGL then select the setting with the highest anti aliasing, that may help. it worked, and looking awesome. Thank you!!

Hi Mike, hi folks. Make sure that the values that you will import are not too small. SU has problem creating faces with edges smaller than about 1 mm. It is best to import using meters as units and/or scale the points after import, if you want to make surfaces with them. Just ideas.

if you want to render the image, just use a normal map (or bump map) with blurred black stripes on white background for your material. If you want more detalied effect, and your rendering engine allows it, you can use a displacement map. If you don't want to render the image, you can just make a texture with desired visuals... Trying to model that, will result in lots of polygons and SU might just not be able to handle.