That is one great resource David. Much appreciated...

Hi Milo You might want to try posting on a forum like http://www.nicepapertoys.com/, but please keep us informed, I'd be interested to know the answer also. Cheers

Maybe you are thinking of the ability to import and export keyboard shortcuts?

Does the order of Ctrl and Shift in the Shortcuts window matter in practice? Along with the V they get played more like a three-note chord on a piano than playing the individual notes in succession.

That mesh looks really good. Even with a lot less guide points. And hey, you didn't even have to bang your head against the wall. Man, 400,000+ points. That makes my head hurt.

I think Photogrametry is trying to straighten a face in a photograph of, say the wall of a building, removing perspective & scaling it up to real units to make it the basis for a 2D survey drawing, while Photomatch is what is provided in SketchUP - creates a 3D model based upon the perspective captured within the image. Kubit PhotoPLAN is an example of a Photogrametry package, which works well with AutoCAD.

You have Piranesi which is a real "3D Photoshop" The link seems to be closed just now: http://www.piranesi.co.uk/ (hope it's only temporary ) so I send a You Tube link as example http://www.youtube.com/watch?v=cO2KKD5rJTo&feature=channel_page Fotosketcher is only for PC EDIT: Bad news: searching the Piranesi link to send it here, I found that Informatix editor of Piranesi has some real problems: http://www.informatixsoftware.com/forums/showthread.php/3122-Informatix-Inc.-Japan-Statement

See this link for your chair http://sketchup.google.com/3dwarehouse/details?mid=e0f16ba60e1deefe427d16763a0ff49. How it was done: Saved the Febo chair as an obj file( it has 5825 vertices and 11238 faces); Imported to Mesh lab( it will import 3ds also) and ran through filters: Merge close vertices ( d= .00010251) with 3 merged; removed duplicate faces with abuot 40 removed; Saved as dxf file and imported to DoubleCad free. Scaled by 10e6 in DoubleCad; Imported to SU 7.1 free with the result you see. I have not re-scaled yet. The reason with the step to Mesh lab is because it has a number of filters for clean up and decimation. I have not tried to reduce model size smaller than above. In addition I think Mesh lab is using a older version of dxf and wanted a later one for import to SU. Note the small artifacts at left/ right cushion back rest/ seat are in the basics model as well as the small ones on seat front edge. You can try different filters in Mesh lab to reduce model size to what you find acceptable. Good Luck

@unknownuser said: Fredo's joint push-pull will allow you to do this: http://forums.sketchucation.com/viewtopic.php?t=6708 thx

thanks a bunch Gai, that fixed the problem

Or get Cutlist 4.1.1 here.

It seems that you are using small, tiling textures BUT quite complex modelling (for GE at least). This is not what Google wants however. What they would like to see is that you take a photo of the whole building (or a side) and apply that photo on a whole side of the building without modelling all the openings etc. As for the model approval line; it should not be more than a week or maximum two.

IMO, there is little value with running your own app. in a browser. If however, an occasional user could access SU pro as a server app for pennies, then running SU Pro in a browser could be an attraction.

Hey Jason! Welcome to SketchUcation. If you are just learning SketchUp, I would recommend watching lots of tutorials. This helped me a lot when I was learning SketchUp. In fact, I still am and benefit a lot from watching tutorials. Check some out on Youtube or the SketchUp website. Also, this thread has a lot of other helpful tips. http://forums.sketchucation.com/viewtopic.php?f=79&t=12706 If you have any questions, feel free to drop them in the Noobie forum where you're sure to get a quick response. (The people here are just awesome like that. ) Finally, don't be afraid to post your creations in the Gallery forum so others can see your cool models and where you will get some helpful tips, tricks, and advice to improve your work. You'll probably even teach others a thing or two. Hope you enjoy it here and learn a lot.

@gaieus said: No, this is not working. It's either not the same length or not at 45 degrees. I'll leave it to someone more knowledgeable. It's probably due to the Earth being round instead of flat. We could blame Christopher Columbus. He's not here to defend himself anyway.

But if the folders are deleted or renamed the menu doesn't work.

John Carmack (Quake series, Doom series, and the upcoming Rage engine) chose to use openGL rather than Direct-X -- that's saying something. Also, Call Of Duty series and Far Cry are notable games/series using openGL. OpenGL appeals to me (as an end-user only, I know nothing about programming) because it's free, non-proprietary, and is managed by the Khronos Group who also manages openCL. And openCL seems much more significant, at this point, than openGL -- using the graphics card to significantly speed up general computation. At this point it's CUDA (which is arguably better, at the moment) or openCL -- free and works on everything or proprietary and only works on Nvidia hardware. I realize this is a convoluted and indirect reason to like openGL, but I want the software to succeed that's not owned by a private company that gets to dictate what hardware we have to buy and who can use the software. Besides, have you seen smallluxgpu (openGL rendering program)? Tell me that's not awesome.

Perhaps on the other ones, Google is using some internal optimizations to trim your model's vertex count. Maybe it can't do that on this one. One optimization might be to convert a rectangular face with several polys to two triangles, if they all have the same color or texture information. If you colored those polys differently this time around, then Google can't reduce the poly count anymore.

I was hoping to avoid the overhead of another modeling program if there was a 3D materials editor available that does what max and blender's material editor do and can import/export the model. After spending a few hours on the renderer for the first time, now I understand why to use an existing photo for a texture first. For the 2D normal map software, I'm leaning toward pixplant; make a fake ambient occlusion map in a photo editor from the pixplant's normal map, and combine the 5 maps into 1 texture. I don't know if using the make normal map SU plugin for a 6th map would be overkill or not, considering that it uses the model's geometry instead of a texture to make a normal map.