If your model is simple architectural shape, but with a lot of underlying detail. You could export jpegs and then make a photo-matched version. Can add in as little or much detail as you would like.

If you want to get more accurate textures then instead of appying through PhotoMatch, you can jsut export matching elevations and use those to texture yoru building.

Another thing to consider. Can you or someone else go back and take the photo again. This time noting where they are standing and how high the camera is off the ground. If you use the sun location and time of day, you could even figure out how much pan and tilt the camera has.

Then you just make that same camera in SU, and use the image as a watermark.

You can also just bring your model into Layout.

From that you can export a PNG, but better yet, if you are using vector or hybrid mode (new to Layout2) you can create a PDF and open that in Illustrator. You will have a vector lines and also a rastor image for faces if in hybrid mode. Now you can alter the line styles in Illustrator, or turn lines on or off, modify them, anything you would normlaly do in Illustrator.

I've put in a request for better layer control so for example all lines would be on the same layers they are on in SketchUp. Now they are all in one layer, but are grouped to some extent.

I would think you could make a DC that has one sub-component that is named "tile fixed size" that is as the name suggested a fixed module. Then there would be another tile named "tile trimmed" that would be used for the left over space. You could set if there should be two equally sized trim pieces on each end, or full tile on one side and just one trimmed on the other.

If instead of linear, this is a matrix you would have more tile types.

• tile fixed
• tile trimmed x
• tile trimmed y
• tile trimmed xandy

So, I was just ran across this thread with similar issue with large size model and a lot of 6-8mp images slapped on building facades.

I did a test with some new SU7 features. (Make Texture Unique, and Edit Texture)

Here is a simple building that was 2.6mb with single image on surface.

Step1. - Select texture and make unique, then purge unused. File size now down to 332k without any resolution change.

Step2 - Now I used the "Edit Texture Image" and resized in Photoshop to 512pix wide, and saved with compression of 6. Now file size went down to 108k.

All and all, this is pretty quick, in particular if I skip step 2 and only do step one. Can just jump from one image to the next, don't have to reload anything. Unfortunately it does appear I have to do this one texture at a time. I wish I could select multiple textures at once and make them unique.

Now my workflow for site like this would be to save each building (or each block, depending upon the scale and level of detail) out as a component. Then I would make sure to save a raw component that uses the high res image, that is not yet unique. This way I can still play with the pins to make more accurate later on. then I would make a small component to use in the main site plan.

The way you solve 1. is similar to how you solve 2.

You put dimensions that you want on each view on their own layer. Then for the scene that you are going to use them on set layer visibility to show the correct dimension set.

You can create a dummy scene that does nothing but change layers, and then use second scene to control camera view, shadows etc. In layout I can chose one of these, and then choose the other one. They work with each other so long as one doesn't over-ride the other. For example the scene with your camera view also has layer settings saved.

The approach of cutting section through model and then manually drawing detail elements is the workflow we use in Revit. It has the same update issue, but seems to be the best way to do it for now.

You could always create a surface at the section cut and use raster image with the pattern you want.

Gaieus is correct, but could be a work around. If your A4 pages, and then want some A2 pages, you can change your layout for the new pages and work on your layout. The border around the A4 pages will also now be A2 but it doesn't mess with anything you have drawn.

So waht this means is that you will have to toggle your layout back and forth when you want to correctly print either the A4 or the A2 pages; but for editing and page layout they will happily exist in the same document.

Best workflow if they really need to be in the same document would probably be to just setup to the larger page size, and then have a mask saved in Photoshop to batch crop the smaller sized sheets. May also want to ask why they ahve to be in the same document. You can also do single document to lay everything out and then split into two docuemnts (ok, make a duplicate copy of file and deleate the A2 pages from that file and the A4 pages on the other copy of the file)

If you save to PDF your Layout generated objects and text should remain vectors, as well as any SketchUp viewports that are set to vector or hybrid.

If you open this up in program like Adobe Illustrator you can do a lot of manipulation of the linework. You can even open in Illustrator and export back to dwg file at correct scale.

I should point out, that this was not by Sweet Onion Creations, but rather by a local z-corp printer.

I was very happy with the level of detail, but felt our model fell short in a number of areas that it didn't have to be.

• Surface shouldn't be so splotchy. Was later told it was because of the timeline, but never discussed issue with us in advance.
• The seam could have been made much better
• The base was warped, and not square, not not really sanded well or at all. For cost of work, might as well spend more time on the base.

At any rate, I posted this as a demonstration of the level of detail you can get (our walls are about an inch tall) but also so you know not all printers are the same, even if using the same machine.

@chris fullmer said:

And for the onclick event, they all trigger simultaneously. So I have one onlick that starts animating a custom attrbiute in EVERY object that moves or hides. So all the textboxes have an attribute that animates 1-60 over 60 seconds. So one number per second. Then on the first text bubble, for example, it is set to be hidden if its "animator" attribute is less than 1 or greater than 6. So it turns on at 1 second intot he video and off at 6 seconds. Or something like that. THen Sangs bubble turns on at 7 seconds and off at 13. And so on. It took a lot of pacience to set them up. Especially sometimes while making all the timers count right, sometimes one or two would get out of synch and would start from 60 and count down while all the others were at zero counting up....yikes!

Thanks for describing that. This could be very useful for making presentations. I just wish I could set the camera location and attributes in a timed DC.

As far as making this easier to code. I could see this as a good ruby script down the road that let the user enter more basic data, and the ruby does the hard work.

By the way, LayOut2 can export to vector PDF. And I have experimented with opening this in Illustrator. So if you need to drop this into larger presentation, or just want to use some of illustrator's brushes, this is an option. I have asked Google to do some under the hood organization of how data is layered in the PDF, so that selected elements to edit in Illustrator would be easier.

If you want to get fancy, you can rename your layout file to have a ".zip" extension, then open the xml file with a text editor and you can do tons of cool things. I had a minimal post on this back in SU 6, but I'll try to play around in SU 7 and see how it works.

in concept could use excel or mail merge on world to batch create Layout documents from a bunch of different files.

Thanks, I haven't tried large numbers, but I suppose I could add an if function if the number of rows and the frame/mullion widths won't fit in the width of the overall DC.

Also after posting mine to 3DWH I saw that one of the newer Google DC tutorials has a window with screen where it adds rows and columns on its own as needed. They also have it setup where user can turn elements on and off, and pick materials. Its funny what is under the hood can be so different depending on how the DC programmer approaches things.

So I've been meaning to do this for a while and finally got around to it. You can set frame with depth, window depth, mullion width and depth, and also how many rows and columns in the system.
http://sketchup.google.com/3dwarehouse/details?mid=4b3c72c1b3c4b238cb3c55835215aed4

The overall size and depth can be set by "component options" dialog or scaling the DC.

If I add an attribute to have separate width and depth for horizontal vs. vertical mullion, and then place this on the ceiling, then all of the sudden it can be beams and joists.

Ideally I'd like work doors into this DC, and also thought of having the DC determine the number of rows and columns on its own based on user entry for min and max window size. The DC would add or remove rows or columns as needed and also keep things equally spaced.

Let me know what you think. Another feature could be to do this on a curved wall; a lot more math involved in that.

As far as the behavior when that box is unchecked. It has to do with how SU and Layout deal with field of view when the aspect ratio of display changes. It pays attention to the vertical field of view and tries to maintain that. So if you print from SU to letter in landscape vs. Tabloid in landscape you get the same vertical coverage, you just get more model on the sides for the tabloid print.

Similar in Layout. If you widen the view no changes are required to maintain the vertical field of view. While if you make it bigger vertically it needs to adjust the mode. Same thing will happen if you drag a corner diagonally (unless in doing this you make the aspect ratio wider vs. taller)

I think attribute naming conventions would be great, and where better to develop and communicate these that at SketchuCation.

One reason that hasn't been mentioned yet for importance of attribute naming, is that if you use the "swap component" feature to bring in a new component that attribute data will carry over for matching names.

This is different than choosing reload compant and choosing a different file. The "swap comment is new to DC's and how it deals with attributers is how it differs from jut reloading a different file.

Let me give an example. Lets say I metal widow with attributes like "frame width, frame depth, window width, window height, window depth". I have already customized a window to suit specific needs, but now want to change to totally different type of window but with same dimensions. If I swap one window for another with same naming attributes, all of the attributes will carry over and I'll be done. If I don't have matching attribute names, I'll have to resize everything about the window.

Only odd thing here is that for example I wouldn't want "manufacture" to carry over, since the new DC would have a different manufacture that I want to keep. Maybe in SU7.x they can have option for specific attributes to make them "not swappable".

The other thing I'd love to see is ability to create DC's in SketchUp that have matching Revit Families. Then using the standard BIM format (Can't remember name maybe "IFC") could port things over to Revit and save a lot of rework. Using this method the actual SU geometry isn't imported, but rather one DC is swapped for a family using an "IFC Exporter". Revit already has an import option for this, so if SU could export it then this woudl seem possible. This to some extent solves the problem of curves in SU being segments. Since I'm not bringing them into my Revit Construction Documents it doesn't matter.

I don't think walls and floors are well suited for this. I would draw those in Revit. But everything we set in the building from sinks to casework, lights, furniture, windows, and doors would seem fair game for this. Again attribute naming and structure wold be critical for this.

@lewiswadsworth said:

Remus, even with the advancements Layout is a toy. My architecture students have nearly universally preferred Google Docs' simple "Presentation" application to Layout for any number of perfectly valid reasons besides simple stability, and no firm I have worked with uses anything less than Illustrator or InDesign for formal document preparation or Powerpoint or Flash for presentation.

Its worth looking at Layout again. I don't believe their time was wasted in developing advancements to Layout. FYI - with the vector or hybrid export you can open PDF's in Illustrator to change,hide or alter the appearances. Which gives you amazing control of line styles and weights. Complex patter brushes, artistic brushes, dashed lines, etc.

FYI - even with Layout ver1 I created 40 pages SD documents. Sure it was pushing Layout ver1 pretty far and I had to be patient. But this workflow saved me tons of time by avoiding going to Photoshop and InDesign.

Film and stage plug create physical cameras. That would give you what you need use as a reference for your section plane. Not sure if this plugin is still on SU site or available somewhere else?