Hellnbak -- I'm sure you're getting enough cooling, but not the most efficient. More open doesn't equal cooler. Here's an interesting article on air cooling setups.
Smaller more confined spaces direct air current over the parts that need to be cooled where larger, more open cases have to move a lot more air. There's also the issues of pursuing a negative or positive pressured case.

Plus... there are some pretty cool cases if you look around.

I use Thea's unbiased engine almost exclusively, but I believe the biased engine has been much improved as well. Does anyone have more experience using it? How does that compare to V-Ray?

I work at an Arch. firm but we almost never get hired for visualization -- the rendering we do is more to help design and persuade the client of what we'd like to do. For us, Thea makes more sense since it's pretty cheap. V-Ray might not cost too much -- for a specialization firm -- but for us it's not worth the cost.

Solo, could you explain why you model faster with it? My understanding is that it helps your orbit and move around the model better, but those are some of the basic strengths of SketchUP. I've never (other than zooming into very small things away from the origin) had trouble with this. The biggest lack, for me, in SketchUP are a lack of good selection (loop, adjacent, etc.) tools -- although this seems to being addressed by plugins.

I could see a 3D mouse being more useful in other modeling/rendering software (3D Max, Revit, Blender, thea...) that focus more on tools than on orbiting.

If I'm wrong, though, I would love to here it.

The idea of licensing births seems like a practical response to Malthusian Eugenics. How do you set standards for who would receive a license? We have a difficult enough time with things like evaluating teachers... how do you evaluate the worth or practical ability of someone to bear children?

At the risk of being overly sensational this has been tried before.

What's going on with the sharp shadow-line on the ceiling behind the TV wall?
The floor is also wider-plank than I'd expect, but not unreasonable.

The biggest factor keeping it from photorealism is grunge and mistakes. Clutter, that kind of thing.

i7-2600k processor is excellent -- the 3.4 speed is nice and fast but that's somewhat irrelevant, since they're so easy to overclock you could get it to 4.5 with little trouble or risk to the processor itself. I'm not going to try that myself until I start to notice things not working as quickly as I'd like, but it's nice to know you have it you need it...

I'd take 16GB RAM. You might not need it, but you'll probably use it more (renderings and loading textures in models) than even a typical heavy user. Plus, RAM is so cheap (relatively) that it'd be silly to have such a nice system and go cheap on a basic component.

GTX 570 is a great card. I went for an AMD and have had absolutely no problems with it (despite it's bad reputation with SketchUP) but that was at least in part because I like the company more and they seemed relatively comparable other than that. Nvidia is the standard choice.
I would completely avoid the Quadro cards -- not worth the money. From a hardware standpoint they are almost always LESS powerful than the cheaper, gaming counterparts; their advantage is that in specific programs (3d MAX being a major one) the drivers are VERY optimized. SketchUP is NOT optimized and wouldn't see the advantage. They also are supposed to be a little more reliable, but I've never known graphic cards to be unreliable. I'd rather have the gaming card now, and if it broke in a year or two, you could use the money you saved and buy a then top-of-the-line card and be even better off.

Storage looks good. The SSD card is big enough you could store all your programs on that and data on the SATA drive. I've heard that would speed up loading all programs... but you might want to check on that -- maybe there is an advantage to only the OS there.

Last thing -- I don't know if you're planning on having this built for you or building it yourself, but having just built a system myself it's not only very easy, but informative. I'm very aware of what parts my system has and what it's capable of.

Good luck!

It's also bad practice to be painting on back faces. In sketchup it doesn't matter, but if you export to other programs (rendering, usually) colored back-faces can be a problem. When that happens it's a pain to fix.
It might be a problem you never come across, but why do it?

@ thomthom

Thanks for the correction.

Congratulations! It looks great.

Do you mean that it's not tiling correctly, or that the pattern is so obvious it's a VERY noticeable tile?

You can solve the first in GIMP -- there's a filter to make seamless. It's a pretty cheap job, but works if you're in a hurry.

If it's the second, and you're worried about noticeable tiling, I would definitely do it by hand. The first step is to run a high-pass filter over the texture; this evens out the contrast and saturates it some. This is good to eliminate shadows and bright spots.
After that, you can re-colorize the image, offset x and y, and clone stamp (or whatever method) to make the image seamless. Good luck.

Here's an update that will continue the running bond pattern vertically -- massimo's was tiling, but stacking for the first row.

Oops... I'm an idiot! I just noticed I had seen it incorrectly. I would have sworn I had seen "enmissive".

sgbotsford, I'll try to answer some of these as I can.

@unknownuser said:

1. Layers had three controls -- visibility, lockability, and inference ability.
2. A locked layer did not interact with new geometry. This should speed up SU on large models.
3. Being locked but inferenceable would allow you to use faces and lines as endpoints, as well as sources for making guides.
4. A non visible layer did not interact at all.

Layers do have visibility and lockability, although being locked just means it can't be edited or deleted, still inferred to, though. I'm not clear on why turning off the inferencing of locked objects would be desirable. If I need to I just hide an object. Also, I tend to use groups or components (depending on if it will be instanced or not) the way I think you use layers. I usually put all of a type of thing in a group and then edit that group (graying out the rest of the model) when I need to edit it. Nested groups are not a problem.

@unknownuser said:

4. Layers have a hierarchy, and are not just in alphabetic order. Turning on and off a layer does so for the layer members.

This does sound nice and you might already be able to do through the outliner (which I have never bothered to use). Also, you could do a similar thing with nested groups, as above.

@unknownuser said:

I found in map making that for a given class of features it was useful to have up to 3 layers, one for points, one for lines, and one for areas. So for example when working with the hydrology group, points would include dams, stream junctions, coordinate points of a corners. Lines would be stream courses, areas would be lakes, ponds, swamps. Being able to lock the lakes layer made it possible to trace watercourses through the swamp without snapping to edges, center points and so on.

I don't think I understand the purpose of this one. As an aside, SketchUP doesn't use points (at least, they're not selectable natively).

@unknownuser said:

5. Layers have alternates depending on resolution.

I'm not sure quite how this translates in SU. In MM I worked with aerial photos as my base layer. Since I was mapping a fairly large area (for a personal project) -- about 200 square kilometers using 1 meter/pixel resolution, and this with a computer that had 500 MB ram and a 1 GHz processor. It was a while ago.

Anyway: MM had a feature where you could turn on/off layers depending on the current zoom level. So I reprocessed my images into 5 sets at 1, 2, 4, 8, and 16 meters per pixel. At large scales, I used the low resolution images. At small scales I used the finer resolution. This meant that at any given view point, only a few megs of phototiles needed to loaded.

It sounds like you're talking about mip maps -- sketchup doesn't do this. It might be a nice feature but I would just as soon export to some other program for visualization.

@unknownuser said:

If SU layers could have simplified geometery when sufficiently distant, this could speed up the display substantially. (From a sufficient distance a sphere is a cube)

This is useful and might be a nice add. I think you would need to either build a low res model yourself or every now and then have SU calculate (I image huge problems/errors resulting from this) the low res models for you. A better solution would be to use the proximity (similar name) plugin that basically does this. You might also have better luck talking to some of the plugin developer guys to see if you can get one of them to make a better plugin for you.

@unknownuser said:

6. Layers have styles.

This would allow you to do things like turn of textures for the floor surfaces layer, while leaving it on for the countertops and cabinets layer. Or have one layer with hidden edges, one layer in wireframe, and one layer in solid with no edges showing. Or set a layer to have only 15% opacity so it is only ghostly visible.

This would be nice.

@unknownuser said:

Zoom behaviour.

As far as I can tell the model of zoom behaviour is that the view point is some arbitrary distance away, and zooming acts like zooming a camera lens. The position of the camera is constant, the angle of view changes.
This makes working with interiors tricky. So far as a newby, I've had to put exterior walls in a layer of their own, and turn them on/off as needed.

I would prefer the reverse: The angle of view remains constant (but setable) and so zooming was handled by moving the point of view. By doing this, as you zoomed closer you would move through the wall into the interior

(Maybe there is some clever plugin that does this.)

Sketchup zooms to the mouse cursor. If you're point at empty space it zooms really fast. If you (preferred method) point at an object and zoom, the zooming slows down dynamically as you get closer.
Zooming shouldn't change the FOV, although there is a command where you change the FOV by zooming -- do you have that command activated, maybe? Normally, I choose a low FOV for exteriors and raise it when doing interiors. Zooming is actually very convenient once you get accustomed to how SU works.

@unknownuser said:

Multicore use. I don't think that this is unreasonable. There is a lot of locality in CAD. Items on the right side of the screen don't interact point wise with items on the left, so it should be possible to parallel-ize a lot of the computation. Dedicating a processor to textures. Dedicating on to shadows Dedicating one to compositing. It should be possible to effectively use a dozen cores.

I don't know the technical details, but multi-core seems to be always requested and the SU developers tell us it would be a LOT of work for very little performance improvement. They say there are more efficient ways to increase performance and they are focusing on those. Not sure either way, but SU has been getting noticeably faster with each release... so I tend to believe them.

@unknownuser said:

64 bit is more problematic. In the mac world 64 bit gives a process the ability to address more than 3-4 GB of memory. At least on the benchmarks I've seen it doesn't make the process faster. Not sure if it's the win that many people claim.

SU is now large address aware, which, as I understand, effectively allows use of more than 4gb memory, even if it's not officially 64-bit.

...

Lastly, since it sounds like you are still learning SU, can I recommend this tutorial?
It's technically a tutorial for exporting SU to the Crysis editor (which is VERY interesting, by the way) but it does a great job of establishing best practices for SU use, including texturing and grouping.
Good luck!

Option 1:

1. import image into autocad -- scale approximately
2. find a known dimension in the image and draw a line (e.g. a 200' long building)
   a. on the image (on the 200' portion of building
   b. a nearby line in model space the actual dimension (a 200' line approximately parallel to line in image)
3. select image, use align command, connecting line on image to scaled line
   a. use only 2 points -- select to scale the image
   Done!

Alternatively, my office uses BlueBeam (PDF editor alternative to acrobat), which has a calibration option where you select 2 points in an image and type in the distance between those. Then, all the measurement/editing commands will be scaled correctly.

Excellent work!

This does a great job of presenting the professional side of sketchup. I love the relationship between the perspective and orthographic drawings -- nicely done!

Also, when drafting switched over to computers there wasn't a fundamental shift in the way things were done. The same drawings were presented in the same way with the advances being in ease of use in creating the drawings (copy being essential here) and editing drawings (move, erase). These were conveniences, though, but didn't change the way design was communicated.
Drawings like yours (BIM in general can, but rarely does) present a new way of understanding the project, in a way that you couldn't (practically) do without a computer.

I look forward to seeing more of your work.

I can't help you with VRay, but that's a VERY reflective/shiny roof. Clay tiles would have a much more diffuse, almost imperceptible reflection.

Thanks, although I'm not really worried that it lists 1 less than actual. I just checked and the limit is 64,000 vertices -- so being off by 1 is irrelevant. I mostly wanted a tool to see how many object groups I could combine for single export.
If I think about it I'll change the code when I get home tonight -- but either way it's working for what I need it to do.

I actually used the code offered by sdmitch (after having separated the command as thomthom suggested), and saved it as a plugin rather than having to cut+paste whenever I wanted to check.

Gaieus,

wiki as in she'll take in whatever any random person tells her?
Let's hope for an authoritative review in the future...

j/k

I finally had a chance to try this out and it works (almost) perfectly!

Almost because for some reason it always returns 1 less vertex than is actually present. For example, a square plane shows 3, a cube shows 7, etc.
This is a complete non-issue, though, since I'm just trying to make sure the number is less than (I think) 25,000.

Thanks again!