Hello, could you tell us a bit more about your experience with SketchUp + vectorworks?
Does the program import textures correctly?
If there is a change in the design, can you update the model, keeping the same reference?
Is it possible to make a solid section with vectorworks, differentiating several materials with different fills? (as in Skalp)
Hello, could you tell us a bit more about your experience with SketchUp + vectorworks?
I haven’t gone too far into testing this as yet. I have just updated to VW 2018, and it imports SU2017, does not import SU 2018. Models seem to come in generally well, textures seem to come in perfectly including transparency. Theres the option of importing SU items as meshes / 3dPolygons or Architectural items, both methods will take some comparing to decide what to use. I would imagine if the design changed the model would need be reimported and replaced. From what I tried I could not get VW to fill the sections created, I would guess this is because VW is a solid modeller and would see those objects as hollow.
Well, I’m happy to report that LayOut just sped up considerably for me. I thought I’d share as it might help others in the same situation.
The other day, out of desperation, I thought I’d try moving my 16MB LayOut file from my Synology NAS onto my workstation’s local drive.
To my surprise, LayOut is considerably quicker with the file installed locally. It is still laggy, but it has become usable.
On the surface, the improvement does not seem to make much sense. I access the Synology through a Zyxel 802.11ac router that has a 1733 Mbps throughput at 5 GHz per the manufacturer. That should be plenty enough to handle the limited I/O one would think LayOut generates, and in fact the same hardware setup and connection work just fine with SketchUp itself, as well as with far more data intensive tasks.
Maybe LayOut has some kind of issue handling SMB / Samba?
In any event, I can now work in LayOut while remaining relatively sane. It has by no means become blindingly fast, but text edits that used to take 5 or 6 seconds now take one second, so there is definitely a significant improvement.
Well… Here I was thinking 1 sec was slow. I work on a SSD but still have to zoom into texts to have them editing faster.
@JQL, how big is your LayOut file? How many pages / viewports?
Mine is pretty small, only 5 pages (24"x36") and 13 viewports.
I think this is a very interesting conversation. It’s all constructive criticism. Probably difficult for the developers to read but a lot just has to be said.
+1 for “Layout has tremendous potential in the industry”
+1 for “i would pay a lot more if it worked well”
- for “LO can’t do large and complex projects quickly or reliably”
LO does appear like it is created for, (and presumably tested with) relatively small and simple models. (carpentry projects, maybe a single level house, etc?).
So what is Layout designed to achieve? It seem to be like LO attempts, but doesn’t do a great job of being:
- a competent 2D drafting package
- a document production/desktop publishing (eg Indesign, Publisher) or publish-to-web app (sketchfab).
- a raster or vector illustration tool (eg adobe photoshop, adobe illustrator, corel)
- a rendering solution (vray, indigo, maxwell)
- a presentation package (eg Powerpoint or Keynote)
- a way to import/exporting CAD/3d design information in an accurate way
- a tool for project collaboration, design markups & review
- a project information output tool (scheduling, bills of quantities and materials, costings)
The functions listed above are all things that most design firms need to be able to do. However in many of these areas, LO barely meets professional standard of output.
So could LayOut do some(if not all) of those things much better ? To our simple (non-programmer) minds, it feels like a few tweaks and additions would go a long way.
LO’s Drafting tools, for example - why not take them directly from SU? (and include some extras, eg from the 2D Tools extension). Same for the Material browser, Entity Info, Components Layers…etc.
LayOut feels like it borrowed its toolset from Powerpoint 2010
LO has been getting better recently, in subtle but important ways…so my fingers remain crossed (for good luck). A number of architects, landscape architects, civil and structural engineers, surveyors, builders, etc, on the Beta testing/development team would be super-important to achieving this We’re a group of overworked professionals with little free time to contribute to forums etc, but the occasional online rant seems to be a good way to vent .
Haha I used to post scathing rants over at the Revit forums when my frustration levels had risen beyond acceptable levels. Now that I have made the commitment to remove Autodesk from my architectural life (Fusion 360 for product design is great!), the stress created by the thought of having to use Revit or Autocad for the rest of my career has evaporated.
I’m hoping the next 2 layout releases will be huge leaps forward in terms of tool-set and performance.
The thing is that I believe that, when it started, Layout followed the idea behind sketchup. A simple 3d tool to create simple and fast presentations in a world dominated by 2D cad, which, honestly, noone understands but architects.
As a simple presentation tool for sketchup models where simple annotations and dimensions could be added, a title block could be added to each page and some easy scrapbooks would make it look professional Layout wouldn’t need much actually. After all people were drafting in CAD, then would present things in Sketchup, but never would abandon CAD.
So, Sketchup and Layout were meant to be sidekicks, managed by Google.
The problem is that they are fun to work with and when I first saw that Layout could present scaled 2D sketchup scenes, create dimensions and annotations, when I understood that sketchup could create sections and I learnt about the Section Cut Face plugin, I opened my eyes and ditched away 2D CAD.
I guess many of us followed the same logic and Sketchup+Layout started to be pushed beyond what they were meant for.
Sketchup, if you remember well, wouldn’t cope well with big buildings. Nowadays it does, but it had to evolve.
It was under Trimble that 64bits came up. Our computers got unchained in Sketchup. It can handle more geometry and more textures now. I can now work with shadows turned on with only a slight wait, and there’s no building it can’t handle… (Though it still can’t handle trees, or heavy realistic chairs, but that’s what render proxies are meant to do.)
The basis is still there: intuitive and fun and accurate with an incredible inference system in 3D. Much better than 2D CAD imho.
In the meantime Layout had already been created for Sketchup thinking in Sketchup as a sidekick and not as the main basis for our architectural work.
So, the real problem for Layout was a mismatch between reality (what the users started doing with it) and intention (being a fast presentation tool for simple sketchup models).
Considering that now, what users want, is a full featured and truly pro solution for presenting 2D output all the way to construction documentation. I think a full revision of Layout should be implemented. However, this revision should go so deep that it would probably be better to have a new addition to the Sketchup Pro package called Documents (or whatever).
Sketchup would have:
- Sketchup - for 3d models;
- Style builder - which I never used in my life;
- Layout - which I’m forced to used but isn’t exactly what I need.
- Documents - which wouldn’t follow the Layout logic of beautiful presentations but a pro logic of immaculate tools for construction documentation and 3D model data extraction.
Maybe layout could also be documents, but I think that would require a massive backstage change and some feature drop in terms of viewport handling.
I have a suggestion that I’ve tried describing before: what if Layout would loose the 3D viewports and would stick to a 2D image of the sketchup viewport?
I’ve noticed that a very complex 2D drawing in sketchup (the equivalent to a model section with full details) renders in a fraction of a second in Layout, while the same output rendered directly from the original sketchup model, takes seconds if not minutes.
So, this means, the solution for Layout could be to have Sketchup convert the 3D scenes into 2D drawings that would open in a new kind of Layout viewport that could not be orbited around. Updating it would mean a regeneration of these 2D drawins and not a full 3D model inside a viewport.
Then Layout would be fast and useful for our standard architectural output and development team would be able to focus on tools for drafting, annotation and data extraction.
Sorry I couldn’t write a simpler post but I had no time for that…
I made a post, even more complex than this one, that would describe these 2D viewports:
The following image is the folder for a project for the refurbishment of 2 apartments in the center of Lisbon:
A you can see:
- I have several SU files for each apartments (general model, window openings, demolition drawings and original apartment model…);
- I also have several Layout files for each apartment (general drawings, partial drawings, details demolition drawings, openings…)
- The biggest file is 486Mb but I think there might be an unpurged model there… I will check it.
EDIT: The biggest file includes the reference for the 2 apartments and I cannot purge the one I don’t need. Maybe there’s an layout viewport hidden there somewhere that prevents me to purge the file.
As this is a set of files for each apartment, it’s a simple project.
If I’m reading you correctly, the new type of viewport you propose would essentially be a static display of a SketchUp scene. It could not be manipulated in LayOut, but it would reflect changes to the underlying SketchUp model.
In other words, any changes to the viewport content would have to be effected in SketchUp, and would appear in LayOut after the model is updated.
I personally think that’s an excellent idea. Most viewport instances used in construction docs are scaled and static anyway, like floor plans.
It would not even have to replace the existing viewports, it could just be available for anyone who needs it.
What kind of impact would such a static viewport have on the overall performance of LayOut? I wish we could hear developers’ feedback.
Very nice project!
The scope of the project and the file sizes should inform product planners as to the usage-model envelope LayOut should be able to support. Although you’d think they should know, seeing the prevalence of architecture and construction imagery in Trimble’s own website, tutorials, videos, and marketing materials.
You’re reading me correctly. We never want to mess our layout viewports (change camera position or other settings). As far as I’m aware most pro users do the swme so they also don’t treat viewports as 3d models with all the flexibility that implies. They use these viewports as static 2d scaled drawings.
So, 2d viewports simplified as much as possible, in raster, vector or hybrid mode, is all we need and even if the export from sketchup could take some time, from my testing all other Layput work is so much faster that the main Layout issue could be solved.
Thanks, but that is not the project, only some spaces and openings from one of the 2 apartments. What corresponds to the file in question.
I wholeheartedly agree, this is my workflow as well.
+1. Never changing the view in Layout - only using scenes from sketchup
Agreed. And scenes need better method for management. A 2d document of an architectural project can easily have 100+ ‘views’ (though most are Top or Side views)
Yeah this annoys me. It becomes difficult to navigate and check everything.
To combat this I create a series of linked models to limit the number of scenes required.
Model A - the primary design model where all modelling, texturing etc work is done.
Model B - contains a referenced version of Model A which has sections and scenes for plan views
Model C - contains a referenced version of Model A which has sections and scenes for elevation views
Model D - contains a referenced version of Model A which has sections and scenes for section views
etc. etc. I find this keeps things tight once you get used to switching back to Model A to change things. It also helps speed up layout a little bit as each of the “drawing” models is imported into its corresponding layout template (Plan, Elevation, Section etc.) to prevent any of the individual layout files from becoming overloaded.
This is made even more efficient by using these amazing plugins: www.lasuapps.com/
I’m currently rebuilding my templates to incorporate my new design process (including a new render engine, design and presentation VR, contractor 3D model, etc)
If it works well I may post something about it. But it wont be for a few months yet.
I’ve been thinking along these lines very modestly for a while and starting to do more with it. One thing I do it for is I move the model to a document that has the sun set for 45° shadows as viewed looking toward the red-blue plane. Then I create rotated copies of the model so all elevations face south and set scenes for each elevation so every side, even north, gets proper, beaux arts style 45° shadows.
I have looked at that, but it messes with your location on earth and time of day setting. Because of that, he warns that you should do this with a
copy of the model, which brings us back to @benoldays workflow of copying or referencing your model into another file.
I don’t like the idea of having sections in one model, plans in another and elevations in yet another.
I like to switch between sections and plans in the same model so I can fine tune both in fas iteractions, specially while detailing them. I’m constantly switching from a plan to an elevation and model so I can’t spread them all over.
So, what I do is breaking the model in chunks… Like wings or rooms or special parts that I can take out from it and develop in separate parts and detail those parts, then bring back those detailed models into the main generic model.
However, the idea is always the same, how to deal with so many sections and scenes in order to describe all parts of the model.
This is especially hard on the refurbishments we’ve been doing, where there is no orthogonal wall, most windows and doors are unique and every new bathroom, kitchen and closet is special.
Honestly, sketchup allows for an excellent modelling workflow and terrible documentation workflow and I’m not talking about layout here, but about:
- scene management - grouping or hiding scenes would be a must;
- double scenes for each section - in order to heal section cut lines that are wrong but derive from intricate models;
- not being able to crrate multiplanar sections - especially on non ortho buildings some elevations and interior elevations require multiple sections;
- not having a section manager - as sections are special scenes that are usually presented in standard ways I don’t see the need dor having them in scene manager. A section manager would suffice, where activating a section would activate it’s section fill and we could edit this section fill manualky, add details to it and avoid all the fuss involved in the scene management.
I’m with @JQL here, I’m flicking between plans / elevations / sections & model and would find it impossible to have it all separate models. I like to work with a master model and if there is a particular room/ annexe that need attention I’ll pull that out, work on it and put it back into my model master. That said, I’m just about to try Eneroths new thingy - the reference manager, looks interesting.
@JQL ‘s other suggestions are also well thought out features, that sound very SketchUp like - so yeah, wot he said…