A new home for SketchUp
-
@jason_maranto said:
Taking those things into consideration you have to say that for the most part direct 3D modeling will eventually become minority concern. With the wide availability of creating a huge range of 3D images without learning any real modeling skills already a very real phenomena at places like DAZ/Poser.
i think you're cutting out a huge swath of 3D world.. it's not just for creating 3D imagery.. it enables you to completely 'build' something prior to actually building it... sort out problems, develop a sequence/schedule, finer material orders, etc..
as far as creating a model of something that already exists then yeah, i can see that dying out.. we're probably not too far off from '3D cameras for everyone' which will more likely than not, produce superior results in a fraction of the time it takes to create the same model today (fraction of a second vs. days
)but as far as modeling new ideas, i'm not convinced it can be automated..
(that said, i am sure the way this stuff is done in the future will be a lot sweeter than today.. once we get rid of the mouse/keyboard/monitor paradigm) -
...... don't forget fingers, hands and arms, just brains in jars full of some strange greenish liquid
-
@mike lucey said:
...... don't forget fingers, hands and arms, just brains in jars full of some strange greenish liquid
lol..
i don't know.. it's just awkward using computers the way we do.. (and if you think about it, things really haven't changed much over the past 30 or so years regarding this.. and i highly doubt they got it right with the first try
) -
@unknownuser said:
i think you're cutting out a huge swath of 3D world.. it's not just for creating 3D imagery.. it enables you to completely 'build' something prior to actually building it... sort out problems, develop a sequence/schedule, finer material orders, etc..
I think this is fairly well covered under item #3 (parametric modeling). At some point in the future anybody (as in layman) will be able to tell a computer exactly what type of buildings, rooms, furnishings, etc. they envision -- the computer will just adapt the design in realtime, and most likely be able to run complex simulations from this data that we can only dream of. Manual modeling will be redundant (for the most part). Obviously we are some ways off from this, but it is coming... there are already many primitive parts of this in place.
For example, as primitive as they are, Dynamic Components were a step for SketchUp in this direction.
Best,
Jason. -
An area where SU shines, but is very unknown is 3D Printing.
Last summer I attended a seminar on the new version of SolidWorks. If I had the money, that is one app I would buy! (Wash my mouth with soap
)However, at the seminar were 2 3D Printers displaying all sorts of very intricate objects, most with intertwined parts that were molded in place, not assembled after printing. One printer was able to make models with a volume of upto 2ft X 3ft X 1 1/2 ft deep, in multiple contiguous materials from soft aluminum to hard plastic to very pliable silicone, all integrally bonded, like a 1 piece wiper blade. It used inkjet technology and upto 7 material cartridges. Needless to say, I was blown away, but at $250,000 I had to put my credit card away.
What's this to do with SU, you may ask?
I asked about what the software could handle in terms of complexity, and the answer was
"whatever you can create in Google Sketchup and scale to fit"
then pass the SU file through a (still in development) translation app to create the 3D "print" file.It seems that SU was simplest in creating the models , while SolidWorks was better at engineering the models for conventional computerized machining. At least for now anyway.
-
@jason_maranto said:
@unknownuser said:
i think you're cutting out a huge swath of 3D world.. it's not just for creating 3D imagery.. it enables you to completely 'build' something prior to actually building it... sort out problems, develop a sequence/schedule, finer material orders, etc..
I think this is fairly well covered under item #3 (parametric modeling). At some point in the future anybody (as in layman) will be able to tell a computer exactly what type of buildings, rooms, furnishings, etc. they envision -- the computer will just adapt the design in realtime, and most likely be able to run complex simulations from this data that we can only dream of. Manual modeling will be redundant (for the most part). Obviously we are some ways off from this, but it is coming... there are already many primitive parts of this in place.
For example, as primitive as they are, Dynamic Components were a step for SketchUp in this direction.
Best,
Jason.Speaking from my experience with residential (and some light commercial work), as well as custom timber design:
Until there is a sophisticated set of algorithms in place that can take a long form series of essay questions (plus hundreds / thousands of yes / no questions), parse the data, watch the expression and emotion on a client's face as you discuss how they've answered those questions in reference to the images they've shown you of 'what we like' - then adapt that all to budget, climate, local codes, etc. etc. etc. - hands, minds, eyes, and brains from designers collaborating with their clients should still play a huge role in designing spaces for human interaction and occupation.
Otherwise we'll have what? Android for home and offices? A mess of conflicting software / free ware / ad based design tools that the 'layman' (whom I very much respect, but want to help guide) can cobble together? We already have a physical environment that can resemble the worst MySpace pages (I'm looking at you, strip mall developer, and I'm looking at you, lady with all sorts of plastic crap from the last 20 years strewn on your lawn).
Pre-fab, sure, plug together your house from a reputable designer / builder in a web based app, let the model inform the manufacturing in terms of parametric modules that get assembled at the shop in the computer and then output to cut lists, material choices, pricing, and then show up on site.
But, no thanks for some future of fully automated design.
I love the idea of parametric modeling (use it a lot on the timber side of things - wall / window modules, custom timber CNC module, etc, for AutoCAD), and I'd love to see DCs grow up - but simply inputting a bunch of data into a spreadsheet and have a building pop out, no thanks.
-
BMike said (in part) but simply inputting a bunch of data into a spreadsheet and have a building pop out, no thanks.
But doing just that can give you the basis for a building that conforms to a set of rules/standards and dimensions for which designers/architects/engineers can now tweak into what the client wants.
This will avoid continuous re-do's when some standard was violated or overlooked during the fomenting stage. That tends to raise the costs, take extra time and pisses off the client, because it happens every time.
-
Yes, please understand I'm not talking about the need for customer service or good taste/design to go away -- just the need for alot of the manual labor of model building. Lets face it, the world is very much a pre-fab place these days, with a limited amount of options for virtually anything.
Best,
Jason. -
Don't be surprised to see more affordable 3D printing soon.
Not big parts but 6"x6"x6" parts.@jgb said:
An area where SU shines, but is very unknown is 3D Printing.
Last summer I attended a seminar on the new version of SolidWorks. If I had the money, that is one app I would buy! (Wash my mouth with soap
)However, at the seminar were 2 3D Printers displaying all sorts of very intricate objects, most with intertwined parts that were molded in place, not assembled after printing. One printer was able to make models with a volume of upto 2ft X 3ft X 1 1/2 ft deep, in multiple contiguous materials from soft aluminum to hard plastic to very pliable silicone, all integrally bonded, like a 1 piece wiper blade. It used inkjet technology and upto 7 material cartridges. Needless to say, I was blown away, but at $250,000 I had to put my credit card away.
What's this to do with SU, you may ask?
I asked about what the software could handle in terms of complexity, and the answer was
"whatever you can create in Google Sketchup and scale to fit"
then pass the SU file through a (still in development) translation app to create the 3D "print" file.It seems that SU was simplest in creating the models , while SolidWorks was better at engineering the models for conventional computerized machining. At least for now anyway.
-
@jason_maranto said:
Yes, please understand I'm not talking about the need for customer service or good taste/design to go away -- just the need for alot of the manual labor of model building. Lets face it, the world is very much a pre-fab place these days, with a limited amount of options for virtually anything.
Best,
Jason.[off:32afqmm4]Probably OT[/off:32afqmm4]
Yes but as you said--understanding how the tech changes will affect our work and, not the least, how much we get paid means bringing up these questions. Would deserve another thread.
-
@unknownuser said:
Don't be surprised to see more affordable 3D printing soon.
Not big parts but 6"x6"x6" parts.Damn...I want one. Too bad I need the money for boring stuff like living.
-
So, does SU create .STL formatted files?
And did you see the price of the ABS filament? $21.50 / lb YIPES!!
Time to cannibalize my weed whacker.
-
@jgb said:
So, does SU create .STL formatted files?
No but it can be converted through a magical process. I did manage to somehow get these http://www.shapeways.com/mymodels on shapesways and get one printed....hmm I forgot
-
@jason_maranto said:
I think this is fairly well covered under item #3 (parametric modeling). ……..
For example, as primitive as they are, Dynamic Components were a step for SketchUp in this direction.
oh.. right..
yeah, i actually use DCs in my own modeling for things like knee walls, flat decks, stairs, (and a couple of other things)..using stairs as an example though.. they're nothing really to do with my final product.. they only make parts of the product accessible and what not.. i've come up with a stout and generally pleasing way to build stairs that conforms to most local US building codes etc and they're a pain to draw new each time.. (as in, if the final height is 1" lower, i have to start (mostly) from scratch in order to redraw them).. so doing them as a dynamic component makes a bunch of sense. (which btw, is pretty hard in itself because sketchup's DCs aren't grown up as you say.. so it involves some workarounds/extra work afterward.. a ruby solution would be best but i don't know ruby)..
but, if my final product were the stairs, i'd get way too bored if i just built the same thing over and over again only with dimension tweaks.. each set would have to be different enough to warrant a fresh drawing each job.. or else, i'd have to find a new job
that said, many builders do that anyway.. and i do agree with you.. parametric models could be used to create drawings of 80-90% of what's being built out there today..
[off:151j2em7]and that should be a good thing as technology should be used to make our lives easier… more leisure time / less work time.. (however, when you throw in today's corps/banks/govts/etc, we seem to be working as much as we were prior to the last 100 years worth of technology)[/off:151j2em7]
so i believe you're right.. many,many designers/architects/engineers etc may find themselves out of a job unless they step up their game..
hopefully though, the creators of the world will stay ahead of the computers as far as coming up with new/better ideas so the more capabilities a computer gains, the more we will require from them (similar to sketchup's poly count race ).. eventually, when computers gain their own intelligence, even the last of the hanger ons will be put out as we'll no longer be able to design the best stuff.. (but maybe i'm going into too much of a scifi thing here.. not saying i don't believe this is where we're heading because i do believe so.. it just seems weird to talk about it at this time) -
@jgb said:
An area where SU shines, but is very unknown is 3D Printing....
I totally agree that Sketchup is excellent for 3D printing, and is not talked about much on this forum, but I think there are many non-members using Sketchup for 3D printing.
Here's a 3D printed project I posted a few months ago: 3D printed Lamp. You can see a few more of my projects here: 3D Printed Sketchup Models (website still in progress)
Here are some 3D printed dice designed in Sketchup that made a big splash in the MMORPG scene last year:
@jgb said:
So, does SU create .STL formatted files?
Yes, with a plugin. I like this one: SKP2STL
-
Hey Mike,
Thanks for calling this out. This is an important point, and we want people to understand what is happening with the 3D Warehouse. I am going to try and chase down some more detail here; I'll do my best to clarify, but I might be a bit hamstrung in what I can share because the sale of SketchUp still has not been approved.
Incidentally, that's also why us SketchUp folks can't comment too much on some of the conversations here.
Best,
Mark
@mike lucey said:
Just read through the - Transfer of licenses and personal data to Trimble - here, http://sketchup.google.com/usernotice.html
This paragraph has me a little confused!
[i]'Opting-out
-
Yeah I share that problem
My 1st color inkjet cost almost that much. It would be really good for custom Lego parts, you know, for the kids @marian said:
@unknownuser said:
Don't be surprised to see more affordable 3D printing soon.
Not big parts but 6"x6"x6" parts.Damn...I want one. Too bad I need the money for boring stuff like living.
-
Not on it's own but this is where I would use AutoDesk's 123D. It is a perfect fit for this.
@jgb said:
So, does SU create .STL formatted files?
And did you see the price of the ABS filament? $21.50 / lb YIPES!!
Time to cannibalize my weed whacker. -
-
@alan fraser said:
That works great except that SU still won't do a real curve no matter what format the file ends up in. Not sure how this will effect the end result as I have not tried a 3D printer yet. Thanks for the link though, seems to be a slick little plug-in
Advertisement