I am currently taking a class in Autodesk Revit, taught by an architect who works as autodesk and is helping to develop new features for the software...
Just started the class, and this is my first time working with parametric modelling, but I wonder: Could this stuff have the potential to make autocad obsolete somewhere down the road?
I mean, here you are actually modelling the entire building, its object oriented, details get worked out for you (with your own control over heirarchies for structure etc.), and its like combing a drafting a modelling software into one package... Its working with real building components, sections are almost point and click...
While it might not be as powerful as other 3D software for concept design... It makes construction documents a snap...
Is this the future of drafting? Are offices moving towards parametric modelling like Revit?
(didn't we have a similar thread before?)
Revit works well, if you know what you are doing. It will cause you a lot of troubles, if you use it for renovation works or like fast track construction where there will be a lot of changes along way. Revit is very, I mean VERY smart program that requires a lot of planning ahead, like floor to floor heights, materials, wall types, etc. Once I tried to move base floor height for multi-story building after bidding set stage, and oh man, it was hell.
how would it be more difficult to change floor heights in revit vs. autocad? doesnt revit make changing floor heights so much easier (basically a click and drag), given that its parametric...? that objects are smart objects, floor heights, walls, etc. adjust automatically, and are connected, have some structural intelligence in terms of the heirarchy?
why would this be more work than conventional autocad?
Unfortunately this "stick" allows architects to make catalog buildings even faster. I remember an old essay I read a year ago that was written in the mid 80s which pretty much decried computer drafting saying it would lead to architects having less understanding of the components of their designs. I refuted it, saying that in CAD good designers would come up with there own details and not use cad block standards. And even when standards where used some understanding was still required, but with Revit or ADT it seems like its too easy to just click 6" wall rather then encouraging innovation. I'm not saying it can't be done in these programs just that I percieve the response of the profession will be to develop more and more architecture based on a small set of materials and building methods.
Don't get me wrong I love parametric modelling and I do feel all CAD software will at some point be parametrically based ( I believe the next release of microstation will have parametrics) but I wish these programs were more opening to customized object creation.
or..maybe I'm just blowing this out my ass, I've only had a couple of hours with Revit.
bRink:
This will be so true, as Suture said, that I might had that problem because I didn't know HOW to swing the stick.
Ok, here was the problem: in Revit, when you create detail drawing, you create a viewport in the section view of the model, and then you literally draw your detail, just like the way you draw in AutoCAD. You can choose the option so that when you cut a wall section, you can see insulation and studs and all the fun stuffs, but in some problematic areas, like where roof parapet and roof meet or floor and wall meet, you need to clean up some mess and draw some over it. Problem was that when you select everything and move the entire model like a couple of inches, all the viewport I created didn't move with the model, so that I ended up REdrawing all the details again. Plus, floor cut line, which creates floor plan, didn't move either, so it was just big mess. Why didn't I just change the floor height number? well, Revit is so smart that they don't let you do it. I guess I might have been able to getaway with some cheating, but then there will be no advantage of using parametic CAD software.
OR
maybe I just didn't know it better. :) hey, I'm only a human.
I definatly think so.
Ful if your havin issues try askin some questions here: http://forums.augi.com/forumdisplay.php?f=93
Pretty awsome group of people that know an awful lot about the program.
Programs such as this increase the involvement of the archtect allowing them to work on the project not just hand it off on a drafter. Personaly I find its improving my construction vocabulary as you are dealing with a virtualy building not lines and layers. So far in the office im at we've found the most usefull aspect of Revit in making revisions and keeping everythin coordinated. Having a 3-d model to go with the drawings for cities and clients is nice too.
As to the customization there isnt much you cant model and schedule. Of course you have to know the best way to go about creating the model, make a new one edit an existing one etc.Not to say it doesnt have its quirks, its still a young program and you hit some snaffu's. But i have yet to use a program that doesnt.
BIM - Building In Modeling has been coming for a while. GSA is going to require it in 2008. So yes - AutoCAD will eventually become obsolete in that respect.
ArchiCAD has been doing all this for ages. It's a great program for modelling and drafting. Why have a bazillion DWGs with oodles of doubled information which all has to be changed manually every time there is a change? With a properly modelled ArchiCAD building every change will be reflected in all the plans sections elevations (and details if they are set up correctly) automatically. Walls, roofs and floors all have elevation and sectional materiality. I just don't think I could go back to producing CDs from AutoCAD again. It's just too painfully slow.
Anyway, Revit was started by a bunch of programmers at Autodesk who had seen the light. Autodesk freaked out and just bought the company outright.
If you're gonna use Revit, etc. make sure you can convert it to dumb ol' ACAD, or make sure that your consultant disciplines are good at Revit, as well. Like so much in this profession, it all breaks down when no one's on the same page.
I have been reading about what Archi-F mentioned about the GSA. People are speculating that the whole AE industry will basically be forced to move to BIM apps in the next 5-10 years because clients will expect it, or even demand it like the GSA, in the same way that clients now expect us to use AutoCAD.
Funny, it seems that we all read last month's issue of "Architecture" regarding BIM. I'm kinda dumb, that's the first time I ever heard the term 'BIM". It seems like a new acronym for something that's been around for about 15yrs - attributes & parametric design.
I really look forward to seeing GSA try to enforce BIM and fully interactive parametrics. (this should prove an interesting guffaw) Everyone including the Revit designers knows that for details we are still supposed to be using AutoCad or similar programs. Revit was designed primarily for taking you from Schematics to Design documents, with modeling processes for evaluating design in various manners, energy efficiency, costing, rendering etc. I think they'll have to force all engineers to use something like Solidworks as well, for each floor, imagine that, either that or every engineer has to have the 7,000 dollar Autodesk Building Systems software on their workstation and pay 2000 dollars a year for the training, also losing 10,000 dollars a year because of loss of speed due to complexity of the interface. Couple that with the fact that most engineers (muchless architects) are too busy worrying about codes, permits, clients, and other real-world issues, add in the fact that these software companies have a new "baby" every 9 months (whereas some buildings take 2 or 3 years to design) and eventually you have a backup. Even now most senior architects and engineers can design a building really well, but are not as fast with the software...but I don't think it is possible to simultaneously train everyone under 30 with the "perfect software" and really expect them to have a clue what they are designing or drawing in the REAL world. Perhaps my next reply will be optimistic, for now I vote pessimistic. I won't be wasting my time OR money bidding any GSA contracts.
Jun 28, 05 11:42 pm ·
·
Block this user
Are you sure you want to block this user and hide all related comments throughout the site?
Archinect
This is your first comment on Archinect. Your comment will be visible once approved.
Revit / parametric modelling: the future of practice?
I am currently taking a class in Autodesk Revit, taught by an architect who works as autodesk and is helping to develop new features for the software...
Just started the class, and this is my first time working with parametric modelling, but I wonder: Could this stuff have the potential to make autocad obsolete somewhere down the road?
I mean, here you are actually modelling the entire building, its object oriented, details get worked out for you (with your own control over heirarchies for structure etc.), and its like combing a drafting a modelling software into one package... Its working with real building components, sections are almost point and click...
While it might not be as powerful as other 3D software for concept design... It makes construction documents a snap...
Is this the future of drafting? Are offices moving towards parametric modelling like Revit?
I know of a few that have, at least for a project or two, but I also know of one that it didn't work out well.
The idea is great, and it will happen sometime, but not very soon.
(didn't we have a similar thread before?)
Revit works well, if you know what you are doing. It will cause you a lot of troubles, if you use it for renovation works or like fast track construction where there will be a lot of changes along way. Revit is very, I mean VERY smart program that requires a lot of planning ahead, like floor to floor heights, materials, wall types, etc. Once I tried to move base floor height for multi-story building after bidding set stage, and oh man, it was hell.
fulcrum:
how would it be more difficult to change floor heights in revit vs. autocad? doesnt revit make changing floor heights so much easier (basically a click and drag), given that its parametric...? that objects are smart objects, floor heights, walls, etc. adjust automatically, and are connected, have some structural intelligence in terms of the heirarchy?
why would this be more work than conventional autocad?
Its not about what stick you swing
its about HOW you swing the stick
Suture - very true.
Unfortunately this "stick" allows architects to make catalog buildings even faster. I remember an old essay I read a year ago that was written in the mid 80s which pretty much decried computer drafting saying it would lead to architects having less understanding of the components of their designs. I refuted it, saying that in CAD good designers would come up with there own details and not use cad block standards. And even when standards where used some understanding was still required, but with Revit or ADT it seems like its too easy to just click 6" wall rather then encouraging innovation. I'm not saying it can't be done in these programs just that I percieve the response of the profession will be to develop more and more architecture based on a small set of materials and building methods.
Don't get me wrong I love parametric modelling and I do feel all CAD software will at some point be parametrically based ( I believe the next release of microstation will have parametrics) but I wish these programs were more opening to customized object creation.
or..maybe I'm just blowing this out my ass, I've only had a couple of hours with Revit.
bRink:
This will be so true, as Suture said, that I might had that problem because I didn't know HOW to swing the stick.
Ok, here was the problem: in Revit, when you create detail drawing, you create a viewport in the section view of the model, and then you literally draw your detail, just like the way you draw in AutoCAD. You can choose the option so that when you cut a wall section, you can see insulation and studs and all the fun stuffs, but in some problematic areas, like where roof parapet and roof meet or floor and wall meet, you need to clean up some mess and draw some over it. Problem was that when you select everything and move the entire model like a couple of inches, all the viewport I created didn't move with the model, so that I ended up REdrawing all the details again. Plus, floor cut line, which creates floor plan, didn't move either, so it was just big mess. Why didn't I just change the floor height number? well, Revit is so smart that they don't let you do it. I guess I might have been able to getaway with some cheating, but then there will be no advantage of using parametic CAD software.
OR
maybe I just didn't know it better. :) hey, I'm only a human.
I definatly think so.
Ful if your havin issues try askin some questions here: http://forums.augi.com/forumdisplay.php?f=93
Pretty awsome group of people that know an awful lot about the program.
Programs such as this increase the involvement of the archtect allowing them to work on the project not just hand it off on a drafter. Personaly I find its improving my construction vocabulary as you are dealing with a virtualy building not lines and layers. So far in the office im at we've found the most usefull aspect of Revit in making revisions and keeping everythin coordinated. Having a 3-d model to go with the drawings for cities and clients is nice too.
As to the customization there isnt much you cant model and schedule. Of course you have to know the best way to go about creating the model, make a new one edit an existing one etc.Not to say it doesnt have its quirks, its still a young program and you hit some snaffu's. But i have yet to use a program that doesnt.
BIM - Building In Modeling has been coming for a while. GSA is going to require it in 2008. So yes - AutoCAD will eventually become obsolete in that respect.
ArchiCAD has been doing all this for ages. It's a great program for modelling and drafting. Why have a bazillion DWGs with oodles of doubled information which all has to be changed manually every time there is a change? With a properly modelled ArchiCAD building every change will be reflected in all the plans sections elevations (and details if they are set up correctly) automatically. Walls, roofs and floors all have elevation and sectional materiality. I just don't think I could go back to producing CDs from AutoCAD again. It's just too painfully slow.
Anyway, Revit was started by a bunch of programmers at Autodesk who had seen the light. Autodesk freaked out and just bought the company outright.
Wake up and smell the parametrics: www.graphisoft.com
If you're gonna use Revit, etc. make sure you can convert it to dumb ol' ACAD, or make sure that your consultant disciplines are good at Revit, as well. Like so much in this profession, it all breaks down when no one's on the same page.
I have been reading about what Archi-F mentioned about the GSA. People are speculating that the whole AE industry will basically be forced to move to BIM apps in the next 5-10 years because clients will expect it, or even demand it like the GSA, in the same way that clients now expect us to use AutoCAD.
Funny, it seems that we all read last month's issue of "Architecture" regarding BIM. I'm kinda dumb, that's the first time I ever heard the term 'BIM". It seems like a new acronym for something that's been around for about 15yrs - attributes & parametric design.
I really look forward to seeing GSA try to enforce BIM and fully interactive parametrics. (this should prove an interesting guffaw) Everyone including the Revit designers knows that for details we are still supposed to be using AutoCad or similar programs. Revit was designed primarily for taking you from Schematics to Design documents, with modeling processes for evaluating design in various manners, energy efficiency, costing, rendering etc. I think they'll have to force all engineers to use something like Solidworks as well, for each floor, imagine that, either that or every engineer has to have the 7,000 dollar Autodesk Building Systems software on their workstation and pay 2000 dollars a year for the training, also losing 10,000 dollars a year because of loss of speed due to complexity of the interface. Couple that with the fact that most engineers (muchless architects) are too busy worrying about codes, permits, clients, and other real-world issues, add in the fact that these software companies have a new "baby" every 9 months (whereas some buildings take 2 or 3 years to design) and eventually you have a backup. Even now most senior architects and engineers can design a building really well, but are not as fast with the software...but I don't think it is possible to simultaneously train everyone under 30 with the "perfect software" and really expect them to have a clue what they are designing or drawing in the REAL world. Perhaps my next reply will be optimistic, for now I vote pessimistic. I won't be wasting my time OR money bidding any GSA contracts.
Block this user
Are you sure you want to block this user and hide all related comments throughout the site?
Archinect
This is your first comment on Archinect. Your comment will be visible once approved.