Related Posts with Thumbnails

« Help On-line | Main | Adding to the Revit Masses »

September 10, 2009


Feed You can follow this conversation by subscribing to the comment feed for this post.

Definately not impossible...
Additional references are not required.
Revit is already natively capable of this function. (with some minor setup for each project and with the exception of elevations and live sections) Live views and drafting views can snap to a pre-determined locations:

I posted this on AUGI a while back and have been using this method on various projects ever since.

The trick for plans is to push the titleblock around via parameters.
The trick for drafting views is to push your detail around to match the titleblock location.

I stand corrected. I'll admit though these capabilities aren't obvious to most and you qualified it as "digging deep" into your bag of tricks. All said I am continually impressed with the clever solutions. thanks for pointing me to the post as it captures a lot of the requirements.

I don't think the titleblocks should be view-type specific, since no plan is the same from project to project.

When pressing and dragging a view, the current alignment option is fine. From sheet to sheet, I wouldn't have a problem with being able to align plans in the same place all at once, so you don't have to change each sheet for every time you adjust the size of the plan.

I'd really also like to be able to snap to elements in the view. Drafting views are a good example. I've got standard casework sections that are not easily aligned to another because the objects in a drafting view don't have an automatic centerpoint, similar to plans and elevations/sections. We are stuck aligning the view title and that means nothing when trying to align the drawings.
Ideally, I could use the align tool, and align one drafting view's floor line to another drafting view's floor line on the same sheet. I'd then align the view titles later. Currently, I place the views almost on top of the other, line up the floor lines and move them apart.

Being able to align a reference plane in any view to a line in the titleblock would suffice for most of my needs.

I don't mind placing a few reference planes into the model for alignment purposes, especially since I can make them unprintable.

I think an Align Views tool should be pretty ambivalent about what we choose. Let us reach through the viewport and pick two elements that define how to move the viewport into alignment with one another.

If you had a dummy view (of the floor plan) set up that could be placed on multiple sheets you could place the required view over the top to align and then remove the dummy view. The dummy view would also need to have the ability to be copy/pasted on to other sheets.

Work flow would be: -
create floor plan
duplicate as coordination plan (dummy plan)
place floor plan view on sheet in required location
place dummy view over top of floor plan and align using current methods
copy dummy view and place on next sheet
place new floor plan over dummy view in next sheet and align using current methods

Or a more simplistic approach would be to be able to cut/paste between sheets

Work flow would be: -
place floor plan on sheet
place another floor plan on same sheet and align with first floor plan
cut the new floor plan out and paste aligned into new sheet

Revit would only allow this to happen if the floor plan being cut/pasted does not appear on another sheet.

I would like to see the following:

Ability to place views in specific locations. Where the view origin is does not matter as long as its consistant

Ability to have right justified view titles. Many frims use these with good reason and the left only makes more work.

The ability for multiple views to be referenced together ie. interior elevations are usually grouped as a single detail with sub numbers for N-E-S-W etc. However this really needs the elevation tool to be fixed to allow this to happen.

Better sheet formatting options for schedules

Polygonal viewports or better masking

The frosting would be autonumbering based on position like Archicad. their method is really slick. Lowest on my priority list for sheets.

Finally... this issue seems to be getting some much needed attention. My first "support requests" pre-date Autodesk's acquisition of Revit. And subsequent SRs and "suggestions" during Beta sessions, have been all but ignored.

That said...

Jeff Shaver's approach is clever, but limited in application.

Some kind of "view handle" that could be accurately placed (and locked) by the user in each view, and that could then be referenced (ie, grabbed, or aligned to) when in sheet space, would make view placement less of a chore. I would suggest that this "handle" be a new, unique object type (ie, not a reference plane)for easier filtering, etc. In my view, view titles should still align with others on a sheet, but independent of the actual viewport.

Schedules are a different animal. The ability to "grab" a schedule's grid (eg, a corner or intersection) when in sheet space, would allow for accurate placement. But stacked or adjacent schedules still need some way of aligning columns and/or rows. Why can't we specifically define column width and row height within the schedule properties? My current workflow is to set up a work grid for each sheet, with sufficiently small increments, and zoom-zoom-zoom [in Thin Lines mode] to eyeball the little "blue carrots" into faux alignment... way too time consuming!

Ultimately, Steve Stafford's "ambivalent alignment" would be the best solution.

I also agree with Chris Hubbard's "need" for right-justified view titles. After all, as long as we use paper print sets, they will usually be bound along the left edge, and having the view title and number closer to the free [right] edge, for quicker visual location, only makes sense.

I have also "requested" a move-to-new-sheet function, similar to PaulB's comments.

Thanks for finally getting around to discussing this PITA, thorn-in-my-side issue.

I was going to make a suggestion then I read PaulB's post - but here goes as mine is subtly different:
It would be good to be able to place a view on a sheet, then select and cut or copy it; On the next sheet, when you paste aligned, if you cut it then it would place the same view in the same spot, but if you copied it, then it would place the dummy view; the user would select the dummy view and a list of other views would appear in the Type Selector, so the correct one would replace the dummy view.

We currently use scope boxes to define a consistent boundary for similar views (then line them up by eye with a spot on the titleblock) - if there is a limitation on what can be snapped to in a view, then scope boxes are a must (along with reference planes).

We also need to be able to crop drafting views - this would greatly help with view alignment (but also for hiding rubbish in linked dwg files). If this were possible, then it would be good to be able to place (2D) scope boxes on the drafting views, and be able to copy them and paste aligned to other drafting views so we could have a consistent origin/cropping boundary for when we have the ability to exchange one view for anoher on a sheet.

On the simplest of terms... it seems just activating alignment of model objects/reference planes/grids/etc. while view is not activated would go a long way. Then a user could simply have an invisible line(s) in their titleblock (or temporarily draw a line(s) on the current sheet) and align a view via self made lines and model references.

Regarding 'guidelines'... what makes a guide line different than a reference line/plane? I don't think that another object type is needed in that respect.

On a different (and probably more complex) view... it would seem to make sense (think REVITNESS) that when a sheet has been created and a plan view has been placed on sheet, Revit would "know" the alignment of said view with respect to said titleblock... then in the next sheet, when a similar view (plan/elevation/section) is placed it would snap to that same alignment (and give visual feedback of what the view is aligning to). Exploring a bit further... when hundreds of sheets/views are in play... you might want to shift a view to better fit. Revit would prompt "Hey, you moved a globally aligned view, do you want to move the rest the views to stay globally aligned with this view?" or something similar. Expanding the story again, someone may want to select the global alignment to intentionally move all views and they would then not get the prompt and Revit would be “Revit like” and move all the views accordingly. Maybe this still requires a new object type, but it would be transparent to the user for the most part... They wouldn't need to know the global alignment element existed, because the prompt helps them accomplish what they need to.

Most of the other suggestions seem CAD like to me, not that there's anything wrong with that, but there's got to be a better way of placing things initially without having to do it twice to get it right. (i.e. place a view, then align/move it) Why can't I just place it right to begin with. In Microstation we could tentatively snap to something then enter two offset coordinates to get it right the first time. This mindset/workflow of placing an object then align/move it to where you really wanted it, has got to be one of the most frustrating learning points for users of CAD making the switch. I've been using Revit for 4 years and I still despise having to move things after I've placed them.

Thanks all for sharing these detailed responses. Reading through them all there are many consistent needs and requirements as well as some clever suggestions. I also appreciate the sharing of workarounds which are very informative. The tentative snap..yes..Accudraw is a powerful feature.

On updating other views to match an alignment I think its required that there be a choice - that aligning other views be an implicit rather than automatic to avoid large unwanted changes.

Allowing Revit to globally move views [on sheets not currently onscreen] is a recipe for disaster. The user may not realize what such an action does to every effected sheet, and could easily result in overlapping views (or notes, schedules, legends, etc.).

In an earlier post, my idea for a "view handle" as a unique object, I think, would make it easier on the coders. An object that could "exist and operate" in both model and sheet space, but is part of neither. The problem with using model elements as alignment objects, is making Revit understand that the user doesn't want to move the reference line (or other element). And what if the same reference line (or other element) is used as an alignment object for other views on the same or different sheets. Do the views start a tug-o-war? Which view wins?

I think that what Troy and Steve said above are the simplest and most flexible ways to go about this. Erik, thanks again for showing interest in another long running concern. It is appreciated.

I don't really see the need to 'align' objects on the sheet. That is, I don't think there needs to be a link between the two, that if one moves then the aligned views move also. How about just letting us move a view from point to point with snaps? Wouldn't that be incredibly simple, or does it not work because it looks too much like AutoCAD?

I think that would help a great deal yet there would still potentially be a case where many plans on a significant number of sheets need to all be adjusted in the same manner. This would need to be done carefully of course but there is a good potential for time savings.

One more thing to keep in mind...

It might not be all that long before we all want to align 3D-Iso views (details anyone?) on a sheet, a la an exploded construction assembly.
Therefore, any coding that takes place should also consider that eventuality.

As far as views having a tug of war, I don't think it would be wise to allow locking of an alignment from sheet view to model object (if that is the chosen path), for that reason. I also don't buy the recipe for disaster so much... although a very valid concern. The way I see it if the views were globally aligned to each other some way, they were meant to be that way. Hence the prompt...(yes user tend to be click happy and not realize what they click on, but that's not any different than any other Revit prompt or warning that I keep having to decipher for other users). Views would only be aligned if a user said they were to be... which is including drawing intent. Yes, there would be some due diligence in your use case, but the way I see it the benefits of providing such intelligence far outweigh the risks of overlapping views (that can be overcome by a simple glance & nudge). This argument is like Autodesk saying, "We're not allowing you to schedule unconnected wall heights, because it's a recipe for disaster due to edited profiles and attachments." I see that as dumbing down a product rather than making it more intelligent. There's probably some legal reason for all of this that I don't care about but my opinion (without legal restrictions) is let the user decide where to place their risk in the data exposed.

Wouldn’t the fact that a view is not active, indicate that you want to move the view and not the elements?

They very notion of planning a feature because it makes it easier on the coders just is so wrong on so many levels, as long as what’s being proposed is not a project stopper. If you want something great, you should conceptualize without consideration to coding complexity… of course I say this with very limited programming experience in macro like projects ;D

I don’t disagree with the view handle idea, just the logic that’s used to make it seem like a better option. It really makes no difference to me what Autodesk comes up with for sheet view alignments.

It would be very flexible and yet powerful to provide both snaps/alignments from the model and some other automated intelligent approach. I don't have all the answers on a global like feature (for instance what to do with differing scales of similar views)... just thought I'd throw it out there to chomp on. In fact for this firm we really don't "need" to align views at all let alone do a lot of them (we get maybe 30-50 sheets in a set and maybe half could be aligned). The lack of alignment was questioned at first but didn't really cause anybody a headache. We moved on quickly.

I am really surprised this many people think it is a necessary (or even partially desired) feature. When i do production work in Revit- and even when i did it in AutoCAD- this was never something any of the firms i worked in have worried about.

A LOT of variables have to be considered.

1. Re: Snapping to the model itself for alignment(reference planes). What happens when the Reference planes move in the Model? Does the view move as well?

2. Re: Having cordinates on the sheet of X and Y, like Powerpoint does with images... Where are said coords taken WRT the view? It would have to be the "center" since we have flexible crop regions. Unless you made a global crop region setting as well.

2b. Same issue with having a new item called a View Reference, or an Alignment Line, or whatever. How does THAT behave with respect to a view that can change size? BTW, we have this already: The Viewport Title.

3. Out of my own idle curiousity- Do a lot of users really plot, then overlay based onthe sheet boundary? Seems to me even on the nicest of printers, i end up looking through the sheets to make sure theyre PERFECTLY aligned, at which point im not sure there is much point.

On the whole, what i DO think would be beneficial, is being able to snap to the Modeled/Drafted viewport, when the viewport is not activated. It would allow those that WANT things aligned, to align them with ease. It ALSO has a lot of other benefits: Placing text on sheets over 3D iso's and other views often means placing them OUTSIDE the viewport, so you cant snap. I find it really intriguing the way items placed on the SHEETS in ACA still know to query the DRAWINGS in the viewport. We could make a good use of that in Revit as well.

I normally take the stance that Policy must always lean towards the more conservative and more wanting (after all, if they HAVE it, those that dont want it wont use it, but those that dont have it CANT use it), but this is one feature that i would hate to see a large undertaking get wasted on, when there are so many more important arenas to focus on.

-Rotate/Mirror Project still dont work correctly
-arc'd line based families in RAC (like beams in RST, but for all catagories.).
-plennnnnnnnnnnnty more.

Please Factory people do not spend time working on 2d documentation as frivolous as this. If anything is to be worked on that is 2d related fix the text tool to be as versitle as AutoCAD's text tool. Like Aaron above, work on the items that will promote BIM, IPD and real-time collaboration...

Point taken on the cleverness or amount of effort on a solution. The future AEC deliverable will be the model right?
I will caution again on making too many correlations with what is posted and discussed here vs real factory resource allocation. In some instances there are other items that really don't need further discussion - the needs are well known.

I sure hope that the factory isn't spending much time on this. There are many other features and enhancements that have been suggested that would be much more valuable than this. IMHO when it comes to lining up views on sheets, close is good enough.

Anyway, it seems that most of the suggestions and even Eric's initial post are over complicating the matter (This is Revit, not AutoCAD). A simpler solution could be something similar to "orient to other view". This idea would use the layout of a previous sheet to layout a new sheet. On the new sheet the user would select a view and right-click on it, in the context menu could be an option "Align to other sheet/view". This would launch a dialog with a list of sheets and views in the project, selecting a view from this list would move the currently selected view to the same location of the sheet. There could even be check-boxes to copy the view title location and crop boundaries to the selected view. I don't think it needs to be any more complex than that.

As suggested above, we do need the ability to right justify stuff in view title families.

Also as suggested above, simply having the ability to object snap to model objects or grids when moving views would pretty much solve the view placement issue.

Non-plot lines and text sure would be nice, but there is no need to add new object categories just for sheet layouts.

I'm also of the opinion that this is a minor bug-bear. I would not like to see the added complexity and hardware/RAM requirements of Revit needing to keep a record of the references of lots of views. I just want to be able to have my floor plans line-up and a simple copy/cut/paste would be fine for what I want.

In terms of overall features of the software I would prefer to see more time spent on how we can build a model in the easiest possible way to replicate what is likely to be built on site. My aim for the documentation (Sheets) is for it to be read and understood, easily, by the contractors to avoid any mistakes on site. That said, the question was relating to aligning sheet views and so from my standpoint a simple, low-tech solution is fine by me.

As I read this, I thought of the idea of "sheet underlays". What if you could underlay a sheet complete with it's views under another sheet? Then the underlay would allow new views to snap just as overlaying views on each other does. This would ensure that plans, RCPs, etc all fall at the same position on sheets. I thought this might cause printing issues if someone forgot to turn off the underlay, but then just make "sheet underlays" never print. (or selectively not print) Just a qucik brainstorm....thoughts?

That is an interesting idea. In evaluating possible solutions to small features (there are always many) I prefer ones that are simple and leverage existing concepts. If there were many sheets though it might take more steps than simply snapping to a guide or line in a Title block. You would first need to choose the underlay then perform the move.
The earlier mentioned "Align to other sheet/view" is a simple idea. Get one view aligned then translate it to other views on sheets via a simple UI.

Totally agree with Anthony. If EVERY visible object (both modeled, drafted, and overlayed) in every view and sheet possessed complete "snapability", then there would be no urgent need for any sort of custom alignment tool(s). Ten throw in a 2d "non-print" system family construction drafting linetype (ala "defpoints") and we can set this up accurately all day long. Simple problem, simple solution.

Also, I move to strike Bruce's comment from the records. So long as plans and information will be presented in a 2d format, we will have a desperate need for top notch 2D annotating and presentation tools. A huge priority for the Factory should be (and should have been) maturing Revit's 2D annotating and presentation tool sets.

I'm with Steve Stafford, allow us to use the align tool on geometry to align views (any type of view) to other views or sheet elements.

Allow a non-printing visible element in Titleblocks (ref lines in a TB family are snappable, but not visible).

Allow view(s) to be aligned sheet to sheet, so that floor plans stack, elevations line up whatever. I think something similar to "Publish Extents" would do nicely. It would be nifty to "publish" a view's location on sheet, and be allowed to select the views you want from a list dialog, then select the related sheets, and have Revit place and align all the views on the all the sheets in one operation. An "update feature" would be useful too, not sure I want stuff "locked" to each other, though that could be an additional "option" that users could employ if desired.

Hi Erik,

As with others here, the alignment of plans and details on a sheet is, for me, a very low order issue.

It would be kinda nice from time to time, but I don't think it will save any meaningful time in the design and documentation process for me. As you are well aware, there are many, many other wish list items which would add significant efficiency benefits to Revit above and beyond what plan alignment might do.

Nevertheless, it is nice to see the factory looking beyond the immediate horizon.

The comments to this entry are closed.


  • Subscribe