topic Re: The great divide in Installation & update

The great divide

Anonymous — Tue, 28 Jan 2025 13:54:59 GMT

Hi Folks,

As a very new ArchiCAD 9 user, I'd be grateful for a little assistance.

To try to get up to speed with things, I'm rattling through the Training Guide that came with our version of ArchiCAD 9, which involves drawing the Drucker-Brownstein residence.

Unfortunately, I'm having problem inserting some windows.

As part of the training guide, you are required to insert 4 windows on an equal spacing around the perimeter of an arc. However, selecting the division option does not give the correct result (the wall width at the ends of the wall are far greater than the gaps between the windows).

To try and resolve the problem, I've opened the next file in the sequence of training files where the windows appear in the correct locations; however, when I compare the location of the windows with the snap points, the windows locations do not agree with the snap points.

So, come on then guys, what's going on?

Is it me, or are you all rolling abount on the floor saying "he fell for that old nut shell"

I should also say, that I may be a new ArchiCAD user, but I'm an old, old hand when it comes to AutoCAD (if I'm allowed to mention that name in these parts), and with the exception of this little glitch, I'm seriously impressed so far.

Many thanks

John

Re: The great divide

Anonymous — Wed, 24 May 2006 07:52:09 GMT

John_C wrote:
however, when I compare the location of the windows with the snap points, the windows locations do not agree with the snap points.

I am not familiar with the tutorial project but the difference may be in which side of the wall you are snapping to, how the reference lines are set up, or differences in the snap points settings.

Without knowing the specifics, I can only suggest you check all these possibilities (wall reference lines could also affect things) and by the time you figure it out you will have learned a lot about haw ArchiCAD work.

I'm glad to hear you're enjoying the change so far. It will get even better as you go along.

Re: The great divide

Anonymous — Wed, 24 May 2006 21:59:27 GMT

John,
Does the training guide tell you that you can get the center
points of the four windows by setting the divisions to a certain value ?
As you know, the true center points for four equally spaced windows
are equidistant from each other but are different from the points
generated by dividing the length of the wall into equal spaces.
So I can't see how the division method could possibly work.

I believe you need to do it the old fashioned way by adding the widths
of your windows, subtract that from the length of the wall, divide this
number by the number of pieces of blank wall, then you have the
segment lengths between the windows. Add one half of the width of a window
to one segment length and this will be the length from the end of the wall
to the center of the first window and again use this same length to the center
of the next window.

In the case of curved walls, all of these lengths are arc lengths derived from
the arc length of the curved wall and the arc lengths of the widths of the windows.
You can get the arc length of the wall in the element information floating pallet
The formula for arc length is the following.
Arc length=PI times the Radius times the Angle divided by 180.
Where "Radius" is the wall radius and "Angle" is the included angle of the arc segment.
Using this formula you can get the center points for the windows using
the method described above but in arc lengths rather than straight lengths.
Using distances rather than divisions you are able to get these points to display
as distances measured on curved walls are arc lengths rather than chord lengths.

I may be completely wrong here, so please someone show us both
how this can be done using divisions, exempting some trivial case, of course.
Peter Devlin

Re: The great divide

Dennis Lee — Wed, 24 May 2006 22:43:33 GMT

For what it's worth, I had the exact same question at that stage of the tutorials not too long ago. I think what I did was just play with the division so that I divide the snap points to 8 or 9 along the arc, and put the windows in every other snap point, which got me close to what I would be satisfied with if I was the architect for the project.

Re: The great divide

Anonymous — Thu, 25 May 2006 01:25:08 GMT

Cheers for the replies folks.

From reading what you say, it sounds like the training guide is misleading, and whilst that might not surprise me, it does concern me as it doesn't show the product in a particularly good light.

With a training guide designed for new users, things really should work the way book the tells it.

As you say Peter, to achieve a balanced look, it is not as simple as dividing the length by 5 and adding 4 windows at the snap points. If the snap points were truly at the 1/5th points, then the gap between the extreme windows and the end of the wall would be half as wide as the gaps between adjacent windows. However, when I try to follow the guide, the gap at the end of the wall is more than double the gap between adjacent windows which is why I was scratching my head.

I might try using 20% as the snap distance, and see if that makes any difference (I know it should be exactly the same, but lets wait and see).

If anybody has any other bright ideas, I'd be happy to hear them.

Re: The great divide

Anonymous — Thu, 25 May 2006 01:54:24 GMT

John,
When you say " If the snap points were truly at the 1/5th points",
does this mean that when you set divisions to five that the
snap points are not at the 1/5 points ?
Are they not equal, or are they equal but not dividing the wall into
five intervals but some other interval ?

Does AutoCAD have some automatic means of distributing windows
in a wall with equal spaces between them ?

The reason I ask is I am wondering how you accomplished
this task before you started using Archicad.

Peter Devlin

Re: The great divide

Anonymous — Thu, 25 May 2006 02:53:38 GMT

Peter wrote:
I may be completely wrong here, so please someone show us both how this can be done using divisions, exempting some trivial case, of course.

It can't typically be done with divisions unless the width of the window is a whole number divisor of the wall length. Not trivial but rare (except perhaps for Richard Meier). It can easily be done with the multiply/distribute command.

Re: The great divide

Anonymous — Thu, 25 May 2006 03:12:24 GMT

Hello Matthew,
It occured to me to try the multiply/distribute command to
array hotspots or windows along the edge of a curved wall,
but it doesn't work.
Meaning, it does not put the hotspots in the right place
for the present purpose.
Could you explain how this is done.
Thank you,
Peter Devlin

Re: The great divide

Anonymous — Thu, 25 May 2006 07:50:57 GMT

Peter wrote:
... are they not equal, or are they equal ...

The with of the wall sections between the windows appears to be equal, but the start and end interval are far larger than I would expect to see.

As Dennis said earlier, it is possible to archive a similar effect to the desired one by setting the divide snap to 9, and using alternate snaps, however, this isn't what the guide suggests.

I think for now I'll assume that it's me being overly pedantic (just for a change 😉 ), and that the guide's at fault for being a little vague and woolly.

If you were using AutoCAD, there would be many ways of skinning this particular cat. The DIVIDE command inserts the required number of POINTS at the desired interval, or you could use the polar array command to position the objects around the curve segment.

That said, you still need to take Matthew's comments into consideration if you want to do the job propperly.

Many thanks.

Ps Have I mentioned that I can be pedantic from time to time

Re: The great divide

Anonymous — Thu, 25 May 2006 15:46:28 GMT

this thread struck me as interesting so I looked through the reference guide and found what I think may work for you.

I find if you draw a line in the center of your wall and then set your snap points to divisions of 4 in your case

then set the special snap points to be "between intersecting points"

This should put your nodes in the desired place along one side of the wall. Then you can repeat for the other. I hope this helps you.

Re: The great divide

Anonymous — Thu, 25 May 2006 21:35:19 GMT

I very very rarely have designs with curved walls, but i will take a "stab" at this one. This may look like trouble but it was not too bad. I am to lazy (and busy) right now to explain this in words though.

Did I mention that I almost never do curved walls. Too much of an engineer for the fancy curves.

Re: The great divide

Anonymous — Fri, 26 May 2006 05:58:03 GMT

Peter wrote:
Hello Matthew,
It occured to me to try the multiply/distribute command to
array hotspots or windows along the edge of a curved wall,
but it doesn't work.
Meaning, it does not put the hotspots in the right place
for the present purpose.
Could you explain how this is done.
Thank you,
Peter Devlin

Don't do it with hotspots. You need to use the element itself or something that matches it in width. Place one (to be removed later) at the end of the wall (off the end actually - just touching at one point). Multiply with the distribute minus one option dragging the outside corner of the element (the one not touching) to the far end of the wall (or space) to be divided.

I hope this makes sense. It is much easier to do than to describe, but I am too tired right now to make any pretty pictures.

Re: The great divide

Anonymous — Fri, 26 May 2006 15:52:38 GMT

Matthew wrote:
...It is much easier to do than to describe, but I am too tired right now to make any pretty pictures.

Hey, no slamming the Pretty Pictures. That is my "Bread-and-butter" here. I always have to attach a picture, because, one: I can not type very well and Two: I are a engineer and words don't come to easy for me.

No attached picture this time. Hey I am getting the hang of this....

Jay

Re: The great divide

Anonymous — Fri, 26 May 2006 23:53:17 GMT

Hello Matthew,
I tried your method and, amazingly, it works !!
Thank you for teaching me this.
For the general case one has to do some careful setup
but it is quick to do.

Thank you,
Peter Devlin

Re: The great divide

Anonymous — Sat, 27 May 2006 01:01:21 GMT

After learning how to distribute the windows using Matthew's method,
it occurred to me that one could get the divisions method to
work by stretching one end of the curved wall exactly one
window width longer and then setting divisions to five.
This will then show hotspots at one corner of each window.
After placing the windows one stretches the wall back to it's original length.

So the division method does work after a fashion using this "trick".
Thanks to Matthew we are all spared the horrible arithmetic
I had suggested in my previous post.

whew !
Peter Devlin

Re: The great divide

Anonymous — Sat, 27 May 2006 07:19:37 GMT

Peter wrote:
Hello Matthew,
I tried your method and, amazingly, it works !!
Thank you for teaching me this.
For the general case one has to do some careful setup
but it is quick to do.

Thank you,
Peter Devlin

It is my pleasure. Glad to be of service.

I find this trick to be useful in all sorts of circumstances, such as laying out muntin bars in custom doors and windows.

Re: The great divide

Anonymous — Sat, 27 May 2006 14:08:16 GMT

Peter wrote:
After learning how to distribute the windows using Matthew's method, it occurred to me that one could get the divisions method to work by stretching one end of the curved wall exactly one
window width longer and then setting divisions to five. This will then show hotspots at one corner of each window. After placing the windows one stretches the wall back to it's original length. .........

After "getting-my-head-around" Matthew's method (where is a nice picture when you need one ), I now do not see a need for the Divide Hotspots.

Because I was using a single Arc as the wall, (not read practical since there will be most likely other walls attached to this wall to enclose the space) I could not place the first window, which is to be removed later, just touching on one end.

After placing a Hotspot on the Original end location of the wall, I just stretched the wall at least as far as the window dimension or a new wall can be used with the same Radius (no Calculation neccesary). Placed the side of the temporary window at hot spot and sent it into the temporary stretched portion of wall. From there I use the "Matthew Method". Delete temporary window, stretch (squeeze) back the wall to original size, or if a new wall was placed just delete the wall and the temporarily placed window is deleted also.

Not that difficult and no Calculations. Thank you Matthew.

Am I making this more difficult by adding the temporary wall? I could not place the temporary window otherwise.

Re: The great divide

Anonymous — Sat, 27 May 2006 16:21:35 GMT

Jay,
I think you have answered your own question
I did want to ask you about something in your
post of May 26.
You show a formula in your "pretty picture"
that reads arc=(05*180/PI*15)
This looks similar to the arc length formula
but where does the 05 term come from ?
Thanks,
Peter Devlin

Re: The great divide

Anonymous — Sat, 27 May 2006 17:56:17 GMT

Because this is a low pixel count JPG, the decimal point was "non-existent". The number was 0.5 (6" = 0.5 feet) for the arc distance. But this mental routine need not be performed. The formula is helpful though for finding the required angular distance for a given arc and Radius, as you know.

Re: The great divide

Anonymous — Sat, 27 May 2006 18:37:28 GMT

Thanks Jay,
I really was staring at that formula and wondering
what I was missing.
I thought it might be 0.5 but I could not see any evidence
of a decimal point.
Peter Devlin