## London Gerkin

The place to talk about Stella4D, Great Stella, and Small Stella. Feel free to ask questions about them here.
robertw
Site Admin
Posts: 535
Joined: Thu Jan 10, 2008 6:47 am
Location: Melbourne, Australia
Contact:
First enter faceting mode, which is entered with this toolbar button:

As soon as you enter this mode (as with all modes) it will show you a description on-screen of what special operations are available in that mode:

If you forget what the mouse can do, keep your eye on the bottom right corner of the window and see how it changes when you hold down Shift or Ctrl (or both). When in faceting mode and holding down Shift, it looks like this:

That tells you what the mouse can do.

OK, so let's make a truncated pyramid, also known as a podium. Well, actually Ulrich might have had it slightly wrong. He was making an anti-podium, so let's do that (lateral faces are triangles rather than quads).

First, the resulting solid will not have dihedral symmetry, so from the drop-down list that initially says "8-fold Dihedral" select "8-fold Pyramidal".

Let's start with the big base octagon. Just hold down Shift and Left-click on each vertex in turn around the base (actually after doing three you can hit the "..." button to auto-finish the polygon).

Then hit Enter to accept this polygon as complete.

Do same again for the top octagon, using vertices closer to the centre.

Now create the lateral triangles. Here's one being created, but not accepted yet:

And here it is after accepting (it's repeated over the symmetry group, so 8 triangles are created).

There are still green edges. These represent edges that are not complete yet, that is, they only have one face so far (a face still needs to connect to another face at these edges). So create the final type of lateral triangle face to get this:

No more green edges, so all seems well. Create this faceted polyhedron by clicking on the left-and-down arrow button in the titlebar of the faceting preview window

If you really did want a podium rather than an anti-podium, then you would start with an octagonal prism instead of antiprism, and create lateral quads.

Alternative method

As Ulrich mentioned, there's also an easier way to create these, but you don't currently have precise control over sizes (must judge by eye instead). This is to use augmentation mode, which can create anti-podiums directly. Start with an antiprism as before, but don't subdivide the faces. Again, drop to 8-fold pyramidal symmetry. Select the top face (you'll want to leave faceting mode first) and hit "a" for augment (or use "Poly->Augment Polyhedron"). Lots of options come up, but you can ignore most of them. Select Antiprism from the choices at the top and hit enter. You should see something like this:

Hold down Ctrl and notice the mouse tips in the bottom right corner of the window:

So while holding down Ctrl, hold down the right mouse button and drag the mouse left and right. You will see the width of the top of the antiprism get bigger or smaller. Also try the left mouse button and drag to change the height. Hit Enter when you're happy and with to keep the result.

You can then select the new top face and repeat the process to add more layers. Should be quite fast and easy, and the augmentation options box will remember your settings from last time, so you don't even need to keep selecting the Antiprism option.

Hope that helps.
Rob.

Squeaky
Posts: 47
Joined: Wed Jan 16, 2008 3:09 pm
Location: Derbyshire, England

Thanks for your advice .... will read youe advice very carefully

I have been messing about alll afternoon and have built (and saved) the above.

Now, after reading your posting, I will have a go at augmentation

Squeaky
Posts: 47
Joined: Wed Jan 16, 2008 3:09 pm
Location: Derbyshire, England

### Thanks.

Thanks!

With a little more work on this method I am quite sure I will be able to achieve what I need.

I need to do a bit more on this shape ... then work out how to get all my mitre angles and side lengths

Squeaky
Posts: 47
Joined: Wed Jan 16, 2008 3:09 pm
Location: Derbyshire, England

### One or two observations

As you will see, from the above, I have made quite a bit of progress today using your 'alternative' method of construction.

Two points ......

1.... I am not sure how to augment to a specific face .... I think I am missing some important selection process.... before I press the 'a' key

2 .... After about six augmentations the software seems overloaded and some quite odd things start to happen happen.

Can this be ovecome or should I start a new drawing for the next six or seven floors.

Anyway, I must thank you for your great assistance.

If anybody else wishes to construct a similar model, from wood, I would be only too pleased to share with them my methods of working.

Squeaky

robertw
Site Admin
Posts: 535
Joined: Thu Jan 10, 2008 6:47 am
Location: Melbourne, Australia
Contact:

### Re: One or two observations

Squeaky wrote:1.... I am not sure how to augment to a specific face .... I think I am missing some important selection process.... before I press the 'a' key
Just select a face first using Shift+Left-click, or double-click on a face. Make sure you're in the default mode first, not faceting mode or anything (hit Esc to return to normal mode).
2 .... After about six augmentations the software seems overloaded and some quite odd things start to happen happen.
Currently creation of nets is usually done via the stellation process, which becomes expensive for models with lots of faces. So make sure no net or stellation views are open while building the model. Then Stella should handle it easily.

There's a trick at the end to allow Stella to create the nets more easily once you've finished too. Use "Poly->Create Convex Hull". Provided your model is convex, nothing should appear to change, but now Stella knows the model is convex, and can create nets without going via stellation (and thus much faster).

Actually, if it's more than just going slow (and potentially slowing down the whole computer), maybe you can expand on what you mean by "odd things".

Rob.

Squeaky
Posts: 47
Joined: Wed Jan 16, 2008 3:09 pm
Location: Derbyshire, England

### Convex Hull

I have attempted to build as you suggested (using 'convex hull) but I am still experiencing some problems.

After seven augmentations the programme/computer faults with a message like .... 'Shape not accurate enough'.

I can overcome this practically by proceeding just one layer at a time... I only need the mitre angles and side lengths .... but it would be nice to see the whole gherkin shape as one model.

I am posting two shots of the shape that I am currently working on.

Squeaky

robertw
Site Admin
Posts: 535
Joined: Thu Jan 10, 2008 6:47 am
Location: Melbourne, Australia
Contact:
Hmm, I just tried with more than 7 layers and it didn't complain. Sometimes accumulation of errors can be a problem, but mostly only when stellation is involved (which is often the case when creating nets). If you get a message saying "Warning: this model may be too complex to stellate" then you're already doing something wrong, as it should be possible to avoid stellation altogether.

You really need to find the precise measurements of that building. Surely they're available somewhere. This method isn't well-suited to recreating it though, since the sizing must be done by eye.

Rob.

Squeaky
Posts: 47
Joined: Wed Jan 16, 2008 3:09 pm
Location: Derbyshire, England

### Problem Augmenting

Hi Rob,

I am happy with the shape, so far, but cannot seem to be able to augment any further.

If I sent you a copy of the file could you possibly show me how to add a further three layers to each end of the current shape.

The three centre layers (coloured red, yellow and blue) are straight sided
the green mauve bright pink and pale yellow are angled inwards slightly and I would wish the three layers, added to each end of the current shape, to be angled, by eye, a similar ammount.

Squeaky

robertw
Site Admin
Posts: 535
Joined: Thu Jan 10, 2008 6:47 am
Location: Melbourne, Australia
Contact:
Sure, send the .stel file to me and I'll see what I can do.

Rob.

Squeaky
Posts: 47
Joined: Wed Jan 16, 2008 3:09 pm
Location: Derbyshire, England

### London Gherkin

Thanks for your very kind help Rob

I will be using this shape as the base for my model. From the 'nets' I can extract all the triangle details (length of sides and mitre angles).

Without your programme I very, very much doubt that I could even attempt to construct this rather complex model.

The top layers including the 'lens' will be done later as a separate file.

Would it be of any interest to post, from time to time, photographs of stages showing its construction?

Squeaky

robertw
Site Admin
Posts: 535
Joined: Thu Jan 10, 2008 6:47 am
Location: Melbourne, Australia
Contact:

### Re: London Gherkin

Squeaky wrote:Would it be of any interest to post, from time to time, photographs of stages showing its construction?
Sure. You should probably do that in the Polyhedron Models forum.

Rob.

Squeaky
Posts: 47
Joined: Wed Jan 16, 2008 3:09 pm
Location: Derbyshire, England

### Putting an Agmentation into memory

Thinking ahead ...

Can I put the augmentation that produced say layer 16 (from the London Gherkin) into memory?

Then, using this layer as my staring point, build up the remaining layers to produce the top part of the model.

Or...... Will I need to start from scratch with an antiprism with the sizes found from the nets of the base part of my model? Then ignore the base layer, when actually building the model .... because it will be straight sided?

Squeaky

robertw
Site Admin
Posts: 535
Joined: Thu Jan 10, 2008 6:47 am
Location: Melbourne, Australia
Contact:

### Re: Putting an Agmentation into memory

Squeaky wrote:Can I put the augmentation that produced say layer 16 (from the London Gherkin) into memory?
You could use faceting mode to create the faces you need for just that part.

Rob.

dabeard
Posts: 12
Joined: Sat Jun 13, 2015 7:08 am

### The Gherkin revisited

Hi Rob and others

I've decided that for my next modular origami project I'd like to attempt to construct a model of the London "gherkin" and was delighted to find this model had already been discussed on this forum back in 2008.

Like Squeaky, my main use of Great Stella will be to create an approximation of the Gherkin, in turn giving me the dimensional data I need for modelling.

Rather than start from scratch in Stella, I wondered if you still had the .stel file you worked on with Squeaky back in 2008? I messaged Squeaky with a similar request a month or so ago, but haven't had a response and it appears that he is no longer an active member of this forum.

Given that it was 7 years ago, finding one .stel file amongst many is probably a big ask, but if you can help that would be great. If not the previous posts on the topic appear to give me an excellent starting point.

Kind regards

David

robertw
Site Admin
Posts: 535
Joined: Thu Jan 10, 2008 6:47 am
Location: Melbourne, Australia
Contact:
OK dabeard, didn't take me long to find again. I just emailed it to you.

Rob.