Parametric models in the Design Catalog?

[Hirudin]

The short version of my question: will the Design Catalog have parametric customization? (Something similar to the the McMaster Carr site or the various “parts” sites out there, like GrabCAD.)

Long version…
I’ve been playing with my laser cutter and one of the projects I’ve been working on is an organization solution for the board game “Small World”. For any board game players out there that have not tried the game yet, I highly recommend it!

I imagine that many people who have played it, especially if they also have some expansion packs, have noticed that the factory box was apparently designed by a jackass. The molded plastic tray in the box has slots that the pieces sorta fit into, but very few of those slots are sized appropriately. They’re either too big, too small, too shallow, or something else. The core game fits OK, but once you start adding expansions you start running out of places to put all the pieces. Some of the expansions come with additional trays, but they are far from perfect as well. I set out to make something better.

Here’s a pic of a prototype… (I’m only posting this somewhat-low resolution pic because I have a better design in the works)

I cut this one out of baltic birch, but I’ve also experimented with using acrylic (I threw a few boxes into the pic). Of course, these two materials may be different thicknesses (in this case, the acrylic is 2.8mm and the baltic birch is 3.1mm). The little boxes have tabs to help hold them together and align them and those tabs introduce a bit of a problem when designing something intended to be cut from different thicknesses of material: in order for the edge of the tab to be flush with the side of the mated piece, the tab must have a length equal to the thickness of the material.

A lil’ image to show what too short, correct, and too long tabs look like.

If I were to buy a design like this in the Design Catalog, I would prefer that the tabs be the “correct” length for whatever material that I’ve chosen to cut it from. Are there plans to incorporate some parametric customization options into the Catalog?

If parametric customization is in the works, I’d be interested in hearing some details about it. A couple questions that come to mind are… (1) When will customization be available to customers? (2) Can a chosen parameter influence the price? (3) Can things other than dimensions be specified?

1. When?
   Basically, I mean: pre- or post-sale? I can come up with reasons why only pre-sale customization and why only post-sale customization would be preferable. Letting the designers decide would prolly be best, but I could see that causing confusion. I’ve tried to write out my rationales for each case, but they ended up being pretty convoluted. I can attempt to expand on this question if anyone wants…

2. Parameter = Price?
   This is easy enough. If someone designs a box or something, maybe the price for the design could be based on the total volume of the box, or some other parameter. A jewelry box would cost less than a shoe box.

3. Other stuff?
   Say someone makes a… I don’t know… little rack for test-tubes. It would be nice if customers could pick how many test-tubes the rack holds. This, combined with the answers to 1 and 2 have big implications, IMO. Another thing that designers might want to customize using the store/software UI is text. Adding some simple text is easy enough, but it could be made easier within the store interface. For instance, maybe someone might want a word printed on all six sides of a box, it would be cool if there was a provision, provided by Glowforge, to place that text for the customer. Even if six instances of text sounds like something that would be “easy” to do by the customer, imagine the label being printed on all four edges of all six sides of the box.

Sorry for the long post!

[chadmart1076]

First off, kudos on the box insert for Small World. Indeed it is a great game but with many bits that don’t work particularly well in the box insert provided with the game. Box inserts is something I plan to use my GF for heavily.

I must first admit that I have very little experience with parametric design. It would be very cool if anything that fit together (like a box) could be adjusted based on the thickness of the material. I think the more likely solution will be to have multiple designs in the catalog based on thickness, then have varying settings for those designs based on material.

[joe]

First, Awesome game! I’m planning something similar for mine. But I’m probably going to start with a custom Dominion box first; its in much worse shape than my Small World box.

A lot of this is going to depend on what you get when you buy a design from the design catalog. Do you get a 2d illustrator file? Do you get a 3d cad file? Or do you just get access in the cloud and you hit the glowing button to print?

I would prefer to get a file when I buy something from the catalog, that way I can fix any error due to variable thickness.
However, if you just get access in the cloud, I would say they need to have an option to input thickness for joints like this. Or have it auto adjust the joints when using Proofgrade^tm?

Hmmmm… this Design Catalog thingy is more complicated than I thought.

[jacobturner]

Long ago when talking about Glowforge capabilities, this idea of tabbed boxes and material thickness variation had come up.

I seem to recall that you would be able to set certain parameters such as “material thickness” for portions of a design. But thinking back on that memory it makes no sense to me that it was mentioned for Glowforge specifically anywhere except for when posting to the catalog. (Since designing your intended cut all happens outside of Glowforge cloud control, except when cutting a doodle)

[nick07lee]

They have something Kinda, sorta, almost, not really, like this on shapeways.com

On that site you can purchase "variations"
For instance if i was to buy a ring from them, i could select ring size and width.

I will say its a major pain in the butt. For their website i have to load the correct STL for each variation that i generate. This turns into quite the pile of files i have to generate.

I know you can drive parametric models from websites (www.misumi.com is a great example, and mcmaster for that matter) But i would be willing to be it is exceptionally complicated. I have seen several websites try and fail miserably and it leads to a terrible experience.

As much as it pains me, i would prefer the “variations” style of change. My upfront work leads to a much smoother and easier shopping experience and also cuts back on user knowledge. I have no problem telling my grandma to shop there because it is just click and go.

[sqw]

as an aside, I would think EVA foam would be great for this application

[caribis2]

Everything I’ve done for myself in OpenSCAD has been parametric, including tabs. Presumably anything in the catalog would be up to the designer as to whether or not it’s parametric. Some things scale better than others. Personally, as a customer I want parametric designs, but as a seller I would not. If your design is not “parametric-proof” you can wind up with a support headache that isn’t worth the money.

Based on the number of gamers who actively post (and extrapolating the lurkers), and this apparently being a common problem, it would not surprise me if an informal economy in these designs developed.

One potential issue with the tabs is that to make them perfect the material has to be perfect.This is more likely in acrylic and other manufactured materials and less likely the more natural the material. So while my tabs are set to the material thickness, which is an input, I still expect to be doing some sanding if it’s a presentation box as opposed to a workshop box.

[jacobturner]

@nick07lee

I know you can drive parametric models from websites (www.misumi.com is a great example, and mcmaster for that matter) But i would be willing to be it is exceptionally complicated. I have seen several websites try and fail miserably and it leads to a terrible experience.

Thingiverse is the worst. Because any time you customize something with their parametric setup, it by default publishes your insanely minor adjustment.

So many people do not turn that off, flooding any search result with myriad copies of the same damn thing.

[erin]

Based on nothing, I expect the designs to be listed with photographs of the finished products, rather than digital renderings, and a list of supplies or materials. And probably links to the appropriate proofgrade supplies. So, for example, a box might be listed as using 1/4 inch plywood. And the accompanying file would be appropriate for 1/4 inch plywood. Presumably, there would be nothing stopping the end user from tweaking the design to use, say, 1/8 inch acrylic, unless there’s no downloading a file at all and you “print” directly from catalogue to glowforge. Which I suppose is a definite possibility. But I would not expect material variations to be a built in feature of the catalogue. It’s just much easier for a beginner user experience to see project x made with supply y, and to use supply y and get an identical replica to project x.

But that’s all just guessing. I doubt glowforge is far enough in the catalogue process to have really tackled this yet. I bet the decision won’t actually be finalized until there’s a beta version of the catalogue with not-glowforge designers uploading projects.

[erin]

@Hirudin

rameter = Price?This is easy enough. If someone designs a box or something, maybe the price for the design could be based on the total volume of the box, or some other parameter. A jewelry box would cost less than a shoe box.

I don’t understand the reasoning behind why size would matter with regards to the price of a design. Either glowforge will set prices (most likely), or the designer will. If glowforge is setting the price, I would expect maybe 2 or 3 tiers of pricing based on, perhaps, the intricacy of the file - like, a working Ferris wheel might cost more than a picture frame. But I wouldn’t say a 9 inch box is anymore intricate than a 5 inch box. And for sure, an attractive small box can be much harder to design than a utilitarian crate.

@Hirudin

Say someone makes a… I don’t know… little rack for test-tubes. It would be nice if customers could pick how many test-tubes the rack holds. This, combined with the answers to 1 and 2 have big implications, IMO. Another thing that designers might want to customize using the store/software UI is text. Adding some simple text is easy enough, but it could be made easier within the store interface. For instance, maybe someone might want a word printed on all six sides of a box, it would be cool if there was a provision, provided by Glowforge, to place that text for the customer. Even if six instances of text sounds like something that would be “easy” to do by the customer, imagine the label being printed on all four edges of all six sides of the box.

With regards to the test tubes, I would not expect that much customization from one design. At all. It’s not really great for the designer and it’s not great for sales. From my own experience, the vast, vast majority of people can’t see past the first thumbnail. You can explain possibilities and show multiple uses and all that, but they’re sold (or not) by the thumbnail. They don’t consider if a design might work with x changes and they don’t think through possibilities. They make a snap judgement based on the very first thing they see, and then they’ve moved on. I mean, most people can’t even see past color. I know that sounds nuts - they’re just buying a design, so they should realize they can make it in any color… But unfortunately it doesn’t work that way in practice.

[jacobturner]

Okay, just remembered more about the long ago conversation…

You are supposed to (eventually?) be able to specify certain parameters/measurements to be based on material thickness, and then the CAMERA registers the actual thickness of your material, and the cloud adjusts the design to be a spot on fit for the material actually loaded.

@dan should be beckoned to call me out on false memory at this point. Maybe it was just us crazy forum folk postulating what can be done with the camera that would blow away current laser users.

EDIT: The search is strong with me today:

Enhancements: interlocking parts detection

There’s a few moving pieces. (sorry for the bad pun) You can, optionally, mark up your designs to indicate that a certain dimension should be resized to equal the material thickness. “Make this slot as wide as the material is thick, plus 0.002” for clearance". That sort of thing. Plans we sell will be marked up like this. If you do this, as you say, you can swap 1/4 for 1/8 and it will still work. When designing, you can designate “product” and “scrap” separately, so the laser’s kerf la…

[paulw]

Do we even know how designs are going to be sold? Obviously you can download an svg or a dxf or whatever, but since the glowforge is cloud-based and semi-closed, it could make sense that some designs will be one-shot purchases where only the driver waveforms reach the buyer’s machine. I’m thinking that this would open the door to designs that would be really cheap, but also to designs that could be uniquely parameterized, especially with proof-grade material.

[erin]

No, I don’t think they have much locked down about the catalogue. Certainly not much they’ve released. I mean, makes sense. The catalogue, as cool as it is, is totally secondary to actually getting people working lasers. The loveliest catalogue won’t be of much use if no one can use it except beta testers

[mpipes]

Simple things like adjusting a tab size based upon material thickness should be easy as pie at the web interface.

Parametric models built with AutoDesk Inventor or Pro/Engineer can be designed using Family Tables (Pro/E) or iParts (Inventor). These models would have user-adjustable parameters, selected by the designer (so a user can’t “break” the intent of the model). I’m sure Solidworks has a similar feature but Im not familiar enough to say.

This works because both Inventor and Pro/E can use XML data to drive part features, and a web interface can input that XML data, and spit out a model.

[nick07lee]

Solidworks has a similar thing. you can tabulate a collection of dimensions or equations that you want to be driven. this doesnt lock out the other dimensions, so you can still break it. But it lets you drive several dimensions and variations of the model with excel.

[Hirudin]

@erin

I don’t understand the reasoning behind why size would matter with regards to the price of a design. Either glowforge will set prices (most likely), or the designer will. If glowforge is setting the price, I would expect maybe 2 or 3 tiers of pricing based on, perhaps, the intricacy of the file - like, a working Ferris wheel might cost more than a picture frame. But I wouldn’t say a 9 inch box is anymore intricate than a 5 inch box. And for sure, an attractive small box can be much harder to design than a utilitarian crate.

I don’t think it’s strictly morally justifiable, but part of the set price for any item is going to be how desirable it is. Sometimes a larger box will be more desirable than a smaller one and thus the price for the design could be higher. If I were to design a refrigerator accessory that holds eggs, and if I were able to set the price on a per-egg basis, I would probably drop the price for the 13-egg model through the 17-egg model and 6, 12, and 18 would probably have the highest price:capacity ratio.

@jacobturner

EDIT: The search is strong with me today:

Nice! Thanks for finding (and remembering) that! Glowforge is 2 steps ahead on this one.

[Hirudin]

@nick07lee

I will say its a major pain in the butt. For their website i have to load the correct STL for each variation that i generate.

That’s what I’m afraid of. I agree that the additional upfront work of the designer can really streamline the experience for the customer though, and that is definitely what Glowforge is going for.

[erin]

@Hirudin

Part of the set price for any item is going to be how desirable it is. Sometimes a larger box will be more desirable than a smaller one and thus the price for the design could be higher.

It’s all conjecture at this point, obviously, but I don’t think glowforge will set prices based on desirability, i.e. making more popular designs more expensive to get more money. It has nothing to do with being moral or ethical – it’s just not really a good move. And I don’t think the individual designers will have control over pricing. Except maybe indirectly through different licensing options.

But, that’s just my guess. They could be planning something entirely crazy for all we know. Like, a system where you can purchase credits to buy designs, or steal them from other glowforgers by beating them in a contest. Hopefully not something requiring balance or catching. Or chess. I would lose all the credits.

[dan]

No details to announce but making catalog prints work well across some variability (customization, material thickness) is definitely on the todo list.

[caribis2]

@erin

From my own experience, the vast, vast majority of people can’t see past the first thumbnail. You can explain possibilities and show multiple uses and all that, but they’re sold (or not) by the thumbnail. They don’t consider if a design might work with x changes and they don’t think through possibilities. They make a snap judgement based on the very first thing they see, and then they’ve moved on. I mean, most people can’t even see past color.

This describes every marketing department I’ve ever worked with - especially the color part.