How do you make your digital models?

[mdemerchant]

I'm curious what other people's processes are for building their digital models.  Do you build all digital first, build in bricks first and capture later, work in parallel or something else?

I like using the digital tools but I like building with real parts a lot more so usually I tend to just build and making a digital model comes at the end only if I have made something I'd like to be able to build again someday, which isn't that often really.  But if I'm going to build something fairly complicated I usually make a rough digital model first to come up with a basic layout of mechanical parts, motors, battery box, etc. because it's way faster than building and rebuilding (and rebuilding...) with real parts.  Once I have the basics worked out I build it for real, working out the remaining details as I go.  However I find that it's then quite a lot of work to go back and bring the digital model up to date.  I was recently doing this with my mobile crane which is a big model and basically ended up practically rebuilding the digital model from scratch as I disassembled the real thing, otherwise it was hard to be sure if I had old parts left in it that didn't exist in the final model.

Does anyone have a more clever approach?

I guess this question could be more broadly applicable but I am specifically thinking about technic models, which is why I posted the question here.

 

[TechnicRCRacer]

28 minutes ago, mdemerchant said:

I like using the digital tools but I like building with real parts a lot more so usually I tend to just build and making a digital model comes at the end only if I have made something I'd like to be able to build again someday,

This is the method I follow most of the time, build with parts first (the color doesn’t matter) and do a digital model if it’s worth showing. I did do a digital model first once with a non-Technic build and it turned out fantastic, so there is a great advantage if you have a particularly bad sorting system for parts (like my normal brick collection). 

[bonox]

if i've got parts, i'll build combination of real and virtual. If I have new ideas and don't have the parts, i'll use the digital model to work out what to buy, but I don't always get it right first time in the virtual. Some of the tools in stud.io are helpful in this regard, for example finding things that have limited support. (attached by one stud for example is a useful tool). When building for real and finding things that need to be changed, remodelling the digital model is called 'red pen markups' in the old terminology and can be a bug bear, but needs to be done if you want to follow later or share.

 

Edited September 22, 2021 by bonox

[Seasider]

It depends on what I’m building. I often physically build small sub-models to test mechanisms. For TC20 a lot of it was done digitally, then going to physical to check it all hangs together properly. For some of the past competitions I’ve done pure physical build and then built digitally at the end so I have a record of what I made. But most of the builds I do now are definitely digital based. I find it allows me to position mechanisms or parts where I want them in space and then work out how to connect them after.

[metalgeekzy]

I am very critical of my own work. Most things i hate, i always see room for improvement. So to me most of my creations are not worth saving. But once in awhile i make something that i really like. And if i can see that creation being used time and time again, ill save it. Take pics, make instructions. 

Like right now im working on a crane. Technic built. I will probably save it once i figure out the arms. The base and what not im mostly happy with. 

[HectorMB]

5 hours ago, mdemerchant said:

I'm curious what other people's processes are for building their digital models.  Do you build all digital first, build in bricks first and capture later, work in parallel or something else?

I like using the digital tools but I like building with real parts a lot more so usually I tend to just build and making a digital model comes at the end only if I have made something I'd like to be able to build again someday, which isn't that often really.  But if I'm going to build something fairly complicated I usually make a rough digital model first to come up with a basic layout of mechanical parts, motors, battery box, etc. because it's way faster than building and rebuilding (and rebuilding...) with real parts.  Once I have the basics worked out I build it for real, working out the remaining details as I go.  However I find that it's then quite a lot of work to go back and bring the digital model up to date.  I was recently doing this with my mobile crane which is a big model and basically ended up practically rebuilding the digital model from scratch as I disassembled the real thing, otherwise it was hard to be sure if I had old parts left in it that didn't exist in the final model.

Does anyone have a more clever approach?

I guess this question could be more broadly applicable but I am specifically thinking about technic models, which is why I posted the question here.

 

Indeed, I proceed very similar to you. I initially make a scratch in digital, a scafold. Then, i build it phiscally and solve the issues that I see. Then, I digitalize everything (often from scratch). At this point, I often have just a "refined" scratch which basically have all the functiontions and only the appearance is missing. From there, I normally finish the model only digitally and once I have something that I am happy with, I make the instructions and the final physical model.

[Jerry LEGO Creations]

I prefer building with real parts. My approach therefore is I start with the physical model, and build the digital one in parallel, though usually a bit behind the physical model. I do this to make sure I am happy with whatever I made to prevent having to rebuild it digitally later on (though that of course still happens from time to time). Having said that, if you don't have the correct parts, experimenting digitally first can work pretty well

[HectorMB]

For me, the reason of using digital tools is not the parts availability, but the possibility to place pieces in positions and sites that would be impossible without dismantling the physical model. 

But also, the use of parts that I do not have is a plus ;)

[astyanax]

I think we had a rather similar discussion less than 2 months ago?

 

[1963maniac]

I've been there and done what all of you have said and for the same reasons at one time or another. I just posted my roller coaster Mod to Rebrickable yesterday. I worked on that project for almost two years. I kept thinking of better building techniques or found other people's techniques and adopted or adapted them to my project. I needed to document all the things I was doing as I went along. So, I do both together some of the time, some of the time I try things digitally first. I then order parts which takes at least three weeks to get here from Europe. In the mean time I find another technique and then had to order more parts to try that new building technique. On and on it went until everything came together as it needed to be. Let me tell you, rearranging that big roller coaster is a lot of work, especially when there are two 10261 sets. But it's done and it works well.

Edited October 20, 2021 by 1963maniac

[SaperPL]

I'm prototyping physically mechanical principles of the model first, then I try to build it in studio, iterate and make it look good. Then I build a physical prototype based on the design and so on. A tick-tock process, kind of like intel. Sometimes I go all physical though if mechanics are most of the build.

[OnyyX]

When designing my 1/16 cars, I used to always design them 100% with bricks in hand, which was a bad habit, as I would make sloppy design choices based on what I had on hand. Nowadays, with a bit more experience, I design my cars from the beginning in Stud.io, not limited by bricks at all. I then create a parts list, instructions and then sort my parts I have on hand, crossing off parts I have and parts I don't. Then I gather parts I don't have in the needed colour in random colours, and then build the car. I then assess any design issues, making anything more rigid etc. I then create a wanted list on Bricklink of parts that I need, and once they arrive, the car is done! I then add any changes I made to the model on the Stud.io one, and make the instructions for it and put them for sale on Rebrickable.

[mdemerchant]

Thanks for all the input, especially to @astyanax for linking the other thread.  I don't use Stud.io as my tool of choice so I wouldn't have noticed it if you hadn't pointed it out.

It's very interesting to hear everyone's processes and I'm glad to see I haven't missed some obvious way to make things easier.  I think my takeaway from the discussion and my own experience is that the easiest way to get an accurate digital model is probably to build the entire thing digitally first, with test builds as needed.  It may not be the most enjoyable way for me personally but I can see that it would be the least overall effort.

 

[MinusAndy]

How do I build my digital models?

Badly. Is how. 
 

I tend to get stuck for hours on aligning gears and axles and it really puts me off. I’m pretty sure there are simple shortcuts to achieve what I want to do but I haven’t found them yet.

For me, the things I’d want to build virtually are the things I would find hardest, like connections at strange angles and flex axles. As much as building mechanisms and gear trains seems great I often find that the digital solution doesn’t take backlash and power soak into account. 

[Pinnacle]

I like it to draw my MOC first as a Digital model. But for me,  it is almost a necessity too.

Due to the dimensions, I want to design my buildings first. So I know the numbers right. Sizes,  parts, and so on.

I don’t like unwanted surprises during building…