Hello there,
Just sharing a video that was suggested to me on YouTube. I found it interesting that this is the first time Rhino has appeared in the tournament, but I suppose that’s to be expected with such heavy mechanical designs.
First apearance of Rhino on CAD Speed Modelling tournament
What are you thoughts on this ?
I noticed the guy using Solidworks did not center the through bolt hole towards the end. Which would have been only seconds. Nice work by both.—-Mark
Not the first time: CAD Speed Modelling Challenge 
I’ve been doing some of those drawings as well for fun. I uploaded a (slow) recording of some attempt back in May 2024:
I think these are nice challenges in two regards: find the optimum workflow for a specific piece, and as a developer find issues with Rhino.
My thoughts are that I should have put more practice into sheet metal! Heh.
The tournament is a little different than the previous challenge. The one you linked, you could have as many attempts as you want to get the fastest part, optimizing every step.
The tournament had a different feel. You can’t plan on trying different approaches, you kind of have to pick a good path quickly and start moving.
Both were fun, and I’ll definitely be trying again!
Gave it a go, 2.30 with a few misclicks & filleting the wrong edges 
. Do you need a GH script for the mass or could you do it using the volume from a solid?
The GH script was the easiest way I could think of at the time. I’ve now replaced that with a Python script, and file templates for each material & unit system that have a Density built in as DocumentUserText.
Python script takes the part volume, multiplies it by the density, and puts a text dot next to the part.
I have a couple of additional tweaks that I want to make for speed purposes…
Nice, what’s your alias set up like? One thing I’ve always found really useful is mapping Sellast to the middle mouse button.
I find this modelling competition rather useless if not even stupid. From my experience very seldom it is about modelling just one exact object and job done. I’m usually in the situation where something is modelled and then it has to be changed. If the object is modelled more or less the exact same way in Rhino and a solid modeler, the solid modeler is likely to adapt the object in much less time. However, if the solid modeler is asked to create a multitude of shapes with slight variations I think Grasshopper would win…
A bit all over the place, a mish-mash of shortened commands and Ctrl/Shift/Alt usage.
I’ve seen some wild alias maps though, that keep everything left-hand-only for the keyboard. It seemed like too much memorization to me though, having to remember all that. I just type fast.
Edit: It was early and I hadn’t had my coffee yet - I do have some commands mapped to the multifunction button on my mouse. If I press it and move one of the four directions, it does four different things. Down turns Project on and off, right is SelLast, left is ZoomSelected, and up is ZoomExtents.
My “forward” button is mapped to Distance, and my “back” button is mapped to Shift, so that when I hold it and right click, I pan instead of rotate the view.
Sure. I already gave you my 2 cents on how it can be a great way of finding bugs. Even if you don’t try to do it as fast as possible.
If I were to prepare for a specific set of operations and wanted to make it as fast as possible I’d probably program my keyboard to have different layers with anything I want preprogrammed in useful ways. Having a keyboard with fully customizable firmware is great (mine run on QMK).
Addendum:
I also think it is useful to think about workflow. Even if the one case doesn’t translate into a more general setup, doing more multiple can give you insight into creating models and perhaps give some ideas. Especially if you look at how other people use the same software. And why not even how people do the same in other software. There is always something to be learned.
Eh, different strokes for different folks. I kind of enjoy the “gamification” of CAD as a fun little side thing, and it’s nice to nerd out with other folks. Plus there’s a bit of satisfaction from being the odd-man-out in a field that’s largely (or entirely, in this case) SolidWorks.
Technically speaking, I have the same experience as you - for what I’m doing on a daily basis, very seldom am I modeling one perfectly-known thing as fast as possible. What takes the longest is the “figuring out” period, trying to figure out how to mash together client demands, codes/laws, budget constraints, etc; the actual modeling is probably only 30% of that? The tournament doesn’t show the hours of notebook sketches, team meetings, research, etc, but it’s good fun.
When I know that a project needs to be parametric, or that I’m going to be making changes, that’s when I’ll break out Grasshopper. (I’m actually working on some custom millwork GH scripts in my spare time)
Yeah I use a mix of left hand e1, e2, e3 etc, and Autocad-style two letter abbreviations, added over time.
That’s literally what I use rhino for, bespoke one-off objects (architectural models). Experiences vary, and anyway, it’s fun.
I have this post bookmarked, and thought about implementing it one day. It just seems so…involved, heh.
Indeed it is very involving. That is why I haven’t done it much. I really have currently only switching between Rendered and Raytraced mode, TestMooCow, SystemInfo and Audit3dmFile programmed as shortkeys on my keyboard
I was also thinking this was one sided towards solidworks
The drawing is a perfect example why sw is good at doing those type of models while Rhino even tho it has similar tools it’s not it’s strongest side
Cads always seem to have at least something better than the other at certain tasks.
always nice seeing people with similar interests doing something fun tho
I remember there is something called “configurations” in solidworks which does exactly what you are describing:
I’d rather like to see a real high-end product modelled, with draft angles where applicable, and proper G2 continuity - triathlon helmet, competition horse saddle, lacrosse stick, ski-boot, carbon bike frame, rear car bumper, ergonomic garden shears, tuneable golf driver. Contenders get one planar and two perspective images. Standard out-of-the-box Rhino. 3… 2… 1… click!
Haha that’d be an interesting challenge! Especially two hours in when contender 1 shelves the last hour of work to try a different surface layout.
Fluid forms/transitions etc also present other challenges, like which is closest or best represents the reference images. The 2.5d SW speed challenge confirm things with mass… so pretty easy.
Just throwing the holes wherever is the ‘Ivan Exploit’ but is permitted in the rules (which TTT usually mentions beforehand). Basically it’s all about getting to the correct mass the fastest, not much else matters.
If the python scripts seem like too much for mass calcs, this food4rhino package makes it easy and can be customized.
P.S. There is a .rui included that they don’t mention on the food4rhino page.
I totally agree to the first part. But do not underestimate a parametric CAD system. Once you modelled it flexible enough, you can change variations by simple reference or parameter changes. This works much better than you might think. Grasshopper is most of the time the slowest possible tool you can choose for such tasks because of incomplete surfacing tooling. I find this rather entirely stupid, because its not the bottleneck. Similar to speed coding, you can only improve “performance”, if you view on the entire process. Nowadays people wasting time on many other things, like posting in this forum during work 