Center of rotation, mirror plane

I am exploring unmake and have been frustrated in attempting to revolve or mirror objects. Is the center of the revolved object modifiable or assignable? The same question goes for mirror planes? Perhaps once an object and say the revolve command are selected could there be a prompt to select the center of rotation or mirror plane. This would allow for off Cartesian axis mirror planes etc.
I am exploring unmake and have been frustrated in attempting to revolve or mirror objects. Is the center of the revolved object modifiable or assignable? The same question goes for mirror planes? Perhaps once an object and say the revolve command are selected could there be a prompt to select the center of rotation or mirror plane. This would allow for off Cartesian axis mirror planes etc.
1
5 comments

Hi Michael, 


Great questions and thanks for asking! 


So regarding Mirror: Currently mirror and symmetry are locked to the Cartesian Axis, and uMake does not have the ability to move the axis of reflection. I encountered a need for this the other day when I was working on a model of a person sitting in a vehicle-- I wanted to model the person in the drivers seat, and wished I could move the symmetry axis, so I understand the need for this. I will add this to our features request list as I do not believe it is currently on it. I definitely can see a use for this feature. 


With regards to revolve: In the next version of 109, we will be adding snapping to the axis of revolve to create revolved surfaces, so you can snap around true vertical or horizontal axes. With regards  to being able to modify or assign the center of revolve, you can move the axis of the revolve when creating a revolved surface-- you just move the two blue points anywhere you want, so in a way this is modifiable. It is not however assignable to other shapes. 

In addition, with regards to rotating objects, uMake currently does not have the ability to change the origin of rotating an object, however this is on our feature request list. 


I hope that helps to answer your question. What is your background in 3D and what other features would you like to see? Thanks again for sharing your questions with us. 

Nyko, Thanks for your quick and thoughtful response. I am familiar with your work as part of local motors. I have in the past submitted several concepts to local project competitions. I am an ID major and have used a variety of 2D and 3D packages. I am mac centric and currently use Rhino for Mac OS, Cheetah 3D and the last version of Silo which I still use. I was a long time user of Autocad and have used Sketchup, Corel Draw back in the day and a package called Trispectives, which pioneered low cost paramedic 3D solid modeling and had a great interface but was unfortunately bought out renamed Ironcad and was brought back into the VAR vortex with vastly higher prices. I appreciate what you guys are trying to do and wish you success. I appreciate the speed at which your developments are proceeding my hope is that you create as lean and functional an interface as possible. Some examples that stand out from other user interface experience are: Integration of related actions in a elegant visual format, examples like the various three axis tools that incorporate constrained axial movement with scale and rotation tools. Added features to make such a construction tool more useful would be displayed movement feedback in real time. It would also be great to have the axial tool start out rationalized to world coordinates but then have each object have an independent object linked axial toll in the object current aspect. It would also be great to map the interactions for often used commands and innovate to make the user more productive when using these common tools and functions. One quick question is there a way to constrain the axial movement of say curve edit points in this version of the software? I have moved points and then when the view is rotated the points have moved off there original parallel plane. Regards, Michael

Well hello Ma Duce Coupe! Ha. Very cool to reunite here on uMake. I just went over to Local to see what projects you did, and I remember M-Level last year. The DARPA challenge was definitely one of the most fun projects I had the opportunity to be part of. I really enjoyed the time Local and feel it really prepped me for the amazing journey at uMake.


You have quite the 3D modeling background-- very good to know. I always enjoy knowing more about the folks so we can get some context around the suggestions. 


Your suggestions all sound great, and I think in due time, many of them will be possible. The challenge, as you mentioned, is to keep the interface clean, simple, organized, and most importantly intuitive. We are trying to not be 'just another CAD' package-- instead we are focused on creating a new way of creating with a focus on making it not only easy and quick to get work on ideas and projects, but also fun. 


With the touch based interface of the iPad, some things that are possible on the computers with mouse/keyboard based interfaces are a bit more difficult/ less practical, at least at this point in time. While some things would work great on the iPad Pro with the Pencil, as uMake works on the iPad Air 2, we have to make sure that all contact areas are not too small to use with a finger, so something like the typical global transform 'gamble' that you might find in some softwares is a bit less practical. That said, we do believe anything is possible. 


With regards to constraining axial moment, with the snap tools in the Pro version, this addresses the problem somewhat, but I understand what you are asking for. Definitely something that is also in our plans further down the road. 


Live movement feedback is a great idea-- I think it would make sense when we add dimensioning to uMake, which is also near the top of our list. 


With regards to constraining axial movement of curve edit points-- we currently don't have this, and I know your experience of moving the points from one view, and then seeing they moved awkwardly in another view. This can happen depending on the snap tools that are enabled, but there is a need to be able to lock/ align control points for sure. The current and upcoming version don't have this, but again, this is on our list. 


It might be prudent to say that our list is quite long, and we've received a ton of great feedback from several qualified individuals. The future for uMake is very bright. 


As you've mentioned, you are aware that we are pushing an update on average every 2 weeks, and we are trying to push at least one meaningful community requested feature request with each update. Our goal is to continue this pace but be able to add a few very cool features with each update in the future. We are really excited about it. It may take some time, but we will get there, together. In the near future I can tell you that we are trying to focus on features that will really enhance the usability of uMake (like Layers did). There are still some features like this that we feel will really make a difference, and once these are addressed, we will be able to look at a new set of features-- some of which may be things like you are suggesting. 


Really great feedback! Thanks for it! 





Nyko, again thanks for the thoughtful response. I am excited about the evolution of 3D sketching rationalized to the emerging touch/gesture interface. And hopefully the new tools will express the unique capability of the new approach. The unmake team is in the enviable position of exploring a new paradigm or frontier. it will be exciting to see what emerges from the new context and constraints. Here is an attempt at some qualitative feeling that have been generated from interacting with software from the desktop, keyboard, graphic tablet paradigm. The need for contextual intelligence: Whenever possible the code and interface should be as contextually aware of the design object as the user. Example some software packages allow you to assign an objects Tranformation spacial identity by selecting, world, object, normal, etc. I have had experiences with this approach where an object is created to, "world coordinates" and then the object is rotated in space (think of a subassembly to a larger object) then an additional transformation is required but the objects transformation reference is locked to, "global". So now if you have a circular (tubular, to use an eighties slang reference) component and try to scale it it will become an oval. It would be great if the software could recognize and remember that components past and default to whatever the most intuitive, rational transformation context would be. Our maybe the relationship of the object axis to it's initial XYZ relationship as a transformation default. So if that object was rotated 0,0,45 degrees (X,Y,Z) the transform tools would align to that 0,0,45. This would also be very helpful,in creating future Boolean options, streamlining features like projecting joining surfaces. Another aspect of contextual intelligence concerns proximity. I know I am only partially along my learning curve with make sketch but one frustration is point editing snapping onto a curve point at the far point of the object. Perhaps some of the software algorithms from displacement painting could be adopted to create a sphere of attraction or falloff that could allow the user to have greater productivity and work with confidence especially on larger more complex objects. So if you are in curve edit mode and working with points the software would prevent huge leaps in distance after just a small spacial manipulation of the point being edited. Another annoyance is that inadvertent finger touches generate curves some very small and hard to notice. If the intent of the bottom right view button is a zoom extent button my working model is small and surrounded by white space when I press it.

Michael, 


I think all of the concerns can be addressed with time.


The relationships with regards to transformation are tricky because in one way we would want to keep the interface/ interaction simple and not complex. However I do think for Pro's this is something we could look into for something like the Pro Subscription that would make a big difference-- i.e. the Free versions would be easier and have less of the more 'advanced' tools, but the Pro version would truly be a pro version complete with really pro tools that would be understood by pros and easy to learn from enthusiasts. 

With regards to the snapping, I totally understand what you are talking about and trust me, it is annoying when moving points around in one view only to turn the camera and realize the curve is all wonky and not what you thought. We do want to address this in the future for sure, and I think we will be able to with some settings. 


Regarding the inadvertent finger touches, we are aware of this, and are looking into ways where you could highlight 'small curves' in a drawing to be able to easily clean a drawing up. 


Lastly, with regards to your model-- this could be due to one of two things. 1. your model may be very small (I've noticed several people work very small-- it is hard to tell because there is no real scale and the grid also changes as you zoom in/ out) or b. You may have some line floating around way off that is making the scene big. 


Thanks again for your feedback. Definitely good to hear and in our minds as we move forward. 

I can say that while working here is a total blast, we have SO much we want to do that it is a bit overwhelming while simultaneously very exciting. We are working hard to prioritize each update and ensure upcoming features are meaningful. 


For example Layers was something we had initially delayed in favor of other features, but when we realized just how important layers were and how much demand there was for them, we moved these way up. We are now working on opacity settings and groups (so you can take a complex shape like a wheel and group it, so that tapping on any of the elements that make up the wheel will let you easily transforms and work with it). This is kind of the first step in getting to the various 'coordinate' modes. 


Hope that helps shed a bit of light on things. 

Please sign in to leave a comment.