Sync is Great for Editing, but What About Creation?
One of the nagging questions I keep coming back to is “What about Creation?” By this of course I mean that Synchronous Technology is great for editing prismatic parts. Soon, it will hopefully also be great for editing parts with shape. I agree that editing is 80% of design (create once, iterate endlessly). But I’d like to open a discussion about geometry creation.
What is there to do new in creation? I’ve suggested primitives, which might reduce creation of prismatic shapes by a step or two. That would be a modest improvement, and it may or may not be worth while. But what about creating more complex shapes? ST6 is said to give Edge users new capabilities in editing complex shapes. That will certainly be welcome, but will we still have to create complex shapes in the same way to begin with? The Bluesurf functionality is powerful, but it takes a long time to set up. Planes, edges, sketches, connecting the sketches, creating relationships, it’s all a lot of work. No software is going to read your mind and create shapes based on your thoughts. Complex shapes are always going to take more time to set up and create than prismatic shapes, but there might be some things that could be done to ease the burden to some extent.
Here are some ideas for simplifying set up for complex shapes. These ideas are going to have to apply primarily to surfaces, because it would be inefficient to apply them to solids. The sketches won’t wind up driving the shape, but with complex faces, the U/V curves do drive the shape, and they can be used to manipulate the shape.
- In the spirit of creating primitives for solids, it would be cool to define a “sheet” to be laid down based on existing edges or sketches, and you could specify how many U/V splines connect the sides. To get the shape, just edit the splines that are automatically laid in for you. This would be like a bluesurf where SE creates specified U/V curves for you, which you can edit.
- The automatic patch or sheet should have “rudders” and “pullers” at the corners and other places to control the direction and strength of tangency/curvature controls.
- Tangency and Curvature Continuity are not enough. You need these “pullers” to specify the distance over which the curvature continuity fades. (In SolidWorks the distance is not controllable, and is very short, so it can cause unintended bumps and things that Rick Williams keeps taking about)
- If you have to lay a patch into a non-4sided area, it would be great if Solid Edge had some sort of a magic wand to help you build the patch using the final boundary, but still be able to control the shape (SolidWorks Fill feature is great at n-sided patches, but not so great at allowing you to control the shape)
Do you guys have other ideas about ways to bring something new to creation? Specifically shape creation to complement Sync being able to handle complex shapes. I think there have to be some ideas to help shape creation to catch up to editing.
Actually Matt now that I think about it I seem to remember talking about this pullers idea or something related a few years ago for SW although it obviously fell through the cracks.. Its a good idea though, however I would make it a ‘tangency comb’ for easy visualisation, and perpendicular to the curvature comb which you could then use to manipulate the continuity fade smoothly/predictably along the edges. I guess that would look like a 2 point spline the end points of which pull out perpendicular to the edge and you can prescribe even an S progression between them by their tangent weighting. Hope that made sense..
Matt-
Have you seen the camera body being driven from a block or the glass container demo’s (Derek England)? These are all NX demos on ST and they are all creation driven demos. Mike Rebrukh, from the NX side, has always been a proponent of using ST for creation. I recall watching a presentation of his where he built a bracket for a wing on an airplane- very few prismatic faces on that part by the time he got done with it.
As far as ST for creation, why not? If the change is so easy to make then the creation is a no brainer. Build what you need, drop the sketches and continue on. Once you have your shape start making the changes you need or start making the shape control rules for your parts.
Ryan
Ryan,
Yes, I’ve seen at least the camera demo. That’s been around for a while, and part of the case I used to pitch the whole synchronous surfacing bit in Huntsville a while back. Just so everyone else knows what we’re talking about, here is the video:
I would love to be able to do that is SE. Inclusion of that capability in say ST7 would make SE very attractive indeed and neutralise any ideas that I might have about SE not being useful for surfacing. Throw in Leap Motion controller for some ‘air sculpting’ and it would be vey cool….add a realtime raytraced gpu (not limited to Quadro/Firepro though please) renderer for visualisation and I will be knocking on the door with my money in hand. 🙂
You’re aiming pretty high there Neil. The SE renderer has been outdated for some time now, lagging far behind SW and IV, both in terms of output and usability. That being the case, you would think .obj export would be possible so that third party renders could be used – but no dice.
There was a half-hearted attemp to provide native import into Bunkspeed but it doesn’t work with frames (weldments) so is pretty useless and no sign of being fixed.
It is going to be some time before the SE offering is sufficiently rounded to tempt SW users.
Not the most optimistic post we’ve had here… I do keep asking for SE .obj support for a reason 🙂 but ok they could go the whole way and integrate Octane into the UI for me. Otoy are always looking for more adopters of their SDK… very effective and reasonably priced 😉 Nvidia have their own free one – not biased though – that would do the job well.
Hey if you ask it might* arrive and at the moment the good part is that SE seem to want to listen 😀
…and anyway lets not forget the momentum for DS to add stuff to SW has all but run out as Catia lite waits in the wings. Its unfortunately probably far more likely that any request for something inspired/genuinely useful for ID users would arrive in SE than SW these days 🙁 DS more or less do small stocking fillers to prolong subs, not real enhancements…sad but true.
Mmmmm…….Neil, I raised both points (outdated renderer and lack of .obj) with senior management during a Q&A at last years PLM Connection in th UK. The respondent (not Karsten or Dan) was clearly offended by my suggestion that the Lightwave rendering engine was outdated and could only mumble a sentence about how they were in discussions with someone. The .obj question wasn’t answered.
The Q&A was largely a waste of time as we were given no insight into the management vision and my impression was that he saw it as something to “get through” rather than engage with the customer.
On the flip side, Karsten showed enthusiasm for the product but more importantly went out of his way to make himself available to talk and LISTEN to customers.
The reason I asked the question was to see if SE had a cohesive marketing/development strategy. We had been told that ST6 would have improve ID functionality but unless this is coupled with the ability to efficiently achieve rendered output (either directly or via third party) then ID guys will not buy it.
This would be a deep shame as I’m certain Dan and his team have done a great job with ST6.
Well I’ve got my fingers crossed. The spotlight is really going to be on ST6 come June/July as people look hard at SE as a potential replacement for SW. Hopefully key people have woken up and attitudes have changed. Dan and Karsten seem keen to do their best for their existing and prospective customers. If Siemens demonstrate as a whole however, either directly or by default, they havent got the business awareness to deliver, or a commitment to making SE as good as it can be (including renderers) they are going to let down a lot of people and ultimately ruin Siemens big opportunity. There is only so much we would expect to see arrive in ST6 that we have indicated are the sorts of things we need or would like but come ST7 I would expect at least .obj out to be available. That’s not hard to do and it would earn a lot of bonus points with the user base. I am not sure why Siemens would want to frustrate their users by offering world leading sync technology but then give them old tech for rendering/presentation or limit their downstream choices. To be frank I think ST7 would be Siemens last chance to get themselves out of the management blight that has kept them in third place for so long regardless of how DS and Autodesk cloud endeavours play out. I would encourage Siemens to think about all aspects of their business that obstruct the adoption of SE or currently disappoint their customers.
Ok not as glamour as in the NX video, but this was done in ST1 or 2
http://vimeo.com/3331350
Simple camera body from Solid DNA on Vimeo.
I should revisit and add bluesurf stuff on it 🙂