CAD Forum

I had someone looking for an alternative to their current CAD system and they were thinking Solid Edge but were hesitant about the synchronous approach. I did ease their worries by letting them know that you can run Solid Edge in it's original format by switching from synchronous to ordered in the options menu. This will run Solid Edge more closely to other CAD systems.

Not only that, you can combine methods. What I don't get is how much people complain about the downside of history-based modeling, and yet defend it as if their lives depend on it when you mention a system that doesn't use history.

Getting out from under history is a great thing for simple models.

And for complex models, history is the biggest mess that ever was.

Required reading for all new history-based trainees: https://www.3dcadworld.com/the-failed-p ... etric-cad/

matt wrote: ↑Tue Sep 20, 2022 11:49 am Not only that, you can combine methods. What I don't get is how much people complain about the downside of history-based modeling, and yet defend it as if their lives depend on it when you mention a system that doesn't use history.

Getting out from under history is a great thing for simple models.

And for complex models, history is the biggest mess that ever was.

Required reading for all new history-based trainees: https://www.3dcadworld.com/the-failed-p ... etric-cad/

The paradigm shift may be too great for most to absorb. I'm still not 100% sold on it, even though I see the benefits. Mostly for me, it's a training issue, I'm only tinkering and learning it, not putting it into practice and using it 8 hours a day. For complex parts, some changes are dirt simple, others have me banging my head against the desk as much as any history modeler. I've also "changed" my models into a corner sometimes that seem hard to get out of....absorb faces and such that got Iost and I don't remember what they were when coming back to it a few days or even weeks later. It really is a shift in thinking.

I'm not opposed to the idea, in theory. I've done very little with it in practice, but I have yet to come across a part where I say to myself "It would be easier if I could just drag this face instead of editing the sketch." Having said that, most of what I draw is square and boxy. If I were drawing organic shapes I might have a different opinion.

I think we can all agree that for imported geometry, Synchronous is a "no brainer".

SPerman wrote: ↑Tue Sep 20, 2022 1:10 pm I'm not opposed to the idea, in theory. I've done very little with it in practice, but I have yet to come across a part where I say to myself "It would be easier if I could just drag this face instead of editing the sketch." Having said that, most of what I draw is square and boxy. If I were drawing organic shapes I might have a different opinion.

Use it for a month. Then try going back.

I need to watch this video from Matt and see if it enlightens me.

Using a combination approach allows you to get the best of both. I will agree with @KennyG that synchronous is the best option when working with imported geometry.

bryan5 wrote: ↑Tue Sep 20, 2022 4:40 pm Using a combination approach allows you to get the best of both. I will agree with @KennyG that synchronous is the best option when working with imported geometry.

I definitely see the benefits there, it's just not something I do. The only imported geometry I use is for packaging, and doesn't require modification. (Although there are some vendors whos models are so bad they could use some rework.)

SPerman wrote: ↑Tue Sep 20, 2022 4:42 pm I definitely see the benefits there, it's just not something I do. The only imported geometry I use is for packaging, and doesn't require modification. (Although there are some vendors whos models are so bad they could use some rework.)

Uh huh. Like engines with 1700 surface bodies.....

mike miller wrote: ↑Tue Sep 20, 2022 4:52 pm Uh huh. Like engines with 1700 surface bodies.....

Lol, I get walking floor layouts for our trailers that contain over 30 000 bodies, half of them generally being surfacic.

As Shania Twain would say, that don't impress me much

matt wrote: ↑Tue Sep 20, 2022 11:49 am Not only that, you can combine methods. What I don't get is how much people complain about the downside of history-based modeling, and yet defend it as if their lives depend on it when you mention a system that doesn't use history.

Getting out from under history is a great thing for simple models.

And for complex models, history is the biggest mess that ever was.

Required reading for all new history-based trainees: https://www.3dcadworld.com/the-failed-p ... etric-cad/

I agree... partially. Usually I do reach a point where I say, yes, I'm sure this model is going to look like this. Then I'll history-brainwipe it and generally proceed with direct edit-type operations. (I'm on NX). Sometimes I'm not yet comfortable with junking the history, in which case I'll do a dumb extraction (which can be made associative later if the need arises) and continue with that.

But this goes in stages, and I will usually then end up building up a new history tree while tuning (this new tree being albeit a small one) before brainwiping again. I've worked with totally history-free systems and too often found myself with unmodifiable geometry, that then needed to be reverted via the Frankenboolean method.